PATROLKnowledge Module for Unix User Guide

Supporting
PATROL for Unix and Linux 9.2

March 2005

Contacting BMC Software

You can access the BMC Software website at http://www.bmc.com. From this website, you can obtain information about the company, its products, corporate offices, special events, and career opportunities.

United States and Canada

Address BMC SOFTWARE INC 2101 CITYWEST BLVD HOUSTON TX 77042-2827 USA Telephone 713 918 8800 or 800 841 2031 Fax 713 918 8000

Outside United States and Canada

Telephone (01) 713 918 8800 Fax (01) 713 918 8000

Copyright 2005 BMC Software, Inc., as an unpublished work. All rights reserved. BMC Software, the BMC Software logos, and all other BMC Software product or service names are registered trademarks or trademarks of BMC Software, Inc. IBM is a registered trademark of International Business Machines Corporation. DB2 is a registered trademark of International Business Machines Corporation. Oracle is a registered trademark, and the Oracle product names are registered trademarks or trademarks of Oracle Corporation. All other trademarks belong to their respective companies. Product Name technology holds U.S. Patent Number number. BMC Software considers information included in this documentation to be proprietary and confidential. Your use of this information is subject to the terms and conditions of the applicable End User License Agreement for the product and the proprietary and restricted rights notices included in this documentation.

Restricted rights legend

U.S. Government Restricted Rights to Computer Software. UNPUBLISHED -- RIGHTS RESERVED UNDER THE COPYRIGHT LAWS OF THE UNITED STATES. Use, duplication, or disclosure of any data and computer software by the U.S. Government is subject to restrictions, as applicable, set forth in FAR Section 52.227-14, DFARS 252.227-7013, DFARS 252.227-7014, DFARS 252.227-7015, and DFARS 252.227-7025, as amended from time to time. Contractor/Manufacturer is BMC SOFTWARE INC, 2101 CITYWEST BLVD, HOUSTON TX 77042-2827, USA. Any contract notices should be sent to this address.

Customer support
You can obtain technical support by using the Support page on the BMC Software website or by contacting Customer Support by telephone or e-mail. To expedite your inquiry, please see Before Contacting BMC Software.

Support website
You can obtain technical support from BMC Software 24 hours a day, 7 days a week at http://www.bmc.com/support_home. From this website, you can
s s s s s s s

read overviews about support services and programs that BMC Software offers find the most current information about BMC Software products search a database for problems similar to yours and possible solutions order or download product documentation report a problem or ask a question subscribe to receive e-mail notices when new product versions are released find worldwide BMC Software support center locations and contact information, including e-mail addresses, fax numbers, and telephone numbers

Support by telephone or e-mail

In the United States and Canada, if you need technical support and do not have access to the web, call 800 537 1813 or send an e-mail message to [email protected]. Outside the United States and Canada, contact your local support center for assistance.

Before contacting BMC Software

Before you contact BMC Software, have the following information available so that Customer Support can begin working on your problem immediately:
s

product information product name product version (release number) license number and password (trial or permanent)

operating system and environment information machine type operating system type, version, and service pack or other maintenance level such as PUT or PTF system hardware configuration serial numbers related software (database, application, and communication) including type, version, and service pack or maintenance level

s s s

sequence of events leading to the problem commands and options that you used messages received (and the time and date that you received them) product error messages messages from the operating system, such as file system full messages from related software

PATROL for Unix User Guide

Contents
Chapter 1 Introduction 19 20 20 20 22 25 25 27 29 29 29 29 30 31 33 33 33 34 34 34 35 36 36 36 37 38 38 39 41 42 42 43 44 44 44 45 46 47
5

PATROL KM for Unix and Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Supported Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications and Icons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Accessing Online Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 2 Menu Summary

Location of KM Menu Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL Console for Unix. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL Console for Microsoft Windows Servers . . . . . . . . . . . . . . . . . . . . . . . . . Menu Summary Table Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Computer Instance Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . UNIX OS Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COLLECTORS Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COMPOSITES Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COMPOSITES_COLLECTOR Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CPU Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DCM Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISK Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FILESYSTEM Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HEALTH AT A GLANCE Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . KERNEL Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MEMORY Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NETWORK Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NFS Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROLAGENT Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PRINTER Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PROCESS Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PROCESS_PRESENCE (PROCCONT) Application Menu . . . . . . . . . . . . . . . . . . . . . . PROCPRES Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SCHEDULER Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SECURITY Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SMP Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SNMPHealth Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SWAP Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USERS Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents

Chapter 3

Parameter Summary

49

Parameter Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Parameter Defaults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Chapter 4 Managing Data Collection 87

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Object Hierarchies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 COLLECTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 DCM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 DCM Collection Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Data Collection Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Collector Parameters for DCM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 -Ctrl and -Coll Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 DCM Collection Exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 PSL Collection Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Collector Parameters for COLLECTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 -Coll Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 PSL Collection Exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Managing Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Changing Data Collection Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Activating an Inactive Application Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Activating an Inactive Collector Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Significant Events Related to Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Chapter 5 Monitoring CPU and Kernel 105

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Object Hierarchy for CPU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 CPU and Multiple CPUs (SMP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Viewing CPU Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Viewing CPU Statistics on Interrupts, Switches, and Waits . . . . . . . . . . . . . . . . . 109 Viewing Statistics on CPU Run Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Viewing Statistics on Multiple CPUs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Object Hierarchy for KERNEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 KERNEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Viewing the Kernel Build Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Viewing Kernel Memory Allocation Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Viewing Kernel I-node Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 Viewing Kernel System Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Chapter 6 Using Composite Parameters 117

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Object Hierarchy for COMPOSITE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Working with Composite Parameter Expressions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Examples of Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Syntax Rules for Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . 120 Setting the Status of a Composite Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Creating Composite Parameter Expressions Using the Wizard . . . . . . . . . . . . . . 121
6 PATROL Knowledge Module for Unix User Guide

Creating Composite Parameter Expressions Manually . . . . . . . . . . . . . . . . . . . . . 124 Editing Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Deleting Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 Chapter 7 Monitoring Disks 127 128 128 130 130 131 132 133 135 136 137 138 138 139 141 143 144 145 146 146 148 149 150 153 157 158 158 159 159 160 161 162 163 163 164 166 167 167 168 170
7

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Disks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Excluding a Single Disk from Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selecting Disks to Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Resuming Monitoring Disks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Disk Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 8 Monitoring and Managing File Systems

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitoring File Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Excluding a Single File System from Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . Selecting File Systems to Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Automating File System Cleanup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Running Recovery Actions Only with Operator Confirmation . . . . . . . . . . . . . . Preventing File Systems from Being Monitored . . . . . . . . . . . . . . . . . . . . . . . . . . . Working with File System Types and File System Classes . . . . . . . . . . . . . . . . . . Managing File Systems Using PATROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mounting a File System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Unmounting a File System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synchronizing File System Cache and Disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Searching for Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Removing Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 9 Monitoring Overall System Health

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HEALTH AT A GLANCE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Overall CPU, File System, and Swap Usage . . . . . . . . . . . . . . . . . . . Viewing the Overall Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 10 Monitoring and Managing Memory

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy for MEMORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shared Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Shared-Memory Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Shared-Memory Segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Semaphores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Semaphore Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Semaphore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy for SWAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents

Virtual Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Viewing the Amount of Swap Space. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Viewing the Swap Partition Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Viewing the Percentage of Used Swap Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Chapter 11 Monitoring and Managing Networks 173

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Network Activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Activating the NETWORK Application Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Deactivating the NETWORK Application Class . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Network Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Viewing a Network Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Discovering the Name of the Network Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Establishing Behavior When Unavailable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Viewing Memory Buffer Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Network Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 Viewing Network Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 Listing Host and IP Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Network Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Listing Protocols Running on the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Viewing Protocol Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Determining the Network Internet Protocol Address . . . . . . . . . . . . . . . . . . . . . . 191 Network Traffic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 Viewing the Routing Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 Determining the Number of Packets Entering and Exiting the Network . . . . . . 194 Determining the Number of Packet Errors Entering and Exiting the Network . 194 Network Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Listing TCP/IP Services Running on the Network. . . . . . . . . . . . . . . . . . . . . . . . . 196 Listing Trusted Hosts on the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Copying or Transferring Files by Using FTP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Checking for a Network Resource by Using ping. . . . . . . . . . . . . . . . . . . . . . . . . . 199 Logging In Remotely by Using rlogin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Querying the Name Server for the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Address Resolution Protocol (ARP). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Viewing the Address Resolution Protocol (ARP) Cache . . . . . . . . . . . . . . . . . . . . 202 Deleting a Host from the ARP Cache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Adding a Host to the ARP Cache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Unix to Unix Communication Protocol (UUCP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Checking for UUCP Support Files and Directories. . . . . . . . . . . . . . . . . . . . . . . . . 206 Cleaning Up UUCP Files in the Spool Directory. . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Copying Files Using UUCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 Listing All UUCP Servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Viewing the UUCP Log File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Viewing UUCP Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Chapter 12 Monitoring Network File Systems 213

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
8 PATROL Knowledge Module for Unix User Guide

Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network File System (NFS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network File System (NFS) Servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Resetting Network File System Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network Information System (NIS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network Information System Accounts. . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network Information System Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network Information System Servers . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 13 Monitoring and Managing Printers

214 216 216 217 218 218 219 220 223 224 224 225 225 226 226 227 227 228 228 229 231 231 232 233 234 235 236 237 238 239 241

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Printer Activation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Activating the PRINTER Application Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deactivating the PRINTER Application Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . Printers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Excluding a Printer from Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selecting Printers to Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Number of Print Jobs in a Printers Queue . . . . . . . . . . . . . . . . . . . . Stopping a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Print Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing All Print Jobs Assigned to a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing All Print Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Print Job . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting All Print Jobs from a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Moving a Print Job to Another Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Moving All Print Jobs to Another Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Print Queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stopping a Print Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting a Print Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 14 Monitoring Processes

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 Configuring Multiple Instances of a Process for Monitoring . . . . . . . . . . . . . . . . 247 Configuring a Process to Alarm When the Number of Running Instances Exceeds a Set Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 Configuring a Process to Alarm When the Number of Running Instances Falls Below a Set Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Configuring a Process to Restart When PATROL Detects It Has Stopped. . . . . 256 Configuring a Process to Stop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Configuring a Process to Alarm When Its Parent Process ID (PPID) is Not 1 . . 262 Configuring a Process to Alarm When It is Owned by an Unauthorized User . 265 Configuring a Process to Suspend Its Alert for N Cycles if Patrol Detects It is Not Running . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 Modifying the Configuration of a Monitored Process . . . . . . . . . . . . . . . . . . . . . . 270
Contents 9

Monitoring Process Presence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 Deleting a Monitored Process from the Monitoring Queue . . . . . . . . . . . . . . . . . 278 Specifying a Flag File Directory for Process Presence Blackout Periods . . . . . . . 278 Suspending Process Monitoring for a Process Instance . . . . . . . . . . . . . . . . . . . . . 280 Resuming Process Monitoring for a Process Instance . . . . . . . . . . . . . . . . . . . . . . 280 Reporting on a Selected Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 Process Host Groups in a Cluster. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Rules for Creating Process Host Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Creating a Host Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 Configuring Communication Attributes for a Process Host Group. . . . . . . . . . . 284 Modifying Host Attributes within a Process Group. . . . . . . . . . . . . . . . . . . . . . . . 285 Removing a Host from a Host Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 Querying a Selected Host within a Process Host Group . . . . . . . . . . . . . . . . . . . . 286 Synchronizing the Local Hosts List to a Selected Hosts List for a Process Host Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 Synchronizing a Selected Hosts List to the Local Host List for a Process Host Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 Viewing the Process List for a Selected Host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 Synchronizing the Local Processs List with a Selected Process's List . . . . . . . . . 289 Synchronizing a Selected Process's List with the Local Process's List . . . . . . . . . 290 Viewing the Process Presence List for a Selected Process . . . . . . . . . . . . . . . . . . . 291 Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Viewing the Number of Zombie Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Viewing Zombie Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 Viewing the Total Number of Processes Running . . . . . . . . . . . . . . . . . . . . . . . . . 295 Viewing Processes that Use Excessive CPU. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Viewing the Number of Processes Waiting on Resources . . . . . . . . . . . . . . . . . . . 298 Viewing the Total Number of User Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Viewing the Average Number of User Processes . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Priority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Reprioritizing a Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 Reprioritizing All Processes in a Process Group . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Reprioritizing All Processes Owned by a User . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 Plug-in Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 Configuring the Process KM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Chapter 15 Scheduling Tasks 311

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 Scheduling a Task for Execution via the Operating System Command Line Processor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 Scheduling a Task for Execution Using a PATROL Command Processor . . . . . 315 Scheduling a PSL Task for Execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 Modifying an Existing Scheduled Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Removing a Scheduled Task from the Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Collecting Debug Data for the Scheduler Application . . . . . . . . . . . . . . . . . . . . . . 322
10 PATROL Knowledge Module for Unix User Guide

Blackouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Specifying a Default Directory for Blackout Scripts. . . . . . . . . . . . . . . . . . . . . . . . Creating a Monitoring Blackout Script. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Information about a Blackout Script. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modifying an Existing Blackout Script. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Blackout Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Scheduling a Blackout Script for Execution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Canceling a Scheduled Blackout Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 16 Reviewing Security

326 326 327 330 331 334 335 337 339 340 340 341 341 344 345 345 346 348 351 352 352 353 353 354 354 355 356 357 360 361 362 363 364 364 365 366 367 369 370 373 374 374 375 375

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Files with SUID or SGID Permissions . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Files with Global Write Privileges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . User Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Users that Failed to Execute su/msu Commands . . . . . . . . . . . . . . . . . Viewing Users without Passwords . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Users with Multiple Sessions Running . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 17 Monitoring the Health of SNMP

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL SNMP System Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Role of Each Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SNMP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Changing SNMP Query Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Listing Configuration Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing PATROL SNMP MIB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Status of SNMP Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting the SNMP Master Agent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting the SNMP Subagent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stopping the SNMP Subagent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tests SNMP Functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Testing the SNMP Master Agents Ability to Receive Traps . . . . . . . . . . . . . . . . Generating Manual SNMP Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Querying the PATROL Agent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Querying the SNMP PATROL platform.km . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determining if Any SNMP Managers are Listening . . . . . . . . . . . . . . . . . . . . . . . Testing PATROL Event Managers Automated SNMP Traps . . . . . . . . . . . . . . . Chapter 18 Monitoring Users

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Resources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Print Jobs that Belong to a User. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Contents

11

Viewing Processes that Belong to a User . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376 Viewing the Amount of Disk Space Consumed by a User . . . . . . . . . . . . . . . . . . 377 Numbers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378 Viewing the Number of Users on the System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 Viewing the Number of Sessions Running on the System . . . . . . . . . . . . . . . . . . 380 Appendix A Debugging and Running Diagnostics 381

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382 Unix Diagnostics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382 Capturing DUMP Information for Selected Components . . . . . . . . . . . . . . . . . . . 383 Viewing Installation Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385 Capturing Log File Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 Collector Debug Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388 Debugging FILESYSTEM Data Collection (DFColl) . . . . . . . . . . . . . . . . . . . . . . . . 388 Debugging NETWORK Data Collection (NETColl) . . . . . . . . . . . . . . . . . . . . . . . . 389 Debugging NFS Data Collection (NFSColl) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390 Debugging PRINTER Data Collection (PRINTERColl) . . . . . . . . . . . . . . . . . . . . . 391 Debugging PROCESS Data Collection (PSColl) . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 Debugging USER and PROCESS Data Collection (USRPROCColl). . . . . . . . . . . 392 Debugging CPU, MEMORY, and KERNAL Data Collection (VMColl) . . . . . . . 393 Debugging DCM Data Collection (DCMColl) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394 Application Discovery Debug Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395 Debugging DISK Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 Debugging FILESYSTEM Discovery Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 Debugging NETWORK Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397 Debugging PRINTER Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398 Debugging SMP Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398 Debugging SWAP Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399 Application Class Debug Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400 Debugging the SCHEDULER Application. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400 Appendix B Loading, Unloading, and Navigating KMs in the PATROL Consoles 401

Accessing KM Commands and InfoBoxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402 Accessing Online Help. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402 Using the Mouse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404 Using the Mouse to Perform Tasks in PATROL . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 Appendix C Regular Expressions 407

Components that Use Regular Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408 Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410 Conventions for Using Regular Expressions with PATROL Objects . . . . . . . . . . . . . 411 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411 Appendix D Inclusion and Exclusion Lists 413

Using Exclusion and Inclusion Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414

12 PATROL Knowledge Module for Unix User Guide

Methods for Creating Exclusion and Inclusion Lists . . . . . . . . . . . . . . . . . . . . . . . Storage Location of Exclusion and Inclusion Lists . . . . . . . . . . . . . . . . . . . . . . . . . Benefits and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benefits of Creating Exclusion Lists Using Menu Commands . . . . . . . . . . . . . . . Inclusion List Limitations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Index

414 415 415 415 416 417

Contents

13

14

PATROL Knowledge Module for Unix User Guide

Figures
PATROL KM for Unix Application Class Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 COLLECTORS Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 DCM Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Data Collection Architecture Using PATROL Perform Collector . . . . . . . . . . . . . . . . 91 PSL Collection Confirmation Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 CPU Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 KERNEL Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 KERNEL InfoBox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 COMPOSITE Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 COMPOSITES_COLLECTOR Application Icon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 COMPOSITES Parameters Container Icon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 DISK Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 DISK Exclusion List Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 FILESYSTEM Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 FILESYSTEM Exclusion List Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 Edit Recovery Action Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Run Recovery Action Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Automatic Mount Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Manual Mount Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Automatic Unmount Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Manual Unmount Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Find Files Like Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Remove Files Like Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 HEALTH AT A GLANCE Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . 158 MEMORY Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Remove Shared Memory Segment Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 Remove Shared Memory Segment Confirmation Dialog Box . . . . . . . . . . . . . . . . . . 166 Remove Semaphores Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Remove Semaphores Confirmation Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 SWAP Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 NETWORK Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 NETWORK InfoBox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Ping Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Rlogin Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Delete Host from Cache Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Delete Host From Cache Confirmation Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Add Host To Cache Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 UUCP Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 NFS Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 NFS User Name Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Figures 15

PRINTERS Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 Delete Print Job Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Delete All Jobs Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234 Move Job to Another Printer Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 Move Jobs To Another Printer Confirmation Dialog Box . . . . . . . . . . . . . . . . . . . . . . 236 Move All Jobs to Another Printer Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 Move Jobs To Another Printer Confirmation Dialog Box . . . . . . . . . . . . . . . . . . . . . . 237 PROCESS Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 PROCESS InfoBox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 CPU Hogs Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296 Renice Process Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 Renice Process Confirmation Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 SCHEDULER Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 SECURITY Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340 SECURITY User Name and Password Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341 Find SUID/SGID Files Like Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342 SNMPHealth Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352 PATROL SNMP System Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 Change SNMP Query Port Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 Walk PATROL SNMP MIB Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358 SNMP Query PATROL Master Agent Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 SNMP Query PATROL Platform.km Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368 State Change Messages Based on Automated Traps . . . . . . . . . . . . . . . . . . . . . . . . . . 371 USERS Application Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374 List Print Jobs Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 USERS InfoBox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 Select Categories of Diagnostics Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383 %DUMP Options Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384 Select Log Files to Collect Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 Mouse Button Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404

16

PATROL Knowledge Module for Unix User Guide

Tables
PATROL KM for Unix and Linux Applications, Icons, and Descriptions . . . . . . . . . 22 Accessing Online Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Menu Summary Table Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Computer Instance Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 UNIX_OS Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 COLLECTORS Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Composites Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 DCM Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 DISK Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 FILESYSTEM Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 MEMORY Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 NETWORK Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 NFS Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 PATROL Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 PATROLAGENT Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 PRINTER Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 PROCESS Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 PROCESS_PRESENCE Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 42 PROCPRES Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 SCHEDULER Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 SECURITY Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 SNMPHealth Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 USERS Application Menu Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 PATROL KM for Unix and Linux Parameter Summary . . . . . . . . . . . . . . . . . . . . . . . . 50 PATROL KM for Unix and Linux Parameter Defaults . . . . . . . . . . . . . . . . . . . . . . . . . 80 Data Collection Methods and Associated Application Classes . . . . . . . . . . . . . . . . . . 88 Collection Parameters Used By the DCM Collection Method . . . . . . . . . . . . . . . . . . . 92 DCM Deactivated Parameters and their Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Collection Parameters Used By the PSL Collection Method . . . . . . . . . . . . . . . . . . . . 95 Collector Deactivated Parameters and their Platforms . . . . . . . . . . . . . . . . . . . . . . . . . 96 Find Files Like Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 Remove Files Like Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 List of Shared-Segment Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 List Network Interface Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 List Memory Buffer Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 List Network Addresses Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . 185 List Hosts Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 List Protocols Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190 List Routing Table Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 List Services Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Tables 17

Show ARP Cache Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 List of NIS Accounts Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 List of NIS Groups Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 List of NIS Groups Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 View All Print Jobs Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Variables for Monitoring Process Presence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 List Zombie Process Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 CPU Hogs Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 PROCTopProcs Text Parameter Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 New Instance Dialog Boxes and Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Process tab properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Start/Restart tab properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 Alert and Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Find Files Like Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343 List Files With Global Write Command Output Format . . . . . . . . . . . . . . . . . . . . . . . 344 List Failed su/msu logins Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . 346 List Users Without Passwords Command Output Format . . . . . . . . . . . . . . . . . . . . . 347 List Duplicate User ID Entries Command Output Format . . . . . . . . . . . . . . . . . . . . . 349 PATROL SNMP Component Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354 Walk PATROL SNMP MIB Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . 360 Process List Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 Disk Space Command Output Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378 Capture Unix Diagnostics %DUMP Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384 Select Log Files to Collect Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 Accessing KM Commands and InfoBoxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402 Accessing Online Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403 Regular Expression Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409

18

PATROL Knowledge Module for Unix User Guide

Chapter

1
20 20 20 22 25 25

Introduction
This chapter provides you with a brief overview of the PATROL Knowledge Module for Unix and Linux (also referred to as the PATROL KM for Unix) component. The following topics are discussed: PATROL KM for Unix and Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Supported Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications and Icons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Accessing Online Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 1

Introduction

19

PATROL KM for Unix and Linux

PATROL KM for Unix and Linux

The PATROL Knowledge Module (KM) for Unix component is a set of files that contains knowledge in the form of command descriptions, applications, parameters, and recovery actions that analyze and manage PATROL events. The PATROL KM for Unix and Linux parameters provide a detailed view of all system activity over time so that you can detect problems, optimize systems, analyze trends, plan capacity, and manage multiple hosts simultaneously.

Supported Platforms
For a list of operating system platforms that are supported by the PATROL KM for Unix, see the PATROL for Unix and Linux Release Notes. See the PATROL for Unix and Linux Getting Started guide additional system requirements and platform information. For a list of PATROL for Unix components and PATROL architecture components that are compatible with the PATROL KM for Unix, see the compatibility matrix in the PATROL for Unix and Linux Release Notes.

Object Hierarchy
The PATROL products are organized hierarchically. At the top level of the hierarchy is the computer instance. Beneath it is the Unix operating system object. All the Unix product application classes are subordinate to the Unix object. Many of the application classes contain instances, but not all application classes have instances or parameters. Figure 1 on page 21 illustrates the basic hierarchy. It does not necessarily display all the application classes in the product.

20

PATROL Knowledge Module for Unix User Guide

Object Hierarchy

Figure 1

PATROL KM for Unix Application Class Hierarchy

Chapter 1

Introduction

21

Object Hierarchy

Applications and Icons

Table 1 contains information on each application in the PATROL KM for Unix and Linux. For information on parameter icons, refer to the documentation for your console. Table 1 Icon PATROL KM for Unix and Linux Applications, Icons, and Descriptions (Part 1 of 4) Definition represents a computer instance that corresponds to the host computer that you are monitoring Opening a computer instance icon displays the application classes that are active for that system. represents the COLLECTOR application, which maintains collector parameters

agent host

COLLECTORS represents the application that you can use to construct a new composite parameter that consists of one or more parameters in a logical relationship that evaluates to true or false COMPOSITES_ COLLECTOR You can use this application to set a warning or alarm condition based on the logical condition and not just on the parameter value. For more information, see Chapter 6, Using Composite Parameters represents the application that displays instances of the composite parameters that you construct with the COMPOSITES_COLLECTOR application COMPOSITES represents the CPU application, which monitors CPU usage This application also contains the SMP application on machines with multiple processors. CPU represents the Data Collection Manager (DCM) application, which controls data collection from PATROL Perform collector to the Data Collection Manager and to associated PATROL parameters DCM represents the DISK application, which monitors and provides information about disks and disk usage

DISK
22 PATROL Knowledge Module for Unix User Guide

Object Hierarchy

Table 1 Icon

PATROL KM for Unix and Linux Applications, Icons, and Descriptions (Part 2 of 4) Definition represents the FILESYSTEM application, which monitors and provides information about file system resources

FILESYSTEM represents the HEALTH AT A GLANCE (HAAG) application, which monitors the general health and well-being of the system by providing an overall view of the systems CPU, file system, and virtual memory usage HEALTH_AT_A GLANCE represents the KERNEL application, which monitors system kernel processes and provides kernel resource-usage information, including i-nodes, the file table, and the process table KERNEL represents the MEMORY application, which monitors memory activity, including paging, I/O caching, and swapping

MEMORY represents the NETWORK application, which monitors network activity

NETWORK represents the NFS application, which monitors RPC (Remote Procedure Call) and NFS (Network File System) activities

NFS represents the PATROL application, which shows the number of PATROL Consoles currently running

PATROL represents the PATROLAGENT application, which monitors and manages the agent

PATROL AGENT

Chapter 1

Introduction

23

Object Hierarchy

Table 1 Icon

PATROL KM for Unix and Linux Applications, Icons, and Descriptions (Part 3 of 4) Definition represents the PRINTER application, which monitors the print queue and printer activity and with which you can perform printer administration activities

PRINTER represents the PROCESS application, which displays statistics on active and zombie processes

PROCESS represents the application that holds PROCCONT instances You can use this application to specify which processes you want to monitor. PROCESS_ PRESENCE represents the PROCPRES application, which monitors specified processes running on your system This application is a child application of the PROCESS and PROCESS_PRESENCE (PROCCONT) applications and resides in the PROCESS_PRESENCE applications container window. represents the SCHEDULER application, which provides menu commands that enable you to perform tasks or run jobs once or at regular intervals. SCHEDULER The task or jobs are written by you and stored as script files. represents the SECURITY application, which provides security-related menu commands SECURITY represents the SMP application This application displays information for each processor on multiprocessor machines. The prefix for SMP parameters is SMP. SMP parameters appear under SMP instances in the CPU container window for machines that have multiple processors. represents the SNMPHealth application This application displays information on the status of the PATROL SNMP subsystem. It also provides the ability to test various aspects of the system. SNMPHealth

name

SMP

24

PATROL Knowledge Module for Unix User Guide

Help

Table 1 Icon

PATROL KM for Unix and Linux Applications, Icons, and Descriptions (Part 4 of 4) Definition represents the SWAP application, which monitors and provides information about utilization of system swap files SWAP parameters appear under SWAP instance icons.

SWAP The SWAP application may return different results on different platforms. represents the UNIX_OS application class This application is a container for all other Unix operating system applications. It does not possess any parameters. It does provide menu commands to set debugging for various collectors and to exclude application instances from being monitored. represents the USERS application, which monitors and provides information about user sessions and processes

UNIX_OS

USERS

Help
Help describes the function of the currently displayed window or dialog box and the use of its elements. The tasks in this section describe how to access KM Help topics and context-sensitive Help from the PATROL consoles.

Accessing Online Help

Table 2 on page 26 provides information about how to access online Help on each console.

NOTE
If you are trying to access Help from a Unix console, you may want to refer to the PATROL Installation Guide for specific instructions on how to install and set up a browser in the Unix environment.

Chapter 1

Introduction

25

Help

Table 2 Console

Accessing Online Help To access product help From the console menu bar, choose Help => Help Topics => PATROL Knowledge Modules. To access application class help Double-click the application class in the KM tab of the console. Then click the Help tab and Show Help in the Application Properties dialog box. To access parameter help Right-click the parameter icon, and choose Help On from the pop-up menu. Double-click the parameter icon; then click the ? icon or Help button in the parameter display window. Double-click the parameter in the KM tab of the console; then click the Help tab and Show Help in the properties dialog box.

PATROL Console for Microsoft Windows

PATROL Console for Unix

From the console menu bar, choose Help On => Knowledge Modules.

Right-click the Choose Attributes => Application Classes and parameter icon, and click Help On. double-click the application name. Then click Show Help in the Application Definition dialog box. From the Operator tab of the console, expand or collapse the list to find the application class. Click the Whats this? Help button and then Double-click the parameter icon to open its display window. Click the Help button.

PATROL Central - From the console menu Microsoft Windows bar, choose Help => Edition Help Topics. Then double-click the name of your KM.

click the application class name. PATROL Central Web Edition Click the Help link, and Click the link for the Click the link for the then click the link for application class, and parameter, and then your KM. then click the Page Help click the Page Help link. link.

26

PATROL Knowledge Module for Unix User Guide

Chapter

Menu Summary
This chapter summarizes the application menus and menu commands for the PATROL Knowledge Module for Unix and Linux component. The application menu content is provided in a table for each of the application classes. This chapter describes Knowlege Module (KM) menu commands only. The KM help system provides further details about these menu commands. For descriptions of PATROL Console menu commands, see the appropriate PATROL user guide for your console. The following topics are discussed in this chapter: Location of KM Menu Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL Console for Unix. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL Console for Microsoft Windows Servers . . . . . . . . . . . . . . . . . . . . . . . . . Menu Summary Table Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Computer Instance Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . UNIX OS Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COLLECTORS Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COMPOSITES Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COMPOSITES_COLLECTOR Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CPU Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DCM Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISK Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FILESYSTEM Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HEALTH AT A GLANCE Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . KERNEL Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MEMORY Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NETWORK Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NFS Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROLAGENT Application Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PRINTER Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PROCESS Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PROCESS_PRESENCE (PROCCONT) Application Menu . . . . . . . . . . . . . . . . . . . . . . PROCPRES Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 29 29 29 30 31 33 33 33 34 34 34 35 36 36 36 37 38 38 39 41 42 42 43

Chapter 2 Menu Summary

27

SCHEDULER Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 SECURITY Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 SMP Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 SNMPHealth Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 SWAP Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 USERS Application Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

28

PATROL Knowledge Module for Unix User Guide

Location of KM Menu Commands

Location of KM Menu Commands

When a KM)is loaded, its associated menu commands are added to the KM area of a menu. Each PATROL Console presents them in a slightly different manner.

PATROL Console for Unix

The KM area of a menu is below the menus horizontal line. Menu commands above the line belong to the PATROL Console.

PATROL Console for Microsoft Windows Servers

KM menu commands appear as submenus under the KM Commands menu command.

Menu Summary Table Format

The tables in this chapter use the format described in the following table: Table 3 Menu Summary Table Format Action Description Description Description Description

Menu Item First-level menu Second-level menu Third-level menu Fourth-level menu

Chapter 2 Menu Summary

29

Computer Instance Menu

Computer Instance Menu

Access the Computer Instance menu as described in Accessing KM Commands and InfoBoxes on page 402. The Computer Instance menu has the following menu commands: Table 4 Computer Instance Menu Summary (Part 1 of 2) Action displays a dialog box prompting you for an operating system command and executes the specified command The output is displayed in the PATROL Console window. OS Task displays a dialog box prompting you for an operating system command and executes the specified command A task icon appears. PSL Command displays a dialog box prompting you for a PATROL Script Language (PSL) command and executes the specified command The output is displayed in the PATROL Console window. PSL Task displays a dialog box prompting you for a PSL command and executes the specified command A task icon appears. Utilities Patrol Refresh Process Cache Force Discovery Reinitialize Agent Process Kill Process File System Mount Unmount Find Files Like Remove Files Like runs PATROL, process, and file system utilities refreshes the process cache, force discover, or reinitialize the agent forces the PATROL Agent to reload its process cache from the kernel process list forces application discovery forces the PATROL Agent to restart provides process-related commands kills a process searches for and to mount and unmount file systems mounts file systems that PATROL has detected as offline or that have not been discovered unmounts a file system that PATROL detects as mounted or that have not been discovered lists all file system files whose names match the specified expression removes all file system files whose names match the specified expression

Menu Command OS Command

30

PATROL Knowledge Module for Unix User Guide

UNIX OS Application Menu

Table 4

Computer Instance Menu Summary (Part 2 of 2) Action forces an update of the application parameter values displays a menu item to save your current selection of PATROL KMs so that your KMs automatically load the next time you start the PATROL Agent saves your current selection of PATROL KMs so that the KMs automatically load the next time you start the PATROL Agent starts the SNMP Master Agent on Unix platforms only opens an xterm window for logging into a remote host or performing other system functions accesses and edit registered recovery actions

Menu Command Refresh Parameters Agent Quick Config

Make Current KM Preloaded Start SNMP Agent Login Configure Recovery Actions

UNIX OS Application Menu

Access the UNIX_OS application menu as described in Accessing KM Commands and InfoBoxes on page 402. The UNIX OS application menu has the following menu commands: Table 5 UNIX_OS Application Menu Summary (Part 1 of 3) Action forces an update of the application parameter values provides access to KM administration commands allows you to point the DCM application class to the location of the PERFORM Home directory if it is different than the location that PATROL expects, which is $PATROL_HOME/<platform_target>/ BEST1 displays commands for the DISK application specifies DISK instances that PATROL should ignore removes all DISK instances from the DISK exclusion list displays commands for the FILESYSTEM application specifies FILESYSTEM instances that PATROL should ignore removes all FILESYSTEM instances from the FILESYSTEM exclusion list displays commands for the PRINTER application specifies PRINTER instances that PATROL should ignore
:

Menu Command Refresh Parameters Knowledge Module Admin Specify PERFORM Home Directory

Toggle PSL/DCM Collection selects the collection method for the PATROL KM for Unix

DISK Application Edit DISK Exclusion List Clear DISK Exclusion List FILESYSTEM Application Edit FILESYSTEM Exclusion List Clear FILESYSTEM Exclusion List PRINTER Application Edit PRINTER Exclusion List

Chapter 2 Menu Summary

31

UNIX OS Application Menu

Table 5

UNIX_OS Application Menu Summary (Part 2 of 3) Action removes all PRINTER instances from the PRINTER exclusion list activates monitoring all PRINTER instances that are not in the PRINTER exclusion list (monitoring is inactive by default) stops monitoring of all PRINTER instances (monitoring is inactive by default) displays commands for the NETWORK application activates monitoring all NETWORK instances stops monitoring of all NETWORK instances displays commands for monitoring SNMPHealth activates monitoring of SNMPHealth stops monitoring of SNMPHealth (default setting) provides access to a diagnostic report and debug controls for various collectors gathers basic installation information such as the computers OS version, PATROL target, Perform target, PATROL Agent version, PATROL Console version, PATROL_HOME location, BEST1_HOME location, and license information generates a diagnostic report for Support purposes displays a dialog box that offers several debug options for the DCM functions and binary executable displays a dialog box that offers several debug options for the DISK application Clear PRINTER Exclusion List Activate Monitoring Deactivate Monitoring

Menu Command

NETWORK Application Activate Monitoring Deactivate Monitoring SNMPHealth Activate Monitoring Deactivate Monitoring Debug and Diagnostics Capture Unix Diagnostics

Create Diagnostic Report DCM Debug Admin DISK Debug Admin

FILESYSTEM Debug Admin displays a dialog box that offers several debug options for the FILESYSTEM application NETWORK Debug Admin NFS Debug Admin PRINTER Debug Admin SMP Debug Admin displays a dialog box that offers several debug options for the NETWORK application displays a dialog box that offers several debug options for the NFS application displays a dialog box that offers several debug options for the PRINTER application displays a dialog box that offers several debug options for Symmetric Multi-Processors (SMP) discovery and data collection displays a dialog box that offers several debug options for the SWAP application displays a dialog box that offers several debug options for the PROCESS application

SWAP Debug Admin PROCESS Debug Admin

32

PATROL Knowledge Module for Unix User Guide

COLLECTORS Application Menu

Table 5

UNIX_OS Application Menu Summary (Part 3 of 3) Action displays a report containing shared memory usage, semaphores usage, a list of current shared memory segments, and a list of current semaphores for the current host runs the BGS Collect Look utility

Menu Command IPC Resource Usage Report

BGS Command Look

COLLECTORS Application Menu

Access the COLLECTORS application menus described in Accessing KM Commands and InfoBoxes on page 402. The COLLECTORS application menu has the following menu commands: Table 6 COLLECTORS Application Menu Summary Action forces an update of the application parameter values sets up the correct access privilege to the utilities for accessing the DISK information using PSL collection methods displays a dialog box from which you can specify the number of printers to monitor

Menu Command Refresh Parameters Setup DISK Ping Script Setup PRINTERColl Option

COMPOSITES Application Menu

The COMPOSITES application menu has no KM-specific menu commands. For information on common menu items, see the PATROL Console Online Help.

COMPOSITES_COLLECTOR Application Menu

Access the COMPOSITES_COLLECTOR application menu as described in Accessing KM Commands and InfoBoxes on page 402. The Composites application menu has the following menu commands:

Chapter 2 Menu Summary

33

CPU Application Menu

Table 7

Composites Application Menu Summary Action creates a composite parameter using a wizard or by entering the information manually edits an existing composite parameter deletes an existing composite parameter

Menu Command Create Expressions Edit Expressions Delete Expressions

CPU Application Menu

The CPU application menu has no KM-specific menu commands. For information on common menu items, see the PATROL Console Online Help.

DCM Application Menu

Access the DCM application menus described in Accessing KM Commands and InfoBoxes on page 402. The DCM application menu has the following menu commands: Table 8 DCM Application Menu Summary Action forces an update of the application parameter values

Menu Command Refresh Parameters Setup PRINTERCOLL Option

DISK Application Menu

Access the DISK application menu as described in Accessing KM Commands and InfoBoxes on page 402. The DISK instance application menu has the following menu commands: Table 9 DISK Application Menu Summary Action stops monitoring the selected instance adds a DISK instance to the list of DISKs that PATROL ignores removes all DISK instances from the DISK exclusion list

Menu Command Stop Monitoring Edit DISK Exclusion List Clear DISK Exclusion List

34

PATROL Knowledge Module for Unix User Guide

FILESYSTEM Application Menu

FILESYSTEM Application Menu

Access the FILESYSTEM application menu as described in Accessing KM Commands and InfoBoxes on page 402. The FILESYSTEM application menu has the following menu commands: Table 10 FILESYSTEM Application Menu Summary Action accesses the Maintenance menu that performs file system activities mounts file systems that PATROL has detected as offline or that have not been discovered You must be root to run this command. remounts the selected instance prompts you to specify the file system to mount unmounts a file system that PATROL detects as mounted or that has not been discovered You must be root to run this command. unmounts the selected instance prompts you to specify the file system to unmount flushes all previously written buffers to disk lists all file system files whose names match the specified expression removes all file system files whose names match the specified expression stops monitoring the selected instance adds a FILESYSTEM instance to the list of FILESYSTEMs that PATROL ignores removes all FILESYSTEM instances from the FILESYSTEM exclusion list removes the current FILESYSTEM instance from the FILESYSTEM exclusion list

Menu Command Maintenance Mount

Automatic Manual Unmount

Automatic Manual Sync Find Files Like Remove Files Like Stop Monitoring Edit FILESYSTEM Exclusion List Clear FILESYSTEM Exclusion List Delete from FILESYSTEM Exclusion List

Chapter 2 Menu Summary

35

HEALTH AT A GLANCE Application Menu

HEALTH AT A GLANCE Application Menu

The HEALTH AT A GLANCE application menu has no KM-specific menu commands. For information on common menu items, see the PATROL Console Online Help.

KERNEL Application Menu

The KERNEL application menu has no KM-specific menu commands. For information on common menu items, see the PATROL Console Online Help.

MEMORY Application Menu

Access the MEMORY application menu as described in Accessing KM Commands and InfoBoxes on page 402. The MEMORY application menu has the following menu commands: Table 11 MEMORY Application Menu Summary Action displays information about the shared memory segments lists information about the active semaphores deletes a specified shared memory segment deletes a specified semaphore

Menu Command List Shared Memory Segments List Semaphores Remove Shared Memory Segments Remove Semaphores

36

PATROL Knowledge Module for Unix User Guide

NETWORK Application Menu

NETWORK Application Menu

To access the NETWORK application menu as described in Accessing KM Commands and InfoBoxes on page 402. The NETWORK application menu has the following menu commands: Table 12 TCP/IP Reports List Hosts List Protocols List Services List Trusted Hosts Utilities Ftp Ping Rlogin Query Name Server ARP Show Arp Cache Add Host To Cache UUCP Reports List Server Name Show Log Show Statistics Utilities UUCP Check UUCP Clean UUCP List Network Interface List Memory Buffer List Network Address NETWORK Application Menu Summary (Part 1 of 2) Action views reports and executes TCP/IP utilities displays network and NFS activity reports in a task window lists the hosts on the network displays the list of known protocols and protocol numbers provides a list of available services lists TCP/IP trusted hosts provides access to network utilities launches the ftp (file transfer protocol) utility launches the ping utility on the specified computer launches a remote login to a specified computer launches the internet query name server program adds and deletes hosts from the Address Resolution Protocol (ARP) table displays the hosts and addresses in the ARP table adds hosts to the ARP table views reports and executes UUCP utilities displays UUCP reports in a task window displays the UUCP server name displays the UUCP log displays UUCP statistics provides access to UUCP utilities You must be root to run these commands. performs a UUCP check performs a UUCP clean displays a command entry dialog box in which you can enter UUCP changes displays the state of an automatically configured interface displays memory management statistics displays all the network addresses

Menu Command

Delete Host From Cache deletes hosts from the ARP table

Chapter 2 Menu Summary

37

NFS Application Menu

Table 12

NETWORK Application Menu Summary (Part 2 of 2) Action displays the routing table and statistics for the network displays the statistics for each protocol layer destroys the network icon or sets an alarm if the network interface is down

Menu Command List Routing Table Show Protocol Statistics Toggle Alarm/Destroy if Down

NFS Application Menu

Access the NFS application menu as described in Accessing KM Commands and InfoBoxes on page 402. The NFS application menu has the following menu commands: Table 13 NFS Reports List Servers Zero NFS Counters NIS Reports List Accounts List Groups List Servers NFS Application Menu Summary Action displays NFS options displays report options lists all the NFS servers for the system in a task window resets the NFS counters to zero You must be root to run this command. displays NIS options displays report options displays a list of all NIS accounts in a task window displays a list of all NIS groups in a task window displays a list of all NIS servers in a task window

Menu Command

PATROL Application Menu

Access the PATROL application menu as described in Accessing KM Commands and InfoBoxes on page 402. The PATROL application menu has the following menu commands: Table 14 PATROL Application Menu Summary Action forces an update of the application parameters

Menu Command Refresh Parameters

38

PATROL Knowledge Module for Unix User Guide

PATROLAGENT Application Menu

PATROLAGENT Application Menu

To access the PATROLAGENT application menu as described in Accessing KM Commands and InfoBoxes on page 402. The PATROLAGENT application menu has the following menu commands: Table 15 PATROLAGENT Application Menu Summary (Part 1 of 2) Action displays the contents of the agent logs fine-tunes the agent's performance displays the current setting of the agent tuning variables sets agent tuning variables sets the interval for the system to refresh the process cache sets both simple and PSL discovery poll times sets agent, user, and process cache priority as well as schedule process cache priority sets the priority for the PATROL Agent. sets the priority that the process cache has for gathering process information sets the priority for user commands controls agent execution sets the length of time, in seconds, between successive agent executions sets the time interval (increment) to be used in scheduling parameter execution sets the maximum value that the process scheduling delta can be set sets up the scheduling for agent connection using the following scheduling submenus The next scheduled execution will occur at the current execution end time. The next scheduled execution will occur at the prior execution end time. The next scheduled execution will be automatically determined by the agent to execute at the optimal time.

Menu Command View Agent Log Tune Agent Show Tuning Variables Set Agent Variables Set Get Processes Cycle (appl) Set Appl Check Cycle Set Priority Agent Process Cache

Process Cache Schedule sets the priority of the agent run queue User Set Run Queue Delta Delta Increment Maximum Delta Schedule Policy Schedule From End Schedule From Prev Schedule Automatic

Schedule Force Delta The next scheduled execution will occur at the prior execution start time. Reset Agent to Default Setting resets the agents variables to their default settings

Chapter 2 Menu Summary

39

PATROLAGENT Application Menu

Table 15

PATROLAGENT Application Menu Summary (Part 2 of 2) Action displays information about the processes running on the agent displays all process information that is in the run queue displays all processes that are currently running displays all the task information that is currently in the queue displays all the applications that are registered by PATROL displays the applications that have been discovered by PATROL displays all the parameters registered by PATROL displays a backward trace of the agent's error log displays the current setting of the agent tuning variables searches the agent's process cache executes a refresh of the parameters and process cache and performs application discovery causes the agent to execute all application discovery scripts forces the PATROL Agent to reload its process cache from the kernel process forces an update of the application parameter values

Menu Command List Agent Information Runable Processes Running Processes Executing Task Commands Registered Applications Discovered App Instances Registered Parameters Error Backtrace List Agent Tuning Variables Search Agent Process Cache Trigger Agent Activities Perform Application Discovery Refresh Process Cache Refresh Parameters

40

PATROL Knowledge Module for Unix User Guide

PRINTER Application Menu

PRINTER Application Menu

Access the PRINTER application menu as described in Accessing KM Commands and InfoBoxes on page 402. The PRINTER application menu has the following menu commands: Table 16 PRINTER Application Menu Summary Action lists all the print jobs for the selected printer lists all the print jobs for all monitored printers deletes the specified print job deletes all the print jobs for the selected printer disables the selected printer enables the selected printer disables the selected print queue enables the selected print queue moves the specified print job to the specified printer moves all the print jobs from the selected PRINTER to a specified printer stops monitoring the selected instance adds a PRINTER instance to the list of PRINTER instances that PATROL ignores removes all PRINTER instances from the PRINTER exclusion list

Menu Command List Print Jobs List All Print Jobs Delete Job Delete All Jobs Stop Printer Start Printer Stop Queue Start Queue Move Job To Another Printer Move All Jobs to Another Printer Stop Monitoring Edit PRINTER Exclusion List Clear PRINTER Exclusion List

Chapter 2 Menu Summary

41

PROCESS Application Menu

PROCESS Application Menu

Access the PROCESS application menu as described in Accessing KM Commands and InfoBoxes on page 402. The PROCCONT application menu has the following menu commands: Table 17 PROCESS Application Menu Summary Action displays information about the zombie processes that persist on the system displays processes that use the CPU excessively decreases the execution priority of the specified process decreases the execution priority of all processes in the specified PROCESS group decreases the execution priority for all processes that the specified USER launches forces an update of the application parameter values

Menu Command List Zombie Process CPU Hogs Renice Process Renice Group Renice User Refresh Parameters

For more information about monitoring processes, see Chapter 14, Monitoring Processes.

PROCESS_PRESENCE (PROCCONT) Application Menu

Access the PROCESS_PRESENCE application menus described in Accessing KM Commands and InfoBoxes on page 402. The PROCCONT application menu has the following menu commands: Table 18 PROCESS_PRESENCE Application Menu Summary (Part 1 of 2) Action

Menu Command

Manage List of Monitored use to add, modify, and report on monitored processes in your Processes environment Manage Host Group Synchronize Group List use to add, modify, and report on watched hosts in your environment use to synchronize group lists and process lists synchronizes settings between hosts within a group. Also, specifies group communications (such as timeout settings, and whether to accept or send changes made on one host to the rest of the group). synchronizes the list of monitored processes for hosts within a group

Process List

42

PATROL Knowledge Module for Unix User Guide

PROCPRES Application Menu

Table 18

PROCESS_PRESENCE Application Menu Summary (Part 2 of 2) Action accesses the menu options that allow you to use and configure the ProcSpring Migration Tools provides information about the ProcSpring Migration Tools other ProcSpring menu commands provides options for full or partial migration of ProcSpring log entries removes all related ProcSpring pconfig entries and deactivates ProcSpring restores previous ProcSpring settings and reactivates ProcSpring specifies a directory where PATROL will look for process presence flag files

Menu Command ProcSpring Migration Tools About ProcSpring Migration Tools Migrate ProcSpring Deactivate ProcSpring w/o Migrating Reactivate ProcSpring Identify Flag Directory

PROCPRES Application Menu

Access the PROCPRES application menu commands as described in Accessing KM Commands and InfoBoxes on page 402. The PROCPRES application menu has the following menu commands: Table 19 PROCPRES Application Menu Summary Action displays all the individual processes currently being monitored for process presence

Menu Command Display Included Processes

Chapter 2 Menu Summary

43

SCHEDULER Application Menu

SCHEDULER Application Menu

Access the SCHEDULER application menu as described in Accessing KM Commands and InfoBoxes on page 402. The SCHEDULER application menu has the following menu commands: Table 20 SCHEDULER Application Menu Summary Action schedules a task to run at once or at regular intervals suspends monitoring of applications during routine maintenance or other activities that generate false alarms establishes in which directory the KM is stores blackout scripts writes debug information to the system output window

Menu Command Schedule Tasks Schedule Blackout Set Default Blackout Script Directory Scheduling Debug

SECURITY Application Menu

Access the SECURITY application menu as described in Accessing KM Commands and InfoBoxes on page 402. The SECURITY application menu has the following menu commands: Table 21 SECURITY Application Menu Summary Action provides listings of the group and password files and shows the failed logins displays all of the files that have the suid or the sgid (guid) bit set displays information about each user who tries and fails to execute an su or msu command displays all users who have multiple sessions active displays all files that can be written to by any user

Menu Command Administration List SUID and SGID Files List Failed su/msu Logins

List Users Without Password displays all users who do not have a password List Duplicate User Id Entry List Files With Global Write

SMP Application Menu

The SMP application has no KM-specific menu commands. For information on common menu items, see the PATROL Console Online Help.

44

PATROL Knowledge Module for Unix User Guide

SNMPHealth Application Menu

SNMPHealth Application Menu

Access the SNMPHealth application menu as described in Accessing KM Commands and InfoBoxes on page 402. The SNMPHealth application menu has the following menu commands: Table 22 SNMPHealth Application Menu Summary (Part 1 of 2) Action provides a way to discover the SNMP configuration of PATROL
:

Menu Command SNMP Configuration

Change SNMP Query Port changes the SNMP Query Port, which is used by external processes to make SNMP queries against the PATROL Agent List Configuration Information Start/Stop SNMP Agent Start SNMP Master Agent displays SNMP-related information that includes PATROL Agent namespace variables relating to SNMP, statuses of the master agent and subagent, and port numbers provides a way to manually start and stop components of the PATROL SNMP system starts the SNMP Master Agent When the PATROL Agent starts, the computer applications SNMPStart parameter automatically starts the SNMP Master Agent. This command is used to restart the master agent after it has been intentionally or unintentionally stopped. starts the SNMP subagent The SNMPHEALTH application classs snmp_subagent_start parameter attempts to start the subagent if it is not running. This command is used to restart the subagent between the snmp_subagent_starts scheduled poll time. stops the SNMP subagent This command changes the state of the SNMPHEALTH application classs snmp_subagent_monitor parameter to ALARM. listens for SNMP traps on the port specified in the agent configuration variable /snmp/trap_port, whose default value is 162 If port 162 is in use and you do not change the configuration variable, this command does not function properly. displays a dialog box that sends a customized trap in which you define the traps host, port, community, and four strings displays a dialog box that queries the PATROL Agent displays a dialog box that queries your computers platform knowledge module

Start SNMP sub-agent

Stop SNMP sub-agent

Test PATROL SNMP System provides several ways to test your PATROL SNMP system Listen for SNMP Traps

Generate Manual SNMP Trap SNMP Query PATROL Master Agent SNMP Query PATROL platform.km

Chapter 2 Menu Summary

45

SWAP Application Menu

Table 22

SNMPHealth Application Menu Summary (Part 2 of 2) Action displays a dialog box that enters a host name and object ID (OID) and runs the PSL function snmp_h_get( ) If no SNMP Managers are listening on port 161, the command times out.

Menu Command SNMP Query Using snmp_h_get

Walk PATROL SNMP Mib displays a dialog box that uses BMC systems object ID to view the PATROL Agent SNMP management information base (MIB) if the PATROL Agent is listening on the designated port Test PATROL SNMP System Set SNMPH_testparm OK Set SNMPH_testparm WARN Set SNMPH_testparm ALARM provides a means to test whether automatic SNMP traps are generated when PATROL issues an event and changes an objects state changes the state of the SNMPHEALTH application classs testparm parameter to OK and issues a trap if the SNMP system is set up changes the state of the SNMPHEALTH application classs testparm parameter to WARN and issues a trap if the SNMP system is set up changes the state of the SNMPHEALTH application classs testparm parameter to ALARM and issues a trap if the SNMP system is set up

SWAP Application Menu

The SWAP application menu has no KM-specific menu commands. For information on common menu items, see the PATROL Console Online Help.

46

PATROL Knowledge Module for Unix User Guide

USERS Application Menu

USERS Application Menu

Access the USERS application menu as described in Accessing KM Commands and InfoBoxes on page 402. The USERS application menu has the following menu commands: Table 23 USERS Application Menu Summary Action displays all the print jobs that belong to the specified user displays all the processes that belong to the specified user displays the amount of disk space used by the specified user displays a list of users that are currently active on the host

Menu Command List Print Jobs Process List Disk Space List Current Users

Chapter 2 Menu Summary

47

USERS Application Menu

48

PATROL Knowledge Module for Unix User Guide

Chapter

Parameter Summary
This chapter provides a summary of parameters for the PATROL KM for Unix and Linux. Refer to the PATROL user guide for your console for additional information about the different types of parameters and their functions. See the PATROL KM for Unix and Linux online Help system for details about KM-specific parameters. The following topics are discussed: Parameter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Parameter Defaults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Chapter 3 Parameter Summary

49

Parameter Summary

Parameter Summary
The PATROL KM for Unix and Linux has various parameters that provide statistical information about resources, operating status, and performance. Table 24 lists each application class and its associated parameters. The table also provides information that you can use when selecting or reviewing the appropriate parameters used in monitoring the PATROL KM. Table 24
Parameter ALL_COMPUTERS Application BlockSize converts OS memory to 1-KB blocks Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux BootTime holds a time stamp of the last time the system was rebooted Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux DFBlockSize converts OS disk block values to 1-KB blocks Platforms: Solaris, HP, RS6000, SINIX, Linux Get_700_nproc determines the hardware model (700/800) and HP-UX version on pre-HP-UX 10 model 700 machines Platforms: HP KernelInfo collects information about the kernel, such as when it was last built Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux LPSpoolDirectory maintains a list of possible locations for print queues This parameter is used by PRINTER discovery to limit instance creation to printers that have locally defined queues. Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux MaxConfigValLen contains the maximum number of characters permitted in the value of a variable stored in the agent configuration database and is used for edit checking Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux OSName collects information about the operating system Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux PageSize converts OS page units into 1-KB page units Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux page 84 page 84 page 82 page 82 page 81 page 81 page 81 page 80 page 81

PATROL KM for Unix and Linux Parameter Summary (Part 1 of 30)

Description See Also

50

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 2 of 30)

Description clears the screen for a parameter that generates text output Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux See Also page 85

SetupClearText

SNMPStart

starts the SNMP Master Agent Platforms: Windows, Solaris, HP, OSF1, RS6000, SINIX, Linux

page 86

TerminalEmulator

sets the terminal emulator that PATROL should use; an xterm window is the default on most platforms Platforms: Solaris, HP, OSF1, RS6000, SINIX, Linux

page 86

COLLECTORS CollectorStatus CollectorStatusText DFColl HAAGColl NETColl generates an alert if the pconfig variable DCMStatusText exists page 80 and saves its contents in the CollectorStatusText parameter receives the contents of the pconfig variable DCMStatusText page 80 executes the df command, collects the output, and then sets the page 81 values of the FILESYSTEM parameters collects data for the HEALTH_AT_A_GLANCE application classes executes the netstat command to collect information about incoming packets, outgoing packets, and errors and sets NETWORK parameters page 81 page 83

NFSColl PRINTERColl

executes the nfsstat command to collect information about NFS page 83 and sets the values of the various NFS parameters executes the lpstat -t command, determines the status of each queue, and displays information about all the local and remote printers you are monitoring collects kernel information on the Linux platform and sets KERFileUsedPercent and KERINodeUsedPercent calculates the top 10 processes that use the most cpu resources and sets the PROCTopProcs parameter executes the sar command and sets various parameters Platforms: AIX, HP-UX Note: On AIX systems, the account that executes the sar command must belong to the adm group. page 84

PROCColl PSColl SARColl

page 84 page 85 page 85

SMPColl

executes the mpstat command and collects information about each individual CPU on the system This collector is different from the SARColl and VMColl because these parameter look at multiple CPUs as a whole and report a single value. The SMPColl collector then sets the values of the SMP parameters. Platforms: HP-UX, Solaris

page 85

Chapter 3 Parameter Summary

51

Parameter Summary

Table 24
Parameter UPTColl

PATROL KM for Unix and Linux Parameter Summary (Part 3 of 30)

Description executes the uptime command and sets the CPULoad parameter collects information to set USERS parameters gathers operating system information related to memory by executing the vmstat command and sets various parameters collects information for the defined composite parameters displays the results of the composite parameter calculation for the specified composite parameter Collector: CPARMCollector See Also page 86 page 86 page 86

USRPROCColl VMColl

COMPOSITES_COLLECTOR Application CPARMCollector COMPOSITES Application CPARMResult page 80 page 80

CPARMExpression

displays the composite parameter definition Collector: CPARMCollector

page 80

CPARMGeneratedCodeText

displays the PSL that was generated by the Composites expression generator and used to determine the composite parameter value Collector: CPARMCollector

page 80

CPU Application Class CPUCpuUtil CPUIdleTime CPUInt displays the percentage of time the CPU was being used displays the percentage of time that the CPU was idle displays the number of non-VME device interrupts A non-VME interrupt is an interrupt that does not originate from the backplane. Collector: VMColl Note: When using DCM collection, PATROL deactivates this parameter on all platforms. When using PSL collection, PATROL deactivates this parameter on AIX. CPULoad displays the 1-minute load average from the uptime command Collector: UPTColl CPUProcSwch displays the total number of CPU context switches per second Collector: SARColl or VMColl Note: When using DCM collection, PATROL deactivates this parameter on all platforms. CPURunQSize displays the number of processes in the run queue (RunQ) page 81 page 80 page 80 page 80 page 80 page 80

52

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 4 of 30)

Description displays the percentage of CPU time spent in system mode doing system tasks, including the time consumed by calls to kernel routines Collector: DCMColl, SARColl or VMColl See Also page 81

CPUSysTime

CPUUserTime CPUWio

displays the percentage of CPU time currently being spent in user mode executing commands and performing user tasks displays the percentage of CPU time spent waiting for input and output operations Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1) and Solaris.

page 81 page 81

DCM Application Class PATROL creates this application class only if the PATROL KM for Unix is using the DCM collection method. CPUCollStatus CPUCtrl DCMColl DCMMessages indicates whether the CPU data is collected by the DCM or PSL page 80 controls the poll time and status (active or inactive) for DCM collection of CPU application parameter data passes data from the Data Collection Manager binary program (DCM) to many parameters displays status messages for the DCM process Anytime the DCM_Alert parameter goes into alarm, PATROL documents the problem in this parameter. It also records start/stop/restart events for the DCM process. This parameter manages the number of lines in the text so that it keeps the last 'N' messages where 'N' is determined as follows:
s

page 80 page 81 page 81

If the agent namespace variable /DCM/DCM/DCMStatus/maxLength is set, then that value is used. Otherwise, the parameter uses the PCONFIG variable /DCM/DCM/DCMStatus/maxLength if it is available. Otherwise, the parameter will default to 40 lines.

Once the max number of lines is reached, the parameter uses a FIFO mechanism so that the oldest messages are removed to make room for the newest messages.

Chapter 3 Parameter Summary

53

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 5 of 30)

Description monitors the status of the DCM process and the global channel used by the KM to communicate with the DCM process When it detects a problem, it logs a message to the DCMStatus parameter and closes the channel. Once the channel is closed, the DCMColl parameter exits. The agent reschedules DCMColl. On its next execution, the DCMColl parameter restarts the DCM process. See Also page 81

DCMMonitor

DCMStatus DCMUpdateCount

displays the health of the DCM process displays the number of parameter values sent to the PATROL Agent from the DCM process It provides one measure of the performance of the PATROL Agent/DCM interface. The number parameter values varies based on the characteristics of your system (for example, number of hard disks) and the configuration of the DCM application class. Both the polling rate and the active/inactive state of the DCM control parameters have an effect on how many parameters values are updated at any one time. DCMUpdateTime is set each time DCMUpdateCount is set.

page 81 page 81

DCMUpdateTime

displays the number of seconds spent by PATROL Agent processing a data update from the DCM process It provides one measure of the performance of the PATROL Agent/DCM interface. The time varies depending on the number of data values included in one update and the other factors on your system. Particularly, the configuration of the UNIX KM and the PATROL Agent with respect to the parameter history file. Anytime the PATROL Agent writes or flushes a significant amount of data to disk, it is possible that this parameter will record an update time that is much longer than normal.

page 81

DFColl DiskCollStatus DISKCtrl HAAGColl KernelCollStatus KERNELCtrl MemCollStatus

executes the df command and collects the output. Then, it then sets the values of the FILESYSTEM parameter. controls the poll time and status (active or inactive) for DCM collection of DISK application parameter data gathers data on the systems overall CPU, File System, and Swap file usage indicates whether kernel data is collected by the DCM or PSL controls the poll time and status (active or inactive) for DCM collection of KERNEL application parameter data

page 81

indicates whether the disk data is collected by the DCM or PSL page 81 page 81 page 81 page 81 page 82

indicates whether memory data is collected by the DCM or PSL page 82

54

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 6 of 30)

Description controls the poll time and status (active or inactive) for DCM collection of MEMORY application parameter data executes the netstat command to collect information about incoming packets, outgoing packets, and errors and sets NETWORK parameters See Also page 82 page 83

MEMORYCtrl NETColl

NFSColl PRINTERColl

executes the nfsstat command to collect information about NFS page 83 and sets the values of the various NFS parameters executes the lpstat -t command, determines the status of each queue, and displays information about all the local and remote printers you are monitoring collects kernel information on the Linux platform and sets KERFileUsedPercent and KERINodeUsedPercent page 84

PROCColl ProcessCollStatus

page 84

indicates whether process data is collected by the DCM or PSL. page 84 Issues an alarm if the method of collecting data automatically changes from DCM to PSL. This indicates that the Perform collector is not collecting data from consumer parameters. controls the poll time and status (active or inactive) for DCM collection of PROCESS application parameter data page 84

PROCESSCtrl PSColl SARColl

calculates the top 10 processes that use the most CPU resources page 85 and sets the PROCTopProcs parameter page 85 executes the sar command and collects the output. Then, it parses the output and performs the necessary data conversions, such as converting page size to a standard 1-K page. Finally, this collector sets the values of the appropriate PATROL parameters executes the mpstat command and collects information about page 85 each individual CPU on the system. The SMPColl collector then sets the values of the SMP parameters indicates whether swap data is collected by the DCM or PSL controls the poll time and status (active or inactive) for DCM collection of SWAP application parameter data executes the uptime command indicates whether user data is collected by the DCM or PSL controls the poll time and status (active or inactive) for DCM collection of USERS application parameter data collects operating system information relating to users. The UserProcess collector calculates the following information:
s s s s

SMPColl

SwapCollStatus SWAPCtrl UPTColl UserCollStatus USERSCtrl USRPROColl

page 86 page 86 page 86 page 86 page 86 page 86

the number of users currently on the system the number of user sessions the total number of user processes on the system the average number of user processes

Chapter 3 Parameter Summary

55

Parameter Summary

Table 24
Parameter VMColl

PATROL KM for Unix and Linux Parameter Summary (Part 7 of 30)

Description See Also page 86 gathers operating system information related to system memory and executes the vmstat command and collects the output. Then, it parses the output and performs the necessary data conversions, such as converting page size to a standard 1-K page changes the label for a DISK instance from "sdNN" to "sdNN (cNNtNNdNN)." For example, changes the disk label from sd3 to sd31 (c1t1d0). This parameter makes the disk label more user-friendly by showing both valid label names for a given disk. page 80

DISK Application Class ChangeDiskLabel

DSKAvgQueue

average number of requests in the queue Collector: DMColl Platforms: All except AIX and Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1).

page 81

DSKAvgServ

displays the average service time, in milliseconds, for the device page 81 to complete transfer requests For a disk, it includes seek time, rotational latency, and transfer time. Collector: DMColl Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on AIX and Compaq Tru64 (Digital Unix, OSF/1).

DSKAvgWait

displays the average time that transfer requests wait in the queue for the device Collector: VMColl Platforms: HP-UX, Solaris

page 81

DSKBps

displays the number of 1-KB blocks read from or written to the page 81 device per second Collector: DCMColl

DSKMsps

displays the average disk seek time for the device and indicates page 81 the speed of the device Collector: VMColl

56

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 8 of 30)

Description See Also displays the percentage of time that the device is busy servicing page 81 a transfer request Collector: DCMColl Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1).

DSKPercentBusy

DSKRead

displays the number of kilobytes read from disk per second Collector: DCMColl Platforms: AIX, Solaris, Linux Note: When using PSL collection, PATROL deactivates this parameter on Solaris.

page 81

DSKReadWrite

displays the number of read and write pages made to/from the page 81 device per second Collector: DCMColl Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1).

DSKSps

displays the number of disk seeks per second Collector: DCMColl Platforms: Compaq Tru64 (Digital Unix, OSF/1), HP-UX Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1).

page 81

DSKTps

displays the number of disk transfers per second Collector: DCMColl Platforms: AIX, Compaq Tru64 (Digital Unix, OSF/1), HP-UX, Solaris

page 81

DSKWrite

displays the number of kilobytes written to disk per second Collector: DCMColl Platforms: AIX, Linux, Solaris Note: When using PSL collection, PATROL deactivates this parameter on Solaris.

page 81

Chapter 3 Parameter Summary

57

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 9 of 30)

Description displays the amount of space available for this file system instance Collector: DFColl See Also page 81

FILESYSTEM Application Class FSAvailableSpace

FSAvailableSpaceMB

displays the amount of space available in megabytes for this file page 81 system instance Collector: DFColl

FSCapacity

displays the percentage of the file system storage that is currently in use Collector: DFColl

page 81

FSFreeInodes

displays the number of unused i-nodes on the local file system Collector: DFColl

page 81

FSInodeUsedPercent

displays the percentage of i-nodes used Collector: DFColl

page 81

FSMountStatus

displays the mount status of the file system Collector: DFColl

page 81

FSUsedSpace

displays the amount of used space for the file system instance Collector: DFColl

page 81

HEALTH AT A GLANCE Application Class HAAGCPUUsage displays whether or not the systems overall CPU usage is satisfactory Collector: HAAGColl HAAGFilesystemUsage displays whether or not the systems overall File System usage is satisfactory. Collector: HAAGColl HAAGHealthSummary reports on the system's overall health based on the other HAAG page 81 parameters: HAAGCPUUsage, HAAGFilesystemUsage, HAAGSwapUsage Collector: HAAGColl HAAGSwapUsage displays whether or not the systems overall swap file usage is satisfactory Collector: HAAGColl page 81 page 81 page 81

58

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 10 of 30)

Description See Also

KERNEL Application Class KERDirBlk displays the number of directory-block reads issued per second page 81 Collector: SARColl Platforms: AIX, HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms. KERFileUsedPercent displays the percentage of kernel file slots being used This parameter monitors file table use. Collector: DCMColl or SARColl Platforms: AIX, HP-UX, Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms except HP-UX. When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1). KERIGet displays the number of files located by an i-node entry Collector: SARColl Platforms: AIX, HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms. KERINodeUsedPercent displays the percentage of kernel i-node slots being used Collector: DCMColl or SARColl Platforms: All except HP, AIX, Linux, and Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms except Compaq Tru64 (Digital Unix, OSF/1) and HP-UX. When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1). KERLgAlloc displays the number of bytes allocated for large memory requests Collector: SARColl KERLgFail displays the number of large memory pool requests that were not satisfied Collector: SARColl page 82 page 82 page 82 page 82 page 81

Chapter 3 Parameter Summary

59

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 11 of 30)

Description displays the amount of memory in bytes available in the KERNEL Memory Allocation (KMA) large memory request pool Collector: SARColl See Also page 82

KERLgMem

KERLockUsedPercent

displays the percentage of kernel lock slots being used Collector: SARColl

page 82

KERMsg

displays the number of message operations performed per second You can use this parameter to monitor applications that use message operations to communicate with other processes. Collector: DCMColl or SARColl Platforms: All except Compaq Tru64 (Digital Unix, OSF/1) and Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms except AIX and Compaq Tru64 (Digital Unix, OSF/1). When using PSL collection, PATROL deactivates this parameter on Solaris.

page 82

KERMsgNumQueUsed Percent

displays the percentage of the total number of available message queues being used system-wide Collector: DCMColl Platforms: all except AIX and Compaq Tru64 (Digital Unix, OSF/1)

page 82

KERMsgNumUsedPercent

displays the percentage of the total number of available kernel messages that are being used system-wide Collector: DCMColl Platforms: all except AIX

page 82

KERNameI

displays the number system pathname searches being performed Collector: SARColl Platforms: HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 82

60

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 12 of 30)

Description displays the amount of memory dynamically allocated for oversized requests Collector: SARColl See Also page 82

KEROvzAlloc

KEROvzFail

displays the number of requests for oversized memory that could not be satisfied Collector: SARColl

page 82

KERProcUsedPercent

displays the percentage of kernel process slot used. This parameter monitors the process table utilization Collector: DCMColl or SARColl Platforms: All except AIX and Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms except Compaq Tru64 (Digital Unix, OSF/1) and HP-UX. When using PSL collection, PATROL deactivates this parameter on Solaris.

page 82

KERSemNumUsedPercent

displays the percentage of the total number of available semaphores used system-wide Collector: DCMColl Platforms: All except AIX and Compaq Tru64 (Digital Unix)

page 82

KERSemNumSetUsedPercent

displays the percentage of the total number of available semaphore sets that are being used system-wide Collector: DCMColl

page 82

KERSemOps

displays the number of semaphore operations per second and monitors applications that use semaphore operations to communicate with other processes Collector: DCMColl or SARColl Platforms: All except Compaq Tru64 (Digital Unix, OSF/1) and Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms except AIX, Compaq Tru64 (Digital Unix, OSF/1), and Solaris. When using PSL collection, PATROL deactivates this parameter on Solaris.

page 82

KERShmNumUsedPercent

displays the percentage of the total number of available shared memory segments that are being used system-wide Collector: DCMColl

page 82

Chapter 3 Parameter Summary

61

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 13 of 30)

Description displays the amount of memory in bytes allocated to the small memory request pool by KERNEL Memory Allocation. Collector: SARColl See Also page 82

KERSmlAlloc

KERSmlFail

displays the number of small memory requests that failed Collector: SARColl

page 82

KERSmlMem

displays the amount of memory in bytes that is available in the page 82 small memory request pool allocated by KERNEL Memory Allocation Collector: SARColl

KERSysCall

displays the number of system calls made per second This number is the total of all read, write, fork, and exec system calls.

page 82

MEMORY Application Class MEMActiveVirPage displays the number of active virtual pages Collector: VMColl Note: When using DCM collection, PATROL deactivates this parameter on all platforms. When using PSL collection, PATROL deactivates this parameter on AIX (32 bit). MEMAddrTransFault displays the number of pages that were not in physical memory page 82 when accessed Collector: SARColl; VMColl for OSF/1 Platforms: Compaq Tru64 (Digital Unix, OSF/1) Note: When using DCM collection, PATROL deactivates this parameter on all platforms. MEMBRead displays the number of physical reads per second from the disk page 82 to the buffer cache Collector: SARColl Platforms: AIX Note: When using DCM collection, PATROL deactivates this parameter on all platforms. page 82

62

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter MEMBWrt

PATROL KM for Unix and Linux Parameter Summary (Part 14 of 30)

Description See Also displays the number of physical writes per second from the disk page 82 to the system buffer cache Collector: SARColl Platforms: AIX Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

MEMCow

displays the number of page write faults that have occurred Collector: VMColl Platforms: Compaq Tru64 (Digital Unix, OSF/1) Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 82

MEMFreeMem

displays the number of pages of memory available Collector: DCMColl, SARColl or VMColl Note: When using PSL collection, PATROL deactivates this parameter on AIX (32-bit) and Solaris.

page 82

MEMLRead

displays the number of logical blocks read per second from the page 82 system buffer Collector: DCMColl or SARColl Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1) and Solaris.

MEMLWrt

displays the number of writes per second to the system buffer Collector: DCMColl or SARColl Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1) and Solaris.

page 82

MEMPageAnticipated

displays the anticipated short-term memory shortfall in memory pages Collector: VMColl Platforms: HP-UX, Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 82

Chapter 3 Parameter Summary

63

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 15 of 30)

Description See Also displays the number of pages that were added to the free list by page 82 the page-stealing daemon Collector: VMColl or SARColl Platforms: All except Compaq Tru64 (Digital Unix, OSF/1) and Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms. When using PSL collection, PATROL deactivates this parameter on AIX (32-bit).

MEMPageFreed

MEMPageIn

displays the number of pages that have been swapped in from secondary memory Collector: SARColl or VMColl Note: When using PSL collection, PATROL deactivates this parameter on AIX (32-bit).

page 82

MEMPageOut

displays the number of pages swapped out of main memory to page 82 secondary memory Collector: SARColl or VMColl Note: When using PSL collection, PATROL deactivates this parameter on AIX (32-bit).

MEMPageScanned

displays the number of pages scanned per second by the pagestealing daemon to find more memory Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1).

page 82

MEMPFault

displays the number of page protection faults that caused pages page 82 to be copied Note: When using PSL collection, PATROL deactivates this parameter on AIX, Compaq Tru64 (Digital Unix, OSF/1), HPUX, and Solaris.

MEMPgFil

displays the number of page faults that have been reclaimed by page 82 bringing the pages in from the file system Collector: SARColl

64

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 16 of 30)

Description displays the number of raw reads per second from character device (TTYs) Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1) and Solaris. See Also page 82

MEMPRead

MEMPWrt

displays the number of raw writes per second to character devices (TTYs) Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1) and Solaris.

page 83

MEMRCache

displays the percentage of read misses in the buffer cache Collector: SARColl Platforms: AIX, HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 83

MEMWCache

displays the percentage of logical blocks written in the buffer cache Indicates the effectiveness of system buffering Collector for group A platforms: DCMColl Collector for group B platforms: SARColl Platforms: All except Linux Note: When using PSL collection, PATROL deactivates this parameter on Compaq Tru64 (Digital Unix, OSF/1) , AIX, and Solaris.

page 83

MEMWire

displays the number of 1-KB memory pages that cannot be paged out of main memory This parameter shows locked memory. Collector: VMColl Platforms: Compaq Tru64 (Digital Unix, OSF/1) Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 83

Chapter 3 Parameter Summary

65

Parameter Summary

Table 24
Parameter MEMZero

PATROL KM for Unix and Linux Parameter Summary (Part 17 of 30)

Description This parameter displays the number of memory pages that are zero-filled on demand. Collector: VMColl Platforms: Compaq Tru64 (Digital Unix, OSF/1) Note: When using DCM collection, PATROL deactivates this parameter on all platforms. See Also page 83

NETWORK Application Class NETCollisionPrc displays the percentage of network send collisions Collector: NETColl NETInErrPrc displays the number of incoming data packets containing errors page 83 per second Collector: NETColl NETOutErrPrc displays the number of outgoing data packets containing errors page 83 per second Collector: NETColl NETPacketsIn displays the total number of incoming packets per second Collector: NETColl NETPacketsOut displays the total number of outgoing packets per second Collector: NETColl NETPseudoAddresses displays the virtual address assigned to a network interface Collector: NETWORK Discovery Platforms: All except Linux NFS Application Class NFSCBadCall displays the percentage of all NFS client calls that failed (since the last sample) Collector: NFSColl NFSCCall displays the total number of NFS client calls made (since the last page 83 sample) Collector: NFSColl NFSCCreate displays the percentage of all NFS client calls made to create a new file (since the last sample) Collector: NFSColl page 83 page 83 page 83 page 83 page 83 page 83

66

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 18 of 30)

Description displays the percentage of all NFS client calls made to retrieve file attributes or files statistics (since the last sample) Collector: NFSColl See Also page 83

NFSCFsStat

NFSCGetAttr

displays the percentage of all NFS client calls that are requests to get file attributes (since the last sample) Collector: NFSColl

page 83

NFSCLink

displays the percentage of all NFS client calls made to create hard links (since the last sample) Collector: NFSColl

page 83

NFSCLookUp

displays the percentage of all NFS client calls made to look up directory paths (since the last sample) Collector: NFSColl

page 83

NFSCMkDir

displays the percentage of all NFS client calls that are calls made page 83 to create directories (since the last sample) Collector: NFSColl

NFSCNclGet

displays the number of times that the client had to request a new client handle for an NFS call (since the last sample) Collector: NFSColl

page 83

NFSCNclSleep

displays the number of times that the kernel had to wait to obtain client structure information (since the last sample) Collector: NFSColl Platforms: All except Linux and Solaris

page 83

NFSCNull

displays the percentage of NFS client calls that request no action page 83 other than acknowledgement (since the last sample) Collector: NFSColl

NFSCRead

displays the number of all NFS client calls that are calls to read data from files (since the last sample) Collector: NFSColl

page 83

NFSCReadDir

displays the percentage of all NFS client calls made to read directories (since the last sample) Collector: NFSColl

page 83

NFSCReadLink

displays the percentage of NFS client calls made to read symbolic links (since the last sample) Collector: NFSColl

page 83

Chapter 3 Parameter Summary

67

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 19 of 30)

Description displays the percentage of all NFS client calls made to remove files (since the last sample) Collector: NFSColl See Also page 83

NFSCRemove

NFSCRename

displays the percentage of all NFS client calls made to rename files (since the last sample) Collector: NFSColl

page 83

NFSCRmDir

displays the percentage of all NFS client calls made to remove directories (since the last sample) Collector: NFSColl

page 83

NFSCRpcBadCall

displays the percentage of all NFS client RPC calls that returned page 83 an error other than time-outs or interruptions (since the last sample) Collector: NFSColl

NFSCRpcBadXid

displays the percentage of client RPC calls for which multiple responses were received (since the last sample) Collector: NFSColl

page 83

NFSCRpcCall

displays the total number of NFS client RPC calls made to the selected server by all clients (since the last sample) Collector: NFSColl

page 83

NFSCRpcNewCred

displays the number of client RPC calls that are calls made to refresh authentication information (since the last sample) Collector: NFSColl Platforms: All except Linux

page 83

NFSCRpcNullRecv

displays the number of empty messages received from TCP or LAN (since the last sample) Collector: NFSColl

page 83

NFSCRpcRetrans

displays the number of NFS client RPC requests that had to be retransmitted (since the last sample) Collector: NFSColl

page 83

NFSCRpcTimeOut

page 83 displays the percentage of all client RPC calls that timed out because no client handles were available (since the last sample) Collector: NFSColl

68

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 20 of 30)

Description displays the number of times that a client request had to wait because a client handle was unavailable (since the last sample) Collector: NFSColl Platforms: All except Linux See Also page 83

NFSCRpcWait

NFSCSetAttr

displays the percentage of all NFS client calls made to store file page 83 attributes (since the last sample) Collector: NFSColl

NFSCSymLink

displays the percentage of all NFS client calls made to create symbolic links (since the last sample) Collector: NFSColl

page 83

NFSCWrite

displays the number of all NFS client calls made to write files (since the last sample) Collector: NFSColl

page 83

NFSSBadCall

displays the percentage of all NFS server calls that failed (since page 83 the last sample) Collector: NFSColl

NFSSCall

displays the total number of NFS server calls made (since the last sample) Collector: NFSColl

page 84

NFSSCreate

displays the percentage of all NFS server calls made to create a new file (since the last sample) Collector: NFSColl

page 84

NFSSFsStat

displays the percentage of all NFS server calls made to retrieve file attributes or files statistics (since the last sample) Collector: NFSColl

page 84

NFSSGetAttr

displays the percentage of all NFS server calls that are requests page 84 to get file attributes (since the last sample) Collector: NFSColl

NFSSLink

displays the percentage of all NFS server calls made to create hard links (since the last sample) Collector: NFSColl

page 84

NFSSLookUp

displays the percentage of all NFS server calls made to look up directory paths (since the last sample) Collector: NFSColl

page 84

Chapter 3 Parameter Summary

69

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 21 of 30)

Description displays the percentage of all NFS server calls made to create directories (since the last sample) Collector: NFSColl See Also page 84

NFSSMkDir

NFSSNull

displays the percentage of NFS server calls that request no action other than acknowledgment (since the last sample) Collector: NFSColl

page 84

NFSSRead

displays the percentage of all NFS server calls made to read data from files (since the last sample) Collector: NFSColl

page 84

NFSSReadDir

displays the percentage of all NFS server calls made to read data directories (since the last sample) Collector: NFSColl

page 84

NFSSReadLink

displays the percentage of all NFS server calls made to read symbolic links (since the last sample) Collector: NFSColl

page 84

NFSSRemove

displays the percentage of NFS server calls made to remove files page 84 (since the last sample) Collector: NFSColl

NFSSRename

displays the percentage of all NFS server calls made to rename files (since the last sample) Collector: NFSColl

page 84

NFSSRmDir

displays the percentage of all NFS server calls made to remove directories (since the last sample) Collector: NFSColl

page 84

NFSSRpcBadCall

displays the percentage of all NFS RPC server requests rejected page 84 by the server RPC facility before the request could be passed to the kernel NFS service routines (since the last sample) Collector: NFSColl

NFSSRpcBadLen

displays the total number of server RPC calls that are truncated page 84 and have an invalid length (since the last sample) Collector: NFSColl

NFSSRpcCall

displays the total number of NFS server RPC calls (since the last page 84 sample) Collector: NFSColl

70

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 22 of 30)

Description displays the number of null RPC calls that the server received (since the last sample) Collector: NFSColl See Also page 84

NFSSRpcNullRecv

NFSSRpcXdrCall

displays the total number of NFS server RPC calls whose headers could not be XDR-decoded (since the last sample) This network problem occurs mostly on wide area networks (WANs). Collector: NFSColl

page 84

NFSSSetAttr

displays the percentage of all NFS server calls made to store file page 84 attributes (since the last sample) Collector: NFSColl

NFSSSymLink

displays the percentage of all NFS server calls made to create symbolic links (since the last sample) Collector: NFSColl

page 84

NFSSWrite

displays the percentage of all NFS server calls made to write files (since the last sample) Collector: NFSColl

page 84

PATROLAGENT Application Class PADeltaBetweenExecSecs PAHistoryIndexSize PAOutstandingJobs PATotalErrors PAUserErrors PAWorkRateExecsMin PRINTER Application Class PRNQLength displays the number of print jobs waiting in the print queue Collector: PrinterColl PROCCONT Application Class groupMonitorDaemon PROCPPStatus collects data that is gathered and analyzed by the PROCPPStatus parameter lists the status of monitored processes and host groups Collector: groupMonitorDaemon page 81 page 85 page 84 displays the number of seconds between executions. This value page 84 is the RUNQ_DELTA variable displays the size of the PatrolAgent history index file in bytes page 84 displays the number of currently executing jobs on the PATROL page 84 Agent and the time that they were started displays the total number of errors detected by the PATROL Agent displays the total number of errors detected by the PATROL Agent displays the number of instruction executions performed per minute by the PATROL Agent page 84 page 84 page 84

Chapter 3 Parameter Summary

71

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 23 of 30)

Description displays the average number of processes per nonroot user Collector: DCMColl or USRPROCColl See Also page 84

PROCESS Application Class PROCAvgUsrProc

PROCExec

displays the number of exec system calls per second Collector: SARColl Platforms: AIX, HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 84

PROCNoZombies

displays the total number of zombie processes Collector: DCMColl or USRPROCColl

page 84

PROCNumProcs

displays the total number of active processes on the system Collector: DCMColl or USRPROCColl

page 84

PROCProcWait

displays the number of processes that are currently waiting for resources Collector: DCMColl or VMColl Platforms: AIX, Compaq Tru64 (Digital Unix, OSF/1), HP-UX, Solaris Note: When using PSL collection, PATROL deactivates this parameter on AIX (32-bit) and Compaq Tru64 (Digital Unix, OSF/1).

page 85

PROCProcWaitInt

displays the number of processes in an interruptible wait state Collector: VMColl Platforms: Compaq Tru64 (Digital Unix, OSF/1) Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

PROCProcWaitUnint

displays the number of processes in an uninterruptible wait state Collector: VMColl Platforms: Compaq Tru64 (Digital Unix, OSF/1) Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

72

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 24 of 30)

Description displays the processes (up to ten) using the highest percentage of CPU time during the sample period Machines with multiple CPUs may show more than 100% utilization for a given process. Collector: DCMColl or PSColl Note: When using PSL collection, PATROL deactivates this parameter on AIX. See Also page 85

PROCTopProcs

PROCUserProcs

displays the total number of nonroot user processes Collector: DCMColl or USRPROCColl

page 85

PROCPRES Application Class PROCPPCPUPerc displays the percentage of CPU used by the selected process (calculated on the number of active CPUs in the system. Collector: DCMColl PROCPPCount displays the number of processes started with the same command-line string as the selected process Collector: DCMColl PROCPPCountCheck indicates whether the process count is within the acceptable range specified by the user Collector: discovery PROCPPMem displays the amount of virtual memory that the process is using page 85 Collector: DCMColl PROCPPOwnerCheck displays the results of a verification that a process is owned by an authorized user name PATROL issues an alert if it discovers that the process is owned by an unauthorized user name. Collector: discovery PROCPPParentPID1 displays the results of a verification that the parent process ID (PPID) for a designated process is 1 Processes with a PPID of 1 are owned by the Unix scheduler, or init. Collector: discovery page 85 page 85 page 84 page 84 page 84

Chapter 3 Parameter Summary

73

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 25 of 30)

Description gathers the information that is displayed in the PATROL Scheduling Report You can access the PATROL Scheduling Report by viewing the scheduler parameter. See Also page 85

SCHEDULER Application Class scheduler_daemon

scheduler

displays the PATROL Scheduling Report This report provides information about items in the schedule queue. Collector: scheduler_daemon

page 85

SMP Application Class SMPContextSwitch displays the number of CPU context switches Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms. SMPCrossCalls displays the number of interprocessor cross-calls Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms. SMPIdlePercent displays the percentage of time the CPU was idle Collector: DCMColl Platforms: All except Reliant and Sequent DYNIX/ptx SMPInterrupts displays the number of processor interrupts Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms. page 85 page 85 page 85 page 85

74

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 26 of 30)

Description displays the number of interrupts as threads, excluding clock interrupts Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms. See Also page 85

SMPIntThread

SMPInvContSwitch

displays the number of processor involuntary context switches Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

SMPMajorFaults

displays the number of major faults Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

SMPMinorFaults

displays the number of minor faults Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

SMPRunQLen1Min

displays the average number of processes in the CPU's run queue during the last minute Collector: SMP discovery Platforms: HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

Chapter 3 Parameter Summary

75

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 27 of 30)

Description displays the average number of processes in the CPU's run queue during the last 5 minutes Collector: SMP discovery Platforms: HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms. See Also page 85

SMPRunQLen5Min

SMPRunQLen15Min

displays the average number of processes in the CPU's run queue during the last 15 minutes Collector: SMP discovery Platforms: HP-UX Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

SMPSpinMutex

displays the number of spins on mutexes, or locks that were not page 85 acquired on the first try Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

SMPSpinRdWr

displays the number of reader and writer locks that were not acquired on the first try Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 85

SMPSystemCalls

displays the number of system calls Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 86

SMPSystemPrcnt

displays the amount of processor time spent on system activities Collector: DCMColl

page 86

76

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 28 of 30)

Description displays the number of thread migrations to another processor Collector: SMPColl Platforms: Solaris Note: When using DCM collection, PATROL deactivates this parameter on all platforms. See Also page 86

SMPThMigration

SMPUserPercent

displays the amount of processor time spent on user-requested activities Collector: DCMColl

page 86

SMPWaitPercent

displays the percentage of time the processor spent waiting Collector: DCMColl

page 86

SNMPHealth Application Class SNMPH_diagnostics gathers and displays diagnostic information relating to the PATROL SNMP system In addition to being displayed in the text parameter window, the diagnostic information is also written to the file, SNMPHealth.log in $PATROL_HOME/log. SNMPH_monitorSubAgent monitors the subagent's status If the subagent is running, PATROL sets its value to OK. If the subagent is not running, PATROL sets the value to ALARM. SNMPH_startSubAgent uses the output of snmp_agent_config to determine whether the page 86 SNMP subagent is running If the subagent is not running, PATROL issues the PSL command snmp_agent_start() to start the subagent. SNMPH_testparm performs diagnostics on the PATROL SNMP system Use it to test the automatic SNMP traps that are issued through the event management system when an event is generated. This parameter's test uses Standard Event 11, which is triggered in PATROL when a parameter's value falls within an alarm range. It uses Standard Event 9 when the parameter's value falls within the OK range after having been in the alarm range. Collector: menu command page 86 page 86 page 86

Chapter 3 Parameter Summary

77

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 29 of 30)

Description displays the number of swap pages available on the swap device The page size is represented by the n in the parameter name and is measured in kilobytes (32, 64, 128, 256, 512, 1024, and 2048). Collector: SWAP Discovery displays the size of the available swap space Collector: SWAP Discovery Platforms: All except Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms. page 86 See Also page 86

SWAP Application Class SWPnPageSizeAvail SWP32PageSizeAvail SWP64PageSizeAvail SWP128PageSizeAvail SWP256PageSizeAvail SWP512PageSizeAvail SWP1024PageSizeAvail SWP2048PageSizeAvail SWPSwapFreeSpace

SWPSwapSize

displays the size of a particular swap area Collector: SWAP Discovery Platforms: All except Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 86

SWPSwapUsedPercent

displays the percentage of swap space used for a particular swap area Collector: SWAP Discovery Platforms: All except Linux Note: When using DCM collection, PATROL deactivates this parameter on all platforms.

page 86

SWPTotSwapFreeSpace

displays the total amount of free swap space Appears under the Summary instance only. Collector: DCMColl, SWAP Discovery or SARColl

page 86

SWPTotSwapSize

displays the size of the entire swap space (On Solaris, this value includes the space used in /tmp.) Appears under the Summary instance only. Collector: DCMColl or SWAP Discovery

page 86

78

PATROL Knowledge Module for Unix User Guide

Parameter Summary

Table 24
Parameter

PATROL KM for Unix and Linux Parameter Summary (Part 30 of 30)

Description displays the percentage of the total system-wide swap space in use of active user sessions on this machine (including root sessions) Appears under the Summary instance only. Collector: DCMColl, SWAPCtrl, or SWAP Discovery See Also page 86

SWPTotSwapUsedPercent

USERS Application Class USRNoSession displays the number of active user sessions on this machine (including root sessions) Collector: DCMColl USRNoUser displays the number of unique users that are currently logged on to the system, NOT the number of user sessions Collector: DCMColl page 86 page 86

Chapter 3 Parameter Summary

79

Parameter Defaults

Parameter Defaults
Table 25 lists default values for parameters. Interpret the column headings as follows. Depending on the type of parameter, some information is not applicable.
Parameter Active? Type specifies the parameter name specifies whether the parameter is active or inactive when discovered specifies whether the parameter is a Standard, Consumer, or Collector parameter specifies the thresholds for the first alarm Alarm 1 This information is not applicable to Collectors. specifies the thresholds for the second alarm Alarm 2 Scheduling Icon Units History Level See Also Page This information is not applicable to Collectors. specifies the time interval in the poll cycle specifies whether the icon is a graph, gauge, or text box specifies the type of unit in which the parameter output is expressed, such as a percentage, a number, or bytes specifies the length of time that history is retained specifies other pages in this chapter where you can find more information about the parameter

Table 25

PATROL KM for Unix and Linux Parameter Defaults (Part 1 of 7)

Scheduling See Also
page 50 page 50 page 56 page 51 page 51 page 52 page 52 page 52 page 52 page 53 page 53 page 52 page 52 page 52 page 52 page 52

Active?

Alarm1

Alarm2

Parameter
BlockSize BootTime ChangeDiskLabel CollectorStatus CollectorStatusText CPARMCollector CPARMResult CPARMExpression CPARMGeneratedCodeText CPUCollStatus CPUCtrl CPUCpuUtil CPUIdleTime CPUInt CPULoad CPUProcSwch y y y y y y y y y y y y y y y y

standard standard collector collector collector collector consumer consumer consumer consumer standard consumer consumer consumer consumer consumer

und. und. und. 1 to 1 und. und. 1 und. und. und. und. 90 to 95 und. und. und. und.

und. und. und. 2 to 2 und. und. 2 und. und. und. und. 95 to 100 und. und. und. und.

0 23:59:59 10 min 1 min NA 30 sec NA NA NA 3 min 1 min NA NA NA NA NA

none none none none none none stoplight text text graph none graph graph graph graph graph

NA NA NA NA NA NA NA NA NA DCM or PSL NA percent percent interrupts load switches

80

PATROL Knowledge Module for Unix User Guide

Units

Type

Icon

Parameter Defaults

Table 25

PATROL KM for Unix and Linux Parameter Defaults (Part 2 of 7)

Scheduling See Also
page 52 page 53 page 53 page 53 page 53 page 53 page 54 page 54 page 54 page 54 page 50 page 54 page 54 page 54 page 56 page 56 page 56 page 56 page 56 page 57 page 57 page 57 page 57 page 57 page 57 page 58 page 58 page 58 page 58 page 58 page 58 page 58 page 50 page 71 page 51 page 58 page 58 page 58 page 58 page 54 page 50 page 59 page 59

Active?

Alarm1

Alarm2

Parameter
CPURunQSize CPUSysTime CPUUserTime CPUWio DCMColl DCMMessages DCMMonitor DCMStatus DCMUpdateCount DCMUpdateTime DFBlockSize DFColl DiskCollStatus DISKCtrl DSKAvgQueue DSKAvgServ DSKAvgWait DSKBps DSKMsps DSKPercentBusy DSKRead DSKReadWrite DSKSps DSKTps DSKWrite FSAvailableSpace FSAvailableSpaceMB FSCapacity FSFreeInodes FSInodeUsedPercent FSMountStatus FSUsedSpace Get_700_nproc groupMonitorDaemon HAAGColl HAAGCPUUsage HAAGFilesystemUsage HAAGHealthSummary HAAGSwapUsage KernelCollStatus KernelInfo KERDirBlk KERFileUsedPercent y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y

consumer consumer consumer consumer collector consumer collector consumer consumer consumer standard collector consumer standard consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer standard collector collector consumer consumer consumer consumer consumer standard consumer consumer

und. 90 to 95 90 to 95 und. und. und. NA und. und. und. und. und. und. NA und. und. und. und. und. und. und. und. und. und. und. und. und. 0 to 0 und. und. 1 to 1 und. und. NA NA und. und. und. und. und. und. und. 90 to 95

und. 95 to 100 95 to 100 und. und. und. NA und. und. und. und. und. und. NA und. und. und. und. und. und. und. und. und. und. und. und. und. 0 to 0 und. und. 2 to 2 und. und. NA NA und. und. und. und. und. und. und. 95 to 100

NA NA NA NA 1 min 3 min 1 min NA NA NA 0 2 min 3 min 1 min NA NA none none none none none none none none none none none none none none 2 min none none 10 min 1 min none none none none 3 min 23:59:59 none none

graph graph graph graph none text none graph graph graph none none graph none graph graph graph graph graph gauge graph graph graph graph graph graph graph graph graph graph boolean graph text none none stoplight stoplight text stoplight graph none graph graph

processes percent percent percent NA none NA NA values sent seconds NA NA DCM or PSL NA requests milliseconds milliseconds 1-KB blocks milliseconds percent kilobytes pages per second seeks transfers kilobytes kilobytes megabytes percent i-nodes percent mount status megabytes NA NA NA NA NA NA NA DCM or PSL NA reads percent

Chapter 3 Parameter Summary

Units

Type

Icon

81

Parameter Defaults

Table 25

PATROL KM for Unix and Linux Parameter Defaults (Part 3 of 7)

Scheduling See Also
page 59 page 59 page 59 page 59 page 60 page 60 page 60 page 60 page 60 page 60 page 54 page 61 page 61 page 61 page 61 page 61 page 61 page 61 page 62 page 62 page 62 page 62 page 50 page 50 page 62 page 62 page 62 page 63 page 54 page 63 page 63 page 63 page 63 page 55 page 63 page 64 page 64 page 64 page 64 page 64 page 64 page 65

Active?

Alarm1

Alarm2

Parameter
KERIGet KERINodeUsedPercent KERLgAlloc KERLgFail KERLgMem KERLockUsedPercent KERMsg KERMsgNumQueUsedPercent KERMsgNumUsedPercent KERNameI KERNELCtrl KEROvzAlloc KEROvzFail KERProcUsedPercent KERSemNumSetUsedPercent KERSemNumUsedPercent KERSemOps KERShmNumUsedPercent KERSmlAlloc KERSmlFail KERSmlMem KERSysCall LPSpoolDirectory MaxConfigValLen MEMActiveVirPage MEMAddrTransFault MEMBRead MEMBWrt MemCollStatus MEMCow MEMFreeMem MEMLRead MEMLWrt MEMORYCtrl MEMPageAnticipated MEMPageFreed MEMPageIn MEMPageOut MEMPageScanned MEMPFault MEMPgFill MEMPRead y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y

consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer standard consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer standard standard consumer consumer consumer consumer consumer consumer consumer consumer consumer standard consumer consumer consumer consumer consumer consumer consumer consumer

und. 90 to 100 und. und. und. 90 to 95 und. 90 to 95 90 to 95 und. und. und. und. 90 to 95 90 to 95 90 to 95 80 to 100 90 to 95 und. und. und. und. und. und. und. und. und. und. und. und. 0 to 100 und. und. und. und. 5 to 25 und. 10 to 15 und. und. und. und.

und. und. und. und. und. 95 to 100 und. 95 to 100 95 to 100 und. und. und. und. 95 to 100 95 to 100 95 to 100 und. 95 to 100 und. und. und. und. und. und. und. und. und. und. und. und. 100 to 200 und. und. und. und. 25 to 100 und. 15 to 100 und. und. und. und.

none none none none none none none none none none 1 min none none none none none none none none none none none 23:59:59 10 min (one time) none none none none 3 min none none none none 1 min none none none none none none none none

graph graph graph graph graph gauge graph graph graph graph none graph graph graph graph graph graph graph graph graph graph graph none none graph graph graph graph graph graph graph graph graph none graph graph graph graph graph graph graph graph

files percent kilobytes requests kilobytes percent messages percent percent conversion NA kilobytes requests percent percent percent sem. ops. percent kilobytes requests kilobytes sys calls NA NA pages pages reads/sec writes/sec DCM or PSL number of 1-Kb pages kilobytes reads writes/sec NA number of 1-Kb pages pages pages pages pages page faults pages reads/sec

82

PATROL Knowledge Module for Unix User Guide

Units

Type

Icon

Parameter Defaults

Table 25

PATROL KM for Unix and Linux Parameter Defaults (Part 4 of 7)

Scheduling See Also
page 65 page 65 page 65 page 65 page 66 page 55 page 66 page 66 page 66 page 66 page 66 page 66 page 66 page 66 page 66 page 67 page 67 page 67 page 67 page 67 page 67 page 67 page 67 page 51 page 67 page 67 page 67 page 68 page 68 page 68 page 68 page 68 page 68 page 68 page 68 page 68 page 68 page 69 page 69 page 69 page 69 page 69

Active?

Alarm1

Alarm2

Parameter
MEMPWrt MEMRCache MEMWCache MEMWire MEMZero NETColl NETCollisionPrc NETInErrPrc NETOutErrPrc NETPacketsIn NETPacketsOut NETPseudoAddresses NFSCBadCall NFSCCall NFSCCreate NFSCFsStat NFSCGetAttr NFSCLink NFSCLookUp NFSCMkDir NFSCNclGet NFSCNclSleep NFSCNull NFSColl NFSCRead NFSCReadDir NFSCReadLink NFSCRemove NFSCRename NFSCRmDir NFSCRpcBadCall NFSCRpcBadXid NFSCRpcCall NFSCRpcNewCred NFSCRpcNullRecv NFSCRpcRetrans NFSCRpcTimeOut NFSCRpcWait NFSCSetAttr NFSCSymLink NFSCWrite NFSSBadCall y y y y y n y y y y y n y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y

consumer consumer consumer consumer consumer collector consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer collector consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer

und. 1 to 10 1 to 10 und. und. und. und. und. und. und. und. und. und. und. 25 to 50 und. und. und. 50 to 60 und. und. und. 2 to 50 und. und. 18 to 30 und. 25 to 50 und. und. 40 to 50 und. und. und. und. und. 20 to 50 und. und. 11 to 50 50 to 75 40 to 50

und. 10 to 20 10 to 20 und. und. und. und. und. und. und. und. und. und. und. 50 to 100 und. und. und. 60 to 100 und. und. und. 50 to 100 und. und. 30 to 100 und. 50 to 100 und. und. 50 to 100 und. und. und. und. und. 50 to 100 und. und. 50 to 100 75 to 100 50 to 100

none none none none none 5 min none none none none none none none none none none none none none none none none none 3 min none none none none none none none none none none none none none none none none none none

graph graph graph graph graph none graph graph graph graph graph text graph graph graph graph graph graph graph graph graph graph graph none graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph

writes/sec percent percent number of 1-KB pages number of 1-KB pages NA percent percent percent packets packets NA percent calls percent percent percent percent percent percent nfs call calls percent NA calls percent percent percent percent percent percent percent calls calls messages calls percent calls percent percent calls percent

Chapter 3 Parameter Summary

Units

Type

Icon

83

Parameter Defaults

Table 25

PATROL KM for Unix and Linux Parameter Defaults (Part 5 of 7)

Scheduling See Also
page 69 page 69 page 69 page 69 page 69 page 69 page 70 page 70 page 70 page 70 page 70 page 70 page 70 page 70 page 70 page 70 page 70 page 71 page 71 page 71 page 71 page 71 page 50 page 71 page 50 page 71 page 71 page 71 page 71 page 71 page 51 page 71 page 72 page 51 page 55 page 72 page 55 page 72 page 72 page 73 page 73 page 73

Active?

Alarm1

Alarm2

Parameter
NFSSCreate NFSSFsStat NFSSGetAttr NFSSCall NFSSLink NFSSLookUp NFSSMkDir NFSSNull NFSSRead NFSSReadDir NFSSReadLink NFSSRemove NFSSRename NFSSRmDir NFSSRpcBadCall NFSSRpcBadLen NFSSRpcCall NFSSRpcNullRecv NFSSRpcXdrCall NFSSSetAttr NFSSSymLink NFSSWrite OSName PADeltaBetweenExecSecs PageSize PAHistoryIndexSize PAOutstandingJobs PATotalErrors PAUserErrors PAWorkRateExecMin PrinterColl PRNQLength PROCAvgUsrProc PROCColl ProcessCollStatus PROCExec PROCESSCtrl PROCNoZombies PROCNumProcs PROCPPCount PROCPPCountCheck PROCPPCPUPerc y y y y y y y y y y y y y y y y y y y y y y y y y y y y y y n y y y y y y y y y y y

consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer standard standard standard standard standard standard standard standard collector consumer consumer collector consumer consumer standard consumer consumer consumer consumer consumer

25 to 50 50 to 75 75 to 100 und. und. 50 to 60 und. 2 to 50 und. 18 to 30 40 to 50 25 to 50 und. und. 40 to 50 und. und. und. und. und. 25 to 50 50 to 75 und. und und. 900000 to 950000 und. und. und. 20 to 25 und. 30 to 100 und. und. 1 to 2 und. und. und. und. und. 1 to 1 und.

50 to 100 75 to 100 und. und. und. 60 to 100 und. 50 to 100 und. 30 to 100 50 to 100 50 to 100 und. und. 50 to 100 und. und. und. und. und. 50 to 100 75 to 100 und. und. und. 950000 to 999999 und. und. und. 25 to 60 und. und. und. und. und. und. und. und. und. und. 2 to 2 und.

none none none none none none none none none none none none none none none none none none none none none none 23:59:59 15 min 0 10 min 5 min 45 min 30 min 2 min 2 min none none 5 min 3 min none 1 min none none NA NA NA

graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph graph none graph none graph graph gauge gauge graph none gauge graph none graph gauge none graph graph graph stoplight graph

percent percent percent calls percent percent percent percent percent percent percent percent percent percent percent calls calls calls calls percent percent percent NA seconds NA NA jobs errors errors execs NA jobs processes NA DCM or PSL calls NA processes processes number of processes NA percentage

84

PATROL Knowledge Module for Unix User Guide

Units

Type

Icon

Parameter Defaults

Table 25

PATROL KM for Unix and Linux Parameter Defaults (Part 6 of 7)

Scheduling See Also
page 73 page 73 page 73 page 71 page 72 page 72 page 72 page 73 page 73 page 51 page 51 page 51 page 74 page 74 page 51 page 74 page 74 page 74 page 74 page 75 page 75 page 75 page 75 page 75 page 76 page 76 page 76 page 76

Active?

Alarm1

Alarm2

Parameter
PROCPPMem PROCPPOwnerCheck PROCPPParentPID1 PROCPPStatus PROCProcWait PROCProcWaitInt PROCProcWaitUnint PROCTopProcs PROCUserProcs PSColl SARColl SetupClearText scheduler scheduler_daemon SMPColl SMPContextSwitch y y y y y y y y y y y y y y y y

consumer consumer consumer consumer consumer consumer consumer consumer consumer collector collector standard consumer collector collector consumer

und. 1-1 1-1 und. und. und. und. und. und. und. und. und. und. und. und. und.

und. 2-2 2-2 und. und. und. und. und. und. und. und. und. und. und. und. und.

NA NA NA none none none none none none 1 min 1 min 10 min (once) none 1 hrs 10 min none

graph stoplight stoplight text graph graph graph text graph none none none text none none graph

kilobytes NA NA NA processes processes processes NA processes NA NA NA NA NA NA number of switches per second number of events per second percent number of interrupts per second number of interrupts per second number of switches per second number of major faults per second number of minor faults per second number of processes per minute number of processes per 5 minutes number of processes per 15 minutes number of events per second number of events per second

SMPCrossCalls

consumer

und.

und.

none

graph

SMPIdlePercent SMPInterrupts

y y

consumer consumer

und. und.

und. und.

none none

gauge graph

SMPIntThread

consumer

und.

und.

none

graph

SMPInvContSwitch

consumer

und.

und.

none

graph

SMPMajorFaults

consumer

und.

und.

none

graph

SMPMinorFaults

consumer

und.

und.

none

graph

SMPRunQLen1Min

consumer

und.

und.

none

graph

SMPRunQLen5Min

consumer

und.

und.

none

graph

SMPRunQLen15Min

consumer

und.

und.

none

graph

SMPSpinMutex

consumer

und.

und.

none

graph

SMPSpinRdWr

consumer

und.

und.

none

graph

Chapter 3 Parameter Summary

Units

Type

Icon

85

Parameter Defaults

Table 25

PATROL KM for Unix and Linux Parameter Defaults (Part 7 of 7)

Scheduling See Also
page 76 page 76 page 77 page 77 page 77 page 77 page 77 page 77 page 77 page 51 page 55 page 55 page 78 page 78 page 78 page 78 page 78 page 78 page 78 page 78 page 78 page 78 page 78 page 78 page 79 page 51 page 52 page 55 page 55 page 79 page 79 page 52 page 52

Active?

Alarm1

Alarm2

Parameter
SMPSystemCalls y

consumer

und.

und.

none

graph

number of system calls per second percent number of migrations per second percent percent NA NA NA NA NA DCM or PSL NA pages pages pages pages pages pages pages number of 1-KB blocks number of 1-KB blocks percent 1-KB blocks 1-KB blocks percent NA NA DCM or PSL NA sessions users NA NA

SMPSystemPrcnt SMPThMigration

y y

consumer consumer

und. und.

und. und.

none none

gauge graph

SMPUserPercent SMPWaitPercent SNMPH_diagnostics SNMPH_monitorSubAgen SNMPH_startSubAgent SNMPH_testparm SNMPStart SwapCollStatus SWAPCtrl SWP1024PageSizeAvail SWP128PageSizeAvail SWP2048PageSizeAvail SWP256PageSizeAvail SWP32PageSizeAvail SWP512PageSizeAvail SWP64PageSizeAvail SWPSwapFreeSpace SWPSwapSize SWPSwapUsedPercent SWPTotSwapFreeSpace SWPTotSwapSize SWPTotSwapUsedPercent TerminalEmulator UPTColl UserCollStatus USERSCtrl USRNoSession USRNoUser USRPROCColl VMColl

y n y y y y y y y y y y y y y y y y y y y y y y y y y y y y

consumer consumer standard standard standard consumer standard consumer standard consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer consumer standard collector consumer standard consumer consumer collector collector

und. und. und. 0 to 0 und. 1 to 1 und. und. und. und. und. und. und. und. und. und. und. und. und. und. und. 90 to 95 und. und. und. und. und. und. und. und.

und. und. und. und. und. 2 to 2 und. und. und. und. und. und. und. und. und. und. und. und. und. und. und. 95 to 100 und. und. und. und. und. und. und. und.

none none 10 min 1 min 1 min none 30 min 1 min 1 min none none none none none none none none none none none none none NA 1 min 3 min 1 min none none 5 min 45 sec

gauge gauge text stoplight none stoplight none graph none graph graph graph graph graph graph graph graph graph gauge graph graph gauge none none graph none graph graph none none

86

PATROL Knowledge Module for Unix User Guide

Units

Type

Icon

Chapter

Managing Data Collection

On some platforms, the PATROL Knowledge Module for Unix and Linux product gives you the ability to select the method PATROL uses to gather data for the PATROL KM for Unix and Linux. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Object Hierarchies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 COLLECTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 DCM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 DCM Collection Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Data Collection Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Collector Parameters for DCM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 -Ctrl and -Coll Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 DCM Collection Exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 PSL Collection Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Collector Parameters for COLLECTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 -Coll Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 PSL Collection Exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Managing Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Changing Data Collection Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Activating an Inactive Application Class. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Activating an Inactive Collector Parameter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Significant Events Related to Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

Chapter 4 Managing Data Collection

87

Introduction

Introduction
Two data collection methods are available to the PATROL Knowledge Module for Unix and Linux. The two methods use separate data collection mechanisms and are controlled by separate application classes. Table 26 lists the collection methods and their application classes. Table 26 Method DCM Collection Method PSL Collection Method Data Collection Methods and Associated Application Classes Application Class DCM Application Class COLLECTORS Application Class

Object Hierarchies
In PATROL, collector parameters are not visually represented by icons. However, the collector application classes are. When their respective data collection methods are employed, PATROL displays the following application class object hierarchies.

COLLECTORS
The COLLECTORS application class governs the PSL data collection method. It does not contain any consumer parameters. When the PSL Collection method is employed, PATROL creates the application class in the UNIX OS container (Figure 2). Figure 2 COLLECTORS Application Object Hierarchy
OS container
UNIX OS

Application class

COLLECTORS

88

PATROL Knowledge Module for Unix User Guide

Object Hierarchies

DCM
The DCM application class governs the DCM data collection method. The DCM application class contains several parameters that monitor the status of the DCM process as shown in the application class hierarchy represented by Figure 3. Figure 3 DCM Application Object Hierarchy

Chapter 4 Managing Data Collection

89

DCM Collection Method

DCM Collection Method

The DCM collection method uses the PATROL Perform collection model for gathering and presenting application information. In this model, the Perform Agent makes calls to the operating system to collect information about the system and stores that information in its own data structures. Special standard parameters, designated by -Ctrl suffix, retrieve the data from the Perform data structures and transform it into the PATROL format. Consumer parameters, written in PATROL Script Language (PSL), retrieve the transformed information from these parameters and present this information in a variety of formats (text, gauges, graphs, and so on) to the PATROL Console. PATROL also uses certain PSL collector parameters, designated by -Coll suffix, to gather information that the Perform Agent does not provide. This section describes
s s s

which parameters (-Ctrl and -Coll) are active during DCM data collection which parameters supply information to which application classes which parameters are not supported by the DCM data collection method

90

PATROL Knowledge Module for Unix User Guide

DCM Collection Method

Data Collection Architecture

PATROL for Unix uses two types of data collection:
s s

data collectors that use standard Unix utilities data collectors that read from the kernel

Data collection that uses standard Unix utilities is called PATROL Script Language (PSL) data collection. Data collection that reads directly from the kernel is gathered by the PATROL Perform Agent kernel-reading data collector as shown in Figure 4. The PATROL Perform Agent is not available on all platforms. See the PATROL for Unix Getting Started guide for a list of platforms that support the PATROL Perform Agent. Figure 4 Data Collection Architecture Using PATROL Perform Collector
PATROL Agent Data Path PSL KM Unix-Named Pipe (FIFO) Data Collection Manager (DCM)

PSL popen() control pipe

PATROL Perform Collector

Shared Memory

Figure 4 depicts how the PATROL Perform collector fits into the data collection architecture of PATROL. The PATROL Perform collector collects the data and writes it to shared memory. The Data Collection Manager (DCM) of the PATROL KM for Unix and Linux is the control point and data interface between the PATROL Perform collector and the PATROL Agent.

Collector Parameters for DCM

When the DCM Collection Method is selected, PATROL instantiates the following parameters under the DCM application class. Use this information to customize your PATROL installation to gather only the data you need.

Chapter 4 Managing Data Collection

91

DCM Collection Method

Table 27 Collector CPUCtrl DCMColl DFColl DISKCtrl

Collection Parameters Used By the DCM Collection Method PSL / Perform Perform PSL PSL Perform PSL Perform Perform PSL PSL PSL Perform Perform PSL Perform Active / Inactive Active Active Active Active Active Active Active Inactive Active Inactive Active Active Active Active Consumer Application CPU, SMP DCM FILESYSTEM DISK HEALTH_AT_A_GLANCE KERNEL MEMORY NETWORK NFS PRINTER PROCESS SWAP CPU USERS See Also 53 53 51 54 51 54 55 55 55 55 55 55 55 55

HAAGColl KERNELCtrl MEMORYCtrl NETColl NFSColl PRINTERColl PROCESSCtrl SWAPCtrl UPTColl USERSCtrl

-Ctrl and -Coll Parameters

The DCM data collection method gathers data using both the Perform Agent and the PATROL PSL collectors. Ultimately, consumer parameters display the information gathered by both types of collection. However, understanding how each type of collector gathers data and which type of collector feeds a consumer helps to understand apparent discrepancies in system statistics. These discrepancies may be due to poorly managed discovery cycles or other performance-tuning issues.

-Ctrl Suffix
Parameters with -Ctrl suffix are standard parameters that interact with the Perform Agent through the DCM application. These parameters control the poll time and status (active or inactive) for data collection of application parameter data.

-Coll Suffix
Parameters with the -Coll suffix are collector parameters that interact directly with Unix utilities. This type of parameter issues the system calls and receives the data without having to interact with another data-gathering application.

92

PATROL Knowledge Module for Unix User Guide

DCM Collection Method

DCM Collection Exceptions

When you use the DCM data collection method, PATROL deactivates the parameters listed below on the specified platforms. When a parameter is deactivated, PATROL does one of two things:
s

does not create the parameter does not update the parameter (if the parameter was created by using the PSL collection method).

Restarting PATROL after changing the collection method ensures that parameters that are not updated by the DCM collection method are destroyed. Table 28 CPU CPUInt CPUProcSwch KERNEL KERFileUsedPercent KERInodeUsedPercent KERProcUsedPercent KERMsg KERSemOps KERDirBlk KERNameI KERIGet MEMORY MEMActiveVirPage MEMAddrTransFault MEMBRead MEMBWrt MEMCow MEMPageAnticipated MEMPageFreed MEMRCache MEMWire MEMZero All All All All All All All All All All All except HP-UX All except Compaq Tru64 and HP-UX All except Compaq Tru64 and HP-UX All except AIX and Compaq Tru64 All except AIX and Compaq Tru64 All All All All All DCM Deactivated Parameters and their Platforms (Part 1 of 2) Platforms

Deactivated Parameters

Chapter 4 Managing Data Collection

93

DCM Collection Method

Table 28 PROCESS

DCM Deactivated Parameters and their Platforms (Part 2 of 2) Platforms All All All All All All All All All All All All All All All All All All All All

Deactivated Parameters PROCExec PROCProcWaitInt PROCProcWaitUnint SMP SMPContextSwitch SMPCrossCalls SMPInterrupts SMPIntThread SMPInvContSwitch SMPMajorFaults SMPMinorFaults SMPRunQLen1Min SMPRunQLen5Min SMPRunQLen15Min SMPSpinMutex SMPSpinRdWr SMPSystemCalls SMPThMigration SWAP SWPSwapFreeSpace SWPSwapSize SWPSwapUsedPercent

94

PATROL Knowledge Module for Unix User Guide

PSL Collection Method

PSL Collection Method

The PSL collection method uses the PATROL consumer-collector model for gathering and presenting application information. In this model, collector parameters, written in PSL, make calls to Unix utilities to collect information about the system. Consumer parameters, also written in PSL, retrieve the information from the collector parameters and present this information in a variety of formats (text, gauges, graphs, and so on) to the PATROL Console. This section describes
s s s

which collector parameters are active during PSL data collection which collector parameters supply information to which application classes which parameters are not supported by the PSL data collection method

Collector Parameters for COLLECTORS

When the PSL Collection Method is selected, PATROL instantiates the following parameters under the COLLECTORS application class. Use this information to customize your PATROL installation to gather only the data you need. Table 29 Collector CollectorStatus CollectorStatusText DFColl HAAGColl NETColl NFSColl PRINTERColl PROCColl PSColl SARColl SMPColl UPTColl USRPROCColl VMColl Collection Parameters Used By the PSL Collection Method See Also 51 51 51 51 51 51 51 51 51 51 51 52 52 52 COLLECTORS COLLECTORS FILESYSTEM HEALTH_AT_A_GLANCE NETWORK NFS PRINTER PROCESS PROCESS CPU, KERNEL, MEMORY, PROCESS, SWAP SMP CPU PROCESS, USER CPU, KERNEL, MEMORY, PROCESS

Active / Inactive Consumer Application Active Active Active Active Inactive Active Inactive Active Active Active Active Active Active Active

Chapter 4 Managing Data Collection

95

PSL Collection Method

-Coll Parameters
The PSL data collection method gathers data using only the PATROL PSL collector parameters. The parameters with -Coll suffix are collector parameters that interact directly with Unix utilities. This type of parameter issues the system calls and receives the data without having to interact with another data-gathering application.

PSL Collection Exceptions

When you use the PSL collection method, PATROL deactivates the parameters listed below on the specified platforms. When a parameter is deactivated, PATROL does one of two things:
s

does not create the parameter does not update the parameter (if the parameter was created by using the DCM collection method)

Restarting PATROL after changing the collection method ensures that parameters that are not updated by the PSL collection method are deleted. Table 30 CPU CPUInt CPUWio DISK DISKDSKAvgQueue DSKAvgServ DSKPercentBusy DSKRead DSKReadWrite DSKSps DSKWrite KERNEL KERFileUsedPercent KERInodeUsedPercent KERMsg KERProcUsedPercent Compaq Tru64 Compaq Tru64 Solaris Solaris Compaq Tru64 AIX and Compaq Tru64 Compaq Tru64 Solaris Compaq Tru64 Compaq Tru64 Solaris AIX Compaq Tru64 and Solaris Collector Deactivated Parameters and their Platforms (Part 1 of 2) Platforms

Deactivated Parameters

96

PATROL Knowledge Module for Unix User Guide

PSL Collection Method

Table 30

Collector Deactivated Parameters and their Platforms (Part 2 of 2) Platforms Solaris AIX (32-bit) AIX (32-bit) and Solaris Compaq Tru64 and Solaris Compaq Tru64 and Solaris AIX (32-bit) AIX (32-bit) AIX (32-bit) Compaq Tru64 AIX, Compaq Tru64, HP-UX, Solaris Compaq Tru64 and Solaris Compaq Tru64 and Solaris Compaq Tru64 and Solaris AIX (32-bit) and Compaq Tru64 AIX

Deactivated Parameters KERSemOps MEMORY MEMActiveVirPage MEMFreeMem MEMLRead MEMLWrt MEMPageFreed MEMPageIn MEMPageOut MEMPageScanned MEMPFault MEMPRead MEMPWrt MEMWCache PROCESS PROCProcWait PROCTopProcs

Chapter 4 Managing Data Collection

97

Managing Collection

Managing Collection
The DCM application class and its components and the COLLECTORS application class are both installed in a standard PATROL installation. By default, PATROL attempts to use the DCM data collection method associated with the DCM KM. If DCM is either not available or not supported on the managed systems platform, PATROL uses the PSL collection method associated with the COLLECTORS KM. The data collection management tasks are described in detail on the following pages: Task Changing Data Collection Methods Activating an Inactive Application Class Activating an Inactive Collector Parameter Significant Events Related to Data Collection Page 98 100 100 101

Changing Data Collection Methods

This task describes how to change the data collection method for the PATROL KM for Unix and Linux. See the PATROL for Unix Getting Started guide for more information about the data collection methods that are supported on each platform.

Before You Begin

Determine the collection method that you are currently using by opening the UNIX OS container. If the DCM icon is visible, PATROL is using the DCM data collection method. If the COLLECTORS icon is visible, PATROL is using the PSL data collection method.

To Change the Data Collection Method

To change the collection method, perform the following procedure:

1 Access the UNIX OS application KM menu commands as described in Accessing

KM Commands and InfoBoxes on page 402.

2 Select Knowledge Module Admin => Toggle PSL/DCM Collection.

PATROL displays a confirmation dialog box (Figure 5 on page 99).

98

PATROL Knowledge Module for Unix User Guide

Managing Collection

Figure 5

PSL Collection Confirmation Dialog Box

3 Click Yes to confirm.

PATROL changes the collection method. In the process of changing collectors, PATROL destroys some existing parameters and creates others.

NOTE
Neither collection method gathers data for every parameter available in the KM. For a list of parameters that are deactivated for a specific collection method, see DCM Collection Exceptions on page 93 or PSL Collection Exceptions on page 96.

4 Access the Computer application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

5 Select Update Connection.

PATROL reconnects to the Agent and displays the parameters supported by the chosen data collection method.

NOTE
Steps 4 and 5 are optional. PATROL will automatically reconnect to the Agent. However, due to various discovery intervals, impatient users can reconnect faster by using the menu command than by using the automated procedure.

Chapter 4 Managing Data Collection

99

Managing Collection

Activating an Inactive Application Class

This task describes how to turn on the data collection mechanism for an application class. It is necessary because some application classes are installed in an inactive state. To collect data for those application classes, you must activate the application class and its corresponding collector parameters.

Before You Begin

You must activate the collector parameter and its parent application class. For information on how to activate a collector parameter, see Activating an Inactive Collector Parameter on page 100.

To Start Collecting Data for an Application Class 1 Access the application class properties as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Check the Active checkbox. 3 Click OK.

PATROL closes the dialog box, runs discovery for this application, and creates the associated collector, consumer, and standard parameters.

Activating an Inactive Collector Parameter

This task describes how to turn on a collector parameter. It is necessary because some application classes are installed in an inactive state. To collect data for those application classes, you must activate the application class and its corresponding collector parameters.

Before You Begin

You must activate the application class and collector parameter. For information on how to activate an application class, Activating an Inactive Application Class on page 100.

100

PATROL Knowledge Module for Unix User Guide

Managing Collection

To Start a Collector Parameter 1 Access the application class properties as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Check the Active checkbox. 3 Click OK.

PATROL closes the dialog box, runs discovery for this parameter, and gathers the associated data.

Significant Events Related to Data Collection

PATROL for Unix uses the DCM collector as the default means to gather information for application classes and collector processes. The following events are generated by conditions that affect a change in data collection. Note that when PATROL for Unix switches to PSL data collection, Process Presence functionality does not exist. Look for these events to alert you to a change in data collection methods and the fact that the agent may not be collecting the type of data you want.

Switch to Manual PCM Data Collection

The following event is logged when you manually switch to PSL data collection. Type of Event Severity Event Class Description INFORMATION 3 UNIXToggleCollection This agent is now using classic PSL methods to collect data. This change was initiated from a Console in developer mode by @developer.

Chapter 4

Managing Data Collection

101

Managing Collection

Toggle the DCM/PSL Collection Menu Command to DCM

The following event is logged when you select the DCM/PSL toggle command to switch to DCM data collection. Type of Event Severity Event Class Description INFORMATION 3 UNIXToggleCollection This agent is now using the PERFORM collector to collect data. This change was initiated from a Console in developer mode by @developer.

DCM Binary Dies

One of the following events will be logged when the DCM binary dies. Type of Event Severity Event Class Description Type of Event Severity Event Class Description WARNING 3 41 DcmCol.psl: dcm died 1 times. ALARM 4 11 Alarm # 2 of global parameter 'DCMStatus' triggered on 'DCM.DCM'. 2 <= 2.00 <= 2.

DCM Fails over to PSL

The following event is logged when DCM fails over to PSL data collection. Type of Event Severity Event Class Description WARNING 3 41 DcmCol.psl: Switching to PSL data collection. DcmCol.psl: Event text: DcmCol.psl: Reason for switch to PSL: DCM terminated count exceeds maximum allowed.

102

PATROL Knowledge Module for Unix User Guide

Managing Collection

Deactivate SMP Application on a Single-CPU System

The following event is logged when the SMP application is deactivated on a single-CPU system. (This event only seems to occur when PSL collection is in effect.) Type of Event Severity Event Class Description INFORMATION 3 41 SMP: Single CPU system, deactivating.

Toggle PSL/DCM Collection and Reinitialize the Patrol Agent

The following event is logged after using the PSL/DCM toggle command and reinitializing the PATROL agent. Type of Event Severity Event Class Description INFORMATION 3 0 Event (0,StartStop) not found in catalog: snowcap 1 0.

Defined Process Presence Instance Does Not Exist for pconfig Variable
If the pconfig variable, /PUK/PROCPRES/attach_ps_list", exists and is set to TRUE, the following event is generated when the defined process presence instance does not exist. (This condition is based on a recovery action defined for PROCPPCountCheck parameter.) The following event is logged after using the PSL/DCM toggle and re initializing the PATROL agent. Type of Event Severity Event Class Description INFORMATION 3 41 This event contains a list of top ten processes and monitored processes. It is generated by the recovery action of the PPROCPCountCheck parameter due to an alert condition.

Chapter 4

Managing Data Collection

103

Managing Collection

104

PATROL Knowledge Module for Unix User Guide

Chapter

5
106 106 108 108 109 109 110 111 112 112 113 114 115

Monitoring CPU and Kernel

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor the CPU and kernel of a system. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy for CPU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CPU and Multiple CPUs (SMP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing CPU Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing CPU Statistics on Interrupts, Switches, and Waits . . . . . . . . . . . . . . . . . Viewing Statistics on CPU Run Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Statistics on Multiple CPUs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy for KERNEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . KERNEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Kernel Build Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Kernel Memory Allocation Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Kernel I-node Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Kernel System Usage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 5 Monitoring CPU and Kernel

105

Introduction

Introduction
The PATROL Knowledge Module for Unix and Linux monitors processors through two applications: CPU and SMP. It monitors the kernel through one: KERNEL.
s

The CPU application monitors such processor statistics as percentage of overall CPU utilization percentage of CPU idle time percentage of CPU time spent on user activities percentage of CPU time spent on system activities percentage of time spent waiting on input and output number of context switches per second number of non-VME device interrupts number of pages in and out per second number of pages scanned per second number of processes in the run queue

The SMP application provides basic CPU utilization statistics for multiple processors on a machine. The KERNEL application provides information such as Kernel Memory Allocation (KMA) statistics for small, large, and oversized memory requests i-node usage kernel usage of system resources

Object Hierarchy for CPU

The CPU application class contains parameters on requests made to the CPU, the size of the CPU run queue, the amount of time the CPU is idle, CPU time consumed by the system, CPU time consumed per user, and other statistics. If the system has multiple CPUs, PATROL creates an SMP application instance for each processor. Each instance contains parameters on processor usage. Figure 6 on page 107 illustrates the basic hierarchy but does not necessarily display all instances and parameters.

106

PATROL Knowledge Module for Unix User Guide

Object Hierarchy for CPU

Figure 6

CPU Application Object Hierarchy

OS container
UNIX OS

Application class

Parameters

Application class

Instances

Parameters

Chapter 5 Monitoring CPU and Kernel

107

CPU and Multiple CPUs (SMP)

CPU and Multiple CPUs (SMP)

The CPU application class monitors various aspects of the CPU. This application class does not provide any menu commands because it automatically detects and monitors the CPU or multiple CPUs. It does gather and store a number of vital statistics in the CPU parameters. The CPU tasks on how to access information stored in parameters are described in detail on the following pages: Task Viewing CPU Utilization Viewing CPU Statistics on Interrupts, Switches, and Waits Viewing Statistics on CPU Run Queue Viewing Statistics on Multiple CPUs Page 108 109 109 110

Viewing CPU Utilization

This task describes how to view what percentage of the CPU is being consumed across the system, is in user mode, and is idle.

To View the CPU Utilization Statistics 1 Access the CPU application class so that you can view its parameters. 2 Open the parameters that track the type of CPU utilization that you want to view:
Parameter CPUCpuUtil CPUIdleTime CPUSysTime Description displays the percentage of CPU utilization, which is calculated by subtracting CPU idle time from 100 displays the percentage of CPU time that is spent idle displays the percentage of CPU time that is spent in system mode doing system tasks, including the CPU resources consumed by calls to kernel routines displays the percentage of CPU time that is spent on user mode commands and tasks initiated by users

CPUUserTime

PATROL displays a graph that shows the amount of CPU utilization over time.

108

PATROL Knowledge Module for Unix User Guide

CPU and Multiple CPUs (SMP)

Viewing CPU Statistics on Interrupts, Switches, and Waits

This task describes how to view information about activities that retard the performance of the CPU. These activities include CPU interrupts, context switches, and waits on input and output operations.

To View Statistics on Activities That Interfere with the CPU 1 Access the CPU application class so that you can view its parameters. 2 Open the parameters that track the type of CPU information that you want to view:
Parameter CPUInt CPUWio CPUProcSwch Description displays the number of non-VME device interrupts displays the percentage of time that the CPU spends waiting for input and output operations displays the total number of CPU context switches per second.

PATROL displays a graph that shows the requested information over time.

Viewing Statistics on CPU Run Queue

This task describes how to view specific information about the CPU run queue.

To View the CPU Run Queue Statistics 1 Access the CPU application class so that you can view its parameters. 2 Open the parameters that track and display the run queue information in the form
that you want to view: Parameter CPULoad Description displays the 1-minute load average from the uptime command. Load average is the average number of processes in the kernels run queue during an interval (1 minute in this case). displays the number of processes in the run queue (RunQ)

CPURunQSize

PATROL displays a graph that shows the requested information over time.

Chapter 5 Monitoring CPU and Kernel

109

CPU and Multiple CPUs (SMP)

Viewing Statistics on Multiple CPUs

This task describes how to view basic CPU utilization information for multiple processors running on the same machine.

To View CPU Utilization Statistics for Multiple Processors 1 Access the CPU application class so that you can view its parameters and the SMP
application class.

2 Open the SMP application class so that you can view its instances (one for each
processor).

3 Open an SMP instance so that you can view its parameters. 4 Open the parameters that display the basic CPU utilization statistics that you want
to view: Parameter SMPIdlePercent SMPSystemPrcnt SMPWaitPercent SMPUserPercent Description displays the percentage of time that the CPU was idle displays the amount of processor time that is spent on system activities displays the amount of processor time that is spent waiting displays the amount of processor time that is spent on user-requested activities

PATROL displays a graph that shows the requested information over time.

NOTE
The SMP application class creates four parameters on all the platforms that it monitors. On some platforms such as HP and Solaris, SMP also creates some platform-specific parameters. For more information on these parameters, see the PATROL KM for Unix and Linux Help.

110

PATROL Knowledge Module for Unix User Guide

Object Hierarchy for KERNEL

Object Hierarchy for KERNEL

The KERNEL application class contains parameters on kernel messages, semaphore operations, and system calls as well as other information. Figure 7 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 7 KERNEL Application Object Hierarchy

OS container

Application class Parameters

Chapter 5 Monitoring CPU and Kernel

111

KERNEL

KERNEL
The KERNEL application class monitors various aspects of the kernel. This application class does not provide any menu commands because it automatically detects and monitors the kernel. It does gather and store a number of vital statistics in the KERNEL parameters. The KERNEL tasks on how to access information stored in parameters are described in detail on the following pages: Task Viewing the Kernel Build Information Viewing Kernel Memory Allocation Statistics Viewing Kernel I-node Usage Viewing Kernel System Usage Page 112 113 114 115

Viewing the Kernel Build Information

This task describes how to view the build date of the kernel running on your system.

To View the Build Date of the Kernel

Access the KERNEL application InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the KERNEL InfoBox, which provides the build date of the kernel. Figure 8 KERNEL InfoBox

112

PATROL Knowledge Module for Unix User Guide

KERNEL

Viewing Kernel Memory Allocation Statistics

This task describes how to view kernel memory allocation (KMA) information.

Before You Begin

Memory requests are divided into three categories:
s

Small memory requesta request for less than 256 bytes of memory Large memory requesta request from 512 bytes to 4 KB of memory Oversized memory requesta request for more than 4 KB of memory that is

allocated dynamically

To View the KMA Statistics 1 Access the KERNEL application class so that you can view its parameters. 2 Open the parameter that records the type of Kernel Memory Allocation
information that you want to view: Parameter KERLgAlloc Description displays the amount of memory in bytes that is allocated for large memory requests by KERNEL Memory Allocation (KMA) in the large memory request pool displays the number of large memory pool requests that failed displays the amount of memory in bytes that is available in the KERNEL Memory Allocation (KMA) large memory request pool displays the amount of memory in bytes that is allocated to the small memory request pool by KERNEL Memory Allocation (KMA) displays the number of small memory requests that failed displays the amount of memory in bytes that is available in the small memory request pool allocated by KERNEL Memory Allocation (KMA) displays the amount of memory that is dynamically allocated for oversized requests displays the number of requests for oversized memory that could not be satisfied

KERLgFail KERLgMem KERSmlAlloc

KERSmlFail KERSmlMem

KEROvzAlloc KEROvzFail

PATROL displays a graph that shows the requested information over time.

Chapter 5 Monitoring CPU and Kernel

113

KERNEL

Viewing Kernel I-node Usage

This task describes how to view information about i-node slots, entries, and tables.

Before You Begin

An i-node is normally used to store active files, directories, and file information. It is also used for other services such as pipes, clone drivers, and, on some systems, semaphores.

To View the I-node Statistics 1 Access the KERNEL application class so that you can view its parameters. 2 Open the parameters that records the i-node information that you want to view:
Parameter KERIGet KERINodeUsedPercent KERNameI Description displays the number of files that are located by an i-node entry displays the percentage of kernel i-node slots that are used displays the number of file system pathname searches, which are the conversion of filenames to i-nodes

PATROL displays a graph that shows the requested information over time.

114

PATROL Knowledge Module for Unix User Guide

KERNEL

Viewing Kernel System Usage

This task describes how to view information about the kernel slot usage (for files, locks, and processes), total number of system calls, directory block reads per second, and message operations performed per second.

To View the Kernel System Statistics 1 Access the KERNEL application class so that you can view its parameters. 2 Open the parameter that records the type of kernel system information that you
want to view: Parameter KERDirBlk KERFileUsedPercent KERLockUsedPercent KERMsg KERProcUsedPercent Description displays the number of directory block reads that are issued per second displays the percentage of the kernel file slots that are used in the file table displays the percentage of used kernel lock slots displays the number of message operations that are performed per second displays the percentage of used kernel process slots and monitors the process table utilization Each table entry represents an active process; the number of entries available depends on the number of terminal lines available and the number of processes spawned by each user. KERSemOps KERSysCall displays the number of semaphore operations performed each second displays the total number of system calls, which consists of all read, write, fork, and exec system calls, made per second

PATROL displays a graph that shows the requested information over time.

Chapter 5 Monitoring CPU and Kernel

115

KERNEL

116

PATROL Knowledge Module for Unix User Guide

Chapter

6
118 118 119 120 120 121 121 124 125 126

Using Composite Parameters

The PATROL Knowledge Module for Unix and Linux provides the ability to create an aggregate parameter. This aggregate parameter consists of two or more existing parameters. PATROL evaluates the values of these designated parameters and then determines, based on user-defined criteria, whether the cumulative effect of these parameters states warrants a warning or an alarm. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy for COMPOSITE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Working with Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . Examples of Composite Parameter Expressions. . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax Rules for Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . Setting the Status of a Composite Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Creating Composite Parameter Expressions Using the Wizard. . . . . . . . . . . . . . Creating Composite Parameter Expressions Manually . . . . . . . . . . . . . . . . . . . . . Editing Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting Composite Parameter Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 6

Using Composite Parameters

117

Introduction

Introduction
Composite parameters are parameters whose values are dependent on one or more existing PATROL parameters. You can assign PATROL alarm settings and recovery actions on the newly created parameters in the same way that you use alarm settings and recovery actions on other parameters. A composite parameter expression consists of one or more relational clauses joined by Boolean (logical) operators. A relational clause can have one of the following forms:
s s s

parameter (relational operator) constant or string constant or string (relational operator) parameter parameter (relational operator) parameter

A relational operator can be any of the standard PSL relational operators (that is, <, >, ==, <=, >=). Relational clauses can be combined using the logical operators AND (&&) and OR (||). The result of a composite parameter expression is a PATROL parameter condition, such as alarm when true, or warn when false. For example, you can build a composite parameter that triggers a PATROL alarm indicating low memory conditions only when both of the following conditions are true:
s s

CPUWio > 40 MEMPFault > 15

See Examples of Composite Parameter Expressions on page 120 for detailed examples.

Object Hierarchy for COMPOSITE

The COMPOSITE application class contains instances for each composite parameter that you have designed. Each instance contains three parameters associated with the existing composite parameter. One parameter contains the logical expression that defines the composite parameters threshold, one contains the PATROL-generated PSL code that constitutes your parameter, and one contains the results of the composite parameter over time. Figure 9 on page 119 illustrates the basic hierarchy but does not necessarily display all instances and parameters.
118 PATROL Knowledge Module for Unix User Guide

Working with Composite Parameter Expressions

Figure 9

COMPOSITE Application Object Hierarchy

OS container
UNIX OS

Application class

Instances

Parameters

Working with Composite Parameter Expressions

You can enter composite parameter expressions manually or by using the expression entry wizard. Even if you use the expression entry wizard, you may need to edit the expression manually for complex expressions. You can create, edit, and delete composite parameter expressions using COMPOSITES_COLLECTOR with the application icon shown in Figure 10. Figure 10 COMPOSITES_COLLECTOR Application Icon

After you create a composite parameter, the icon for the composite parameter appears in the COMPOSITES container (see Figure 11 on page 120) after the next scheduled update of the parameter.

Chapter 6

Using Composite Parameters

119

Working with Composite Parameter Expressions

Figure 11

COMPOSITES Parameters Container Icon

Examples of Composite Parameter Expressions

Example 1
CPU System Time for CPU 0 is greater than 91 percent
/CPU/CPU_0/CPUSysTime > 91

Example 2
CPUWio is greater than 40 AND MEMPFault is greater than 15
/CPU/CPU_0/CPUWio > 40 && /MEMORY/MEMORY/MEMPFault > 15

Example 3
DSKAvgQueue is greater than or equal to 20 OR DSKBps is greater than 85 and MEMPFault is greater than 15
/DISK/c0t1d0/DSKAvgQueue >= 20 || /DISK/c0t3d0/DSKBps > 85 && /MEMORY/MEMORY/MEMPFault > 15

Syntax Rules for Composite Parameter Expressions

The following syntax rules apply when you create a composite parameter expression: 1. A valid composite parameter name consists of alphanumeric characters. It cannot contain a space character or any special characters, such as _, /, * or # . 2. All parameters in the expression must be valid PATROL parameters. The parameter name must specify the application class, the application instance, and the parameter name.

120

PATROL Knowledge Module for Unix User Guide

Working with Composite Parameter Expressions

3. Although the expression entry wizard displays all parameters within an application class, including collector parameters, the application cannot use collector parameters in calculations. 4. Expressions cannot contain arithmetic operators. 5. String terms must be enclosed in double quotes ("). 6. You can enclose a double quote (") in a string by escaping it using a backslash, \. A string with an enclosed double quote looks like "this \"string\"" You cannot use the backslash to escape any other special characters. 7. Although the expression entry wizard does not enable you to enter parentheses in an expression, you can enter parentheses manually to control the order of the expression evaluation. If parentheses are not used, expression evaluation is strictly left to right, with the relational operators having higher precedence than the Boolean operators.

Setting the Status of a Composite Parameter

The state of the composite parameter does not affect the state of the PATROL parameters used to define the composite parameter. For example, if you can make the composite parameter inactive, the parameters used in the expression can remain active. If the parameters used in a composite parameter expression are inactive, the composite parameter is not calculated.

Creating Composite Parameter Expressions Using the Wizard

This task describes how to create composite parameters using the expression entry wizard.

Before You Begin

You must be using a PATROL Console in developer mode.

Chapter 6

Using Composite Parameters

121

Working with Composite Parameter Expressions

To Create a Composite Parameter Expression Using the Expression Entry Wizard 1 From the COMPOSITES_COLLECTOR menu, choose KM Commands => Create
Expressions.

PATROL displays the Create Expression dialog box.

NOTE
To return to the previous dialog box or redisplay the Create Expressions dialog box, you can click Cancel at any time while you are using the expression entry wizard.

2 Type a name for the new expression.

A valid composite parameter name consists of alphanumeric characters. It cannot contain a space character or any special characters, such as _, /, * or # .

3 Click Create.
PATROL displays the Edit Expression dialog box.

A If you do not want the composite parameter to be active when you finish
defining it, deselect the Active checkbox.

NOTE
The active state for the composite parameter does not affect the active state for the parameters used in the definition of the composite parameter.

B Select the Warning state or the Alarm state button to indicate the parameter state
that PATROL sets when the condition specified by the expression occurs.

C Select the True or False button to indicate whether PATROL should set the
condition when the expression is true or false.

4 To start the expression entry wizard, click Wizard.

PATROL displays the Select Expression Term Type dialog box.

5 Select the term type that you want to use:

s

If you select PATROL KM Parameter and then click Continue, PATROL displays the Select PATROL application dialog box. Go to Step 6 to continue. If you select Constant value and then click Continue, PATROL displays the Enter Constant Value dialog box. Go to step 8 on page 123 to continue.

122

PATROL Knowledge Module for Unix User Guide

Working with Composite Parameter Expressions

6 Select the application class that contains the parameter that you want to use in the
expression and then click Continue. The parameters associated with the selected application class are displayed in the parameter list box.

7 Select the application parameter that you want to use in the expression. A If there is more than one instance of the application, select the instance you want
to use in the definition and then click Continue.

B If the application has only one instance, select the PATROL application
parameter you want to use in the expression and then click Continue.

NOTE
Although collector parameters are listed in the dialog box, they cannot be used in composite parameter calculations.

PATROL displays the Select Relational Operator dialog box. Go to step 10 to continue.

8 Enter the constant you want to use in the composite parameter expression.
The constant can be either an integer value or a string expression. See Syntax Rules for Composite Parameter Expressions on page 120 for information about string expressions.

9 Click Continue to continue entering the expression.

PATROL displays the Select Relational Operator dialog box.

10 Select the relational operator you want to use in the relational clause and then click
Continue.

After you complete entry of the second term, PATROL displays the Expression Complete dialog box.
s

If you do not want to add another relational clause to the expression, click Done. PATROL displays the Save or Discard dialog box. The procedure continues with step 12 on page 124. If you want to add another relational clause to the expression, click More. PATROL displays the Select Boolean Operator dialog box. The procedure continues with step 11 on page 124.

Chapter 6

Using Composite Parameters

123

Working with Composite Parameter Expressions

11 Select the Boolean operator you want to use to join relational clauses.
PATROL displays the Select Expression Term Type dialog box. Return to step 5 on page 122 to add another relational clause to the expression.

12 Click Save to save the composite parameter expression.

After you click Save, the KM analyzes the expression to make sure that all terms are valid.
s

If the KM detects no errors, the new composite parameter appears in the COMPOSITES container after the next collection cycle. If the KM detects errors in the expression, the Syntax Errors dialog box appears, which contains text identifying the error. Click OK to display the Edit Expression dialog box and correct the error.

Creating Composite Parameter Expressions Manually

This task describes how to create composite parameter expressions manually.

Before You Begin

Before you enter a composite parameter expression manually, you should be familiar with the syntax rules on page 120. You must be using a PATROL Console in developer mode.

To Create a Composite Parameter Expression Manually 1 From the COMPOSITES_COLLECTOR menu, choose KM Commands => Create
Expressions.

PATROL displays the Create Expressions dialog box.

2 Enter the parameter expression in the text entry box. 3 Select Done when you have completed your entry.
PATROL displays the Save or Discard dialog box.

4 Click Save to save the composite parameter expression. The KM analyzes your
entry to make sure that all terms are valid.

124

PATROL Knowledge Module for Unix User Guide

Working with Composite Parameter Expressions

If the KM detects no errors, the new composite parameter appears in the COMPOSITES container after the next collection cycle. If the KM detects errors in the expression, the Syntax Errors dialog box appears, which contains text identifying the error. Click OK to display the Edit Expression dialog box and correct the error.

Editing Composite Parameter Expressions

This task describes how to edit a composite parameter expression.

Before You Begin

You must be using a PATROL Console in developer mode. Understand the syntax rules for composite parameter expressions. For more information, see Syntax Rules for Composite Parameter Expressions on page 120.

To Edit a Composite Parameter Expression 1 Choose KM Commands => Edit Expressions from the Composites menu located on
the Composites application icon. PATROL displays the Edit Expressions dialog box.

2 Select the name of the composite parameter you want to edit and click Edit. 3 Edit the expression as follows:
s

To change the relational clause, edit the expression in the text entry dialog box manually. For more information, see Syntax Rules for Composite Parameter Expressions on page 120. To add a relational clause to the expression, click Wizard. You are then prompted to select a Boolean operator to join the clauses.

4 Click Done to complete the changes.

PATROL displays the Save or Discard dialog box.

5 Click Save to save the changes.

The PATROL KM validates the expression after you click Save and displays any problems.

Chapter 6

Using Composite Parameters

125

Working with Composite Parameter Expressions

Deleting Composite Parameter Expressions

This task describes how to delete composite parameter expressions.

Before You Begin

You must be using a PATROL Console in developer mode. At least one composite parameter expression must exist.

To Delete a Composite Parameter Expression 1 Choose KM Commands => Delete Expressions from the Composites menu located on
the Composites application icon. PATROL displays the Delete Expressions dialog box.

2 Select the name of the composite parameter you want to delete and click Delete.
PATROL removes the expression from the dialog box and deletes the application instance and its associated parameters.

126

PATROL Knowledge Module for Unix User Guide

Chapter

7
128 128 130 130 131 132 133

Monitoring Disks
The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor all disk drives mounted on a system. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Disks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Excluding a Single Disk from Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selecting Disks to Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Resuming Monitoring Disks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Disk Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 7 Monitoring Disks

127

Introduction

Introduction
The DISK application monitors the disk drives that are mounted on a system for information such as
s s s s s s s s

average number of requests in the queue average service time number of blocks transferred per second percentage of time the device was busy servicing a transfer request number of kilobytes read from disk per second number of read and write pages per second number of disk transfers per second number of kilobytes written to disk per second

By default, the DISK application class discovers and monitors all the disks mounted on the system. You use menu commands to select which drives are monitored and which are not.

Object Hierarchy
The DISK application class contains an instance for each disk that is being monitored. Each disk instance contains parameters on queued requests, service time, block transfer rate, and read and write speeds. Figure 12 on page 129 illustrates the basic hierarchy but does not necessarily display all instances and parameters.

128

PATROL Knowledge Module for Unix User Guide

Object Hierarchy

Figure 12

DISK Application Object Hierarchy

OS container
UNIX OS

Application class

Instances

Parameters

Chapter 7 Monitoring Disks

129

Disks

Disks
The DISK application class enables you to select which disks PATROL monitors. With this feature, you can exclude from monitoring test disks and other disks that are not vital to your companys day-to-day activities. The DISK tasks are described in detail on the following pages: Task Excluding a Single Disk from Monitoring Selecting Disks to Monitor Resuming Monitoring Disks Page 130 131 132

Excluding a Single Disk from Monitoring

This task describes how to stop monitoring a single disk.

Before You Begin

To perform this task, you must be in developer mode. If you attempt to stop monitoring a disk when in operator mode, PATROL displays the following error message in the system output window.
DISK: Only Development Console can filter an instance

To Stop Monitoring a Disk

On the disk that you do not want to monitor, perform the following steps.

1 Access the DISK application menu as described in Accessing KM Commands and

InfoBoxes on page 402.

2 Select Stop Monitoring.

PATROL removes the disk icon from the PATROL Object Hierarchy and stops monitoring the selected disk. To resume monitoring of the disk, see Selecting Disks to Monitor on page 131.

130

PATROL Knowledge Module for Unix User Guide

Disks

Selecting Disks to Monitor

By default, PATROL monitors all disks. This task describes how to determine which disks PATROL excludes from monitoring and how to resume monitoring disks that have been excluded.

Before You Begin

s

The exclusion list feature supports regular expressions. To stop monitoring large numbers of disks that have uniform names, use regular expressions. To perform this task, you must be in developer mode. If you attempt to stop monitoring a disk when in operator mode, PATROL displays the following error message in the system output window.

DISK: Only Development Console can filter an instance

To Stop Monitoring Disks

On any disk, perform the following steps.

1 Access the DISK application menu as described in Accessing KM Commands and

InfoBoxes on page 402.

2 Select Edit DISK Exclusion List.

PATROL opens the DISK Exclusion List dialog box (Figure 13), which contains two list boxes: Monitored Disks and Excluded Disks. Figure 13 DISK Exclusion List Dialog Box

Chapter 7 Monitoring Disks

131

Disks

3 In the Monitored Disks list box, select the disk that you want to exclude from
monitoring.

4 Click Apply.
PATROL adds the disk to the list of excluded disks and then removes the disk icon from the PATROL Object Hierarchy.

5 Repeat step 1 through step 4 for each disk that you want to exclude.

Resuming Monitoring Disks

This task describes how to resume monitoring a single disk or all disks.

Before You Begin

To perform this task, you must be in developer mode. If you attempt to resume monitoring a disk when in operator mode, PATROL displays the following error message in the system output window.
DISK: Only Development Console can filter an instance

To Resume Monitoring a Disk

On any disk, perform the following steps.

1 Access the DISK application menu as described in Accessing KM Commands and

InfoBoxes on page 402.

2 Select Edit DISK Exclusion List.

PATROL opens the DISK Exclusion List dialog box (Figure 13 on page 131), which contains two list boxes: Monitored Disks and Excluded Disks.

3 In the Excluded Disks list box, select the disk that you want to resume monitoring. 4 Click Apply.
PATROL places the disk in the Monitored Disks list box, discovers the disk instance during the next discovery cycle, and resumes monitoring it.

132

PATROL Knowledge Module for Unix User Guide

Disk Space

To Resume Monitoring All Disks 1 Access the DISK application menu as described in Accessing KM Commands and
InfoBoxes on page 402.

2 Select Clear DISK Exclusion List.

PATROL opens the Confirmation dialog box, which asks whether you want to clear the DISK exclusion list.

3 Click Yes.
PATROL discovers all the disk instances during the next discovery cycle and resumes monitoring them.

Disk Space
With PATROL, you can monitor disk space by user. For more information, see Chapter 18, Monitoring Users.

Chapter 7 Monitoring Disks

133

Disk Space

134

PATROL Knowledge Module for Unix User Guide

Chapter

8
136 137 138 138 139 141 143 144 145 146 146 148 149 150 153

Monitoring and Managing File Systems

8

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor all file systems in a network. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitoring File Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Excluding a Single File System from Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . Selecting File Systems to Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Automating File System Cleanup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Running Recovery Actions Only with Operator Confirmation . . . . . . . . . . . . . . Preventing File Systems from Being Monitored . . . . . . . . . . . . . . . . . . . . . . . . . . . Working with File System Types and File System Classes . . . . . . . . . . . . . . . . . . Managing File Systems Using PATROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mounting a File System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Unmounting a File System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synchronizing File System Cache and Disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Searching for Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Removing Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 8 Monitoring and Managing File Systems

135

Introduction

Introduction
The FILESYSTEM application monitors the file systems mounted on a system for information such as
s

disk space number of kilobytes available number of kilobytes used percentage of space used

i-nodes percentage of i-nodes used number of i-nodes available

The application also lets you manage a monitored file system within PATROL. Some of the tasks that you can perform are
s s s

mount and unmount file systems synchronize information stored in the file system cache and disk search for and remove files

By default, the FILESYSTEM application discovers and monitors all the disks mounted on the system. You can use menu commands to select which file systems are monitored and which are not.

136

PATROL Knowledge Module for Unix User Guide

Object Hierarchy

Object Hierarchy
The FILESYSTEM application class contains an instance for each file system that is being monitored. Each file system instance contains parameters on free and used space, free and used i-nodes, and system capacity. Figure 14 illustrates the basic hierarchy but does not necessarily display all and parameters. Figure 14 FILESYSTEM Application Object Hierarchy

OS container
UNIX OS

Application class

Instances

Parameters

Chapter 8 Monitoring and Managing File Systems

137

Monitoring File Systems

Monitoring File Systems

The FILESYSTEM application class enables you to select which file systems PATROL monitors. With this feature, you can exclude from monitoring test systems and other file systems that are not vital to your companys day-to-day activities. The FILESYSTEM tasks are described in detail on the following pages: Task Excluding a Single File System from Monitoring Selecting File Systems to Monitor Automating File System Cleanup Running Recovery Actions Only with Operator Confirmation Preventing File Systems from Being Monitored Page 138 139 141 143 143

Excluding a Single File System from Monitoring

This task describes how to stop monitoring a single file system.

Before You Begin

To perform this task, you must be in developer mode. If you attempt to stop monitoring a file system when in operator mode, PATROL displays the following error message in the system output window.
FILESYSTEM: Only Development Console can filter an instance

To Stop Monitoring a Single File System

On the file system that you do not want to monitor, perform the following steps.

1 Access the FILESYSTEM application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Stop Monitoring.

PATROL removes the icon for the file system from the PATROL Object Hierarchy and stops monitoring the selected system. To resume monitoring of the file system, see Selecting File Systems to Monitor on page 139.

138

PATROL Knowledge Module for Unix User Guide

Monitoring File Systems

Selecting File Systems to Monitor

By default, PATROL monitors all file systems. This task describes how to determine which file systems PATROL excludes from monitoring and how to resume monitoring file systems that have been excluded.

Before You Begin

s

The exclusion list feature supports regular expressions. To stop monitoring large numbers of file systems that have uniform names, use regular expressions. The exclusion list feature excludes file systems by type. You can stop monitoring all NFS systems, CD-ROM systems, and process control (proc) subsystems. To perform this task, you must be in developer mode. If you attempt to stop monitoring a file system when in operator mode, PATROL displays the following error message in the system output window.

FILESYSTEM: Only Development Console can filter an instance

NOTE
If the FILESYSTEM application instance has more than 250 instances, the KM displays a message in the system output window (SOW) similar to the following, 20040616162441 PatrolAgent-W-EUSER: Discovery/ pre discovery proc FILESYSTEM may be in an infinite loop - executed 500029 instructions. To prevent this situation, increase the maximum PSL instruction limit from the 500000 default setting to 5000000 or higher.

To Stop Monitoring File Systems

On any file system, perform the following steps.

1 Access the FILESYSTEM application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Edit FILESYSTEM Exclusion List.

PATROL opens the FILESYSTEM Exclusion List dialog box (Figure 15 on page 140), which contains two list boxes: Monitored Filesystems and Excluded Filesystems.

Chapter 8 Monitoring and Managing File Systems

139

Monitoring File Systems

Figure 15

FILESYSTEM Exclusion List Dialog Box

3 Either in the Excluded Filesystems list box, select the file system that you want to
monitor or in the Monitored Filesystems list box, select the file system that you want to exclude from monitoring.

4 If you want to stop monitoring one of the following file system types, select the
corresponding check box. Ignore Filesystem Type NFS Description excludes all NFS file systems, which means that the PATROL Agent running on this managed system monitors only file systems that reside on the managed system; this option does not exclude SAMBA or AFS shares (types include NFS and CACHEFS) excludes mounted CD-ROMs (types include CDFS, HSFS, and ISO9660) excludes file systems that provide access to the image of each active process in the system (types include PROC, USBDEVFS, and DEVPTS) excludes file systems that you designate (possible types include VFS, SYSV, and UMSDOS)

CD-ROM proc

Custom

140

PATROL Knowledge Module for Unix User Guide

Monitoring File Systems

5 Click Apply.
Based on your selections in step 3 and step 4, PATROL adds the file systems to the respective list box and then either removes or inserts the file system icons from the PATROL Object Hierarchy.

6 Repeat step 3 through step 5 for each file system that you want to exclude. Then
click Done.

Automating File System Cleanup

This task describes how to set up the Filesystem Cleanup recovery action, which removes files once the percentage of the file system currently in use, measured by the FSCapacity parameter, exceeds the threshold established in either Alarm1 or Alarm2. This recovery action attempts to remove core dump files, and if the file system /tmp, or /tmp is mounted off this file system, the action attempts to remove from /tmp files that are 30 days or older.

Before You Begin

s

You must be root to run this command. For a given FILESYSTEM instance, the FSCapacity parameter must be active. This recovery action applies to all file system types except NFS, CD-ROM, and proc.

To Set Up Automatic File System Cleanup 1 Access the Computer application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Configure Recovery Actions.

PATROL displays the Register Recovery Actions dialog box, which lists all available recovery actions.

Chapter 8 Monitoring and Managing File Systems

141

Monitoring File Systems

3 Select the FilesystemCleanup recovery action and click Accept.

PATROL displays the Recovery Actions Instance dialog box. This recovery action applies to all FILESYSTEM instances.

4 Select the only item in the list and click Edit.

PATROL displays the Edit Recovery Action dialog box (Figure 16). Figure 16 Edit Recovery Action Dialog Box

5 Select the mode of the recovery action by clicking the corresponding radio button.
s

Run Unattendedruns the recovery action automatically, without prompting

you
s

Run Attendedprompts you before running the recovery action, and does

nothing if you do not respond within the user-defined time-out period

s

Do Not Executedoes not perform the recovery action; this mode is the default

setting

NOTE
If you select Run Attended, use the slide bar to set the amount of time PATROL waits for user input before it cancels the recovery action.

6 Click ACCEPT.
PATROL closes the Edit Recovery Actions dialog box and displays the Recovery Action Instances dialog box.
142 PATROL Knowledge Module for Unix User Guide

Monitoring File Systems

7 Click Close.
PATROL closes the Recovery Action Instances dialog box and displays the Registered Recovery Actions dialog box.

8 Click Close.
PATROL closes the Registered Recovery Actions dialog box and enables the recovery action based on the options that you chose.

Running Recovery Actions Only with Operator Confirmation

This task describes how to manage a recovery action when it is set to run only with operator confirmation. With this feature, you can turn off the operator confirmation check so that the recovery action will run without user input.

Before You Begin

You must have set up a recovery action as described in Automating File System Cleanup on page 141.

To Run Recovery Actions in Attend Mode

When the value of the FSCapacity parameter exceeds the file system percentage used amount established in thresholds, PATROL displays the Run Recovery Action dialog box (Figure 17). Figure 17 Run Recovery Action Dialog Box

Chapter 8 Monitoring and Managing File Systems

143

Monitoring File Systems

Click Acknowledge to run the recovery action. If you want to disable the recovery action or run the recovery action in the future without user input, select the appropriate option.
s

Do not ask me againswitches the recovery action to unattended mode, which runs the recovery action without any user interaction Disableprevents the recovery action from attempting to run and from

producing this dialog box If you click Cancel, PATROL does not run the recovery action or save any of the options that you selected.

Preventing File Systems from Being Monitored

The PATROL KM for Unix and Linux automatically discovers the presence of mounted file systems and also issues an alert on the FSMountStatus parameter if a discovered file system is later unmounted. This feature is not optional and cannot be turned off. In some situations, however, you may want to unmount a discovered file system and cancel any alerts that the PATROL KM for Unix and Linux would subsequently issue, or you may want to exclude a file system from being monitored in the first place. In such cases, you can use one of the following options prevent a file system from being monitored.
s

Use the Stop Monitoring command to add the name of the file system to the FILESYSTEM exclusion list. This command will prevent monitoring of the file system without regard to whether the file system is mounted. Edit the /PUK/FILESYSTEM/moniList variable, which established file system persistence, to remove the file system name from the variable. This option only applies to previously discovered file systems that are no longer available.

144

PATROL Knowledge Module for Unix User Guide

Monitoring File Systems

Working with File System Types and File System Classes

The PATROL Knowledge Module for Unix and Linux uses the FILESYSTEM.km to report file system information. In turn, the FILESYSTEM.km uses the output from the Unix mount command to determine which types of file systems are mounted. The FILESYSTEM.km discovery logic looks at the type field to determine file system type. There are four classes of file systems based on type. You can use these classes to exclude certain types of file systems from monitoring. The following table identifies these classes and the Agent configuration variables that you can use to monitor them. Files System Class Files System Type Agent Configuration Variable NFS NFS The Agent Configuration variable, /UNIX/FILESYSTEM/NFSTokens designates which file system types are considered NFS file systems. By default, this variable is populated with NFS CACHEFS NFS3 NFSV3. The Agent Configuration variable, /UNIX/FILESYSTEM/CDFSTokens designates which file system types are considered CDFS file systems. By default, this variable is populated with CDFS HSFS ISO9660 CDRFS. The Agent Configuration variable, /UNIX/FILESYSTEM/PROCTokens designates which file system types are considered PROC file systems. By default, this variable is populated with PROC USBDEVFS DEVPTS PROCFS. The Agent Configuration variable, /UNIX/FILESYSTEM/CUSTTokens designates which file system types are defined as CUST file systems. By default, this variable is blank.

CDFS

CD

PROC

PROC

CUST

CUSTOM

You can edit the appropriate Token in the agent configuration variable to add, delete, or modify file system types for any of these classes. For example, you can add AUTOFS to the variable defining NFS file system types: /UNIX/FILESYSTEM/NFSTokens. This would cause the FILESYSTEM.km to treat file systems with type=AUTOFS as an NFS file system, which, in turn, would result in inclusion or exclusion based upon the NFS file system exclusion rule.

Chapter 8 Monitoring and Managing File Systems

145

Managing File Systems Using PATROL

Managing File Systems Using PATROL

The FILESYSTEM application class enables you to manage file systems from within PATROL. With this feature, you can mount and unmount systems either automatically or manually. It also gives you the ability to synchronize file system tables and find and remove files. The FILESYSTEM tasks related to managing file systems are described in detail on the following pages: Task Mounting a File System Unmounting a File System Synchronizing File System Cache and Disk Searching for Files Removing Files Page 146 148 149 150 153

Mounting a File System

This task describes how to mount a file system that PATROL has detected as offline or that has not been discovered.

Before You Begin

s

You must be root to run this command. To perform this task, you must be in developer mode. If you attempt to mount a file system when in operator mode, PATROL displays the following error message in the system output window.

FILESYSTEM: Only Development Console can filter an instance

To Mount a File System

On any file system, perform the following steps.

146

PATROL Knowledge Module for Unix User Guide

Managing File Systems Using PATROL

1 Access the FILESYSTEM application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Maintenance => Mount => Automatic or Manual. NOTE

This menu command is only available when the filesystem is offline, when the filesystem is online, it is grayed out. The Manual option is only available on Linux, RS6000, OSF1, and HP platforms only. The Manual mount option is not available on the Solaris platform.

Automatic
PATROL uses the file system table to determine where to mount the file system. It prompts you for the password for the root account (shown in Figure 18). This option is available only if the file system has been unmounted. Figure 18 Automatic Mount Dialog Box

Manual NOTE
You can manually mount a file system on Linux, RS6000, OSF1, and HP platforms only. The manual mount option is not available on the Solaris platform.

PATROL prompts you for the device name, where it should enter the file system, and whether or not it should update the file system table (Figure 19). Figure 19 Manual Mount Dialog Box

Chapter 8 Monitoring and Managing File Systems

147

Managing File Systems Using PATROL

Enter Device Namethe location (absolute path) of the file system

(machine_name:/path)
s

Enter Mounted Onthe path to be assigned to the file system once it is mounted Update Filesystem Tabledetermine whether or not to make a record of this

mount in the file system table; a record in the file system table ensures that the file system will be automatically mounted when the managed system is rebooted

3 Supply the required information for the type of mount that you are performing
and click on Apply. PATROL mounts the file system to your network and writes the result to the system output window.

Unmounting a File System

This task describes how to unmount a file system that PATROL detects mounted.

Before You Begin

s

You must be root to run this command. To perform this task, you must be in developer mode. If you attempt to unmount a file system when in operator mode, PATROL displays the following error message in the system output window.

FILESYSTEM: Only Development Console can filter an instance

To Unmount a File System

On any file system, perform the following steps.

1 Access the FILESYSTEM application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Maintenance => Unmount => Automatic or Manual. Automatic

PATROL prompts you for the password for the root account (Figure 20 on page 149).

148

PATROL Knowledge Module for Unix User Guide

Managing File Systems Using PATROL

Figure 20

Automatic Unmount Dialog Box

Manual
PATROL prompts you for the mount location (Figure 21), which is the path assigned to a file system once it has been mounted. Figure 21 Manual Unmount Dialog Box

3 Supply the required information for the type of dismount that you are performing
and click Apply. PATROL unmounts the file system from your network and writes the result to the system output window.

Synchronizing File System Cache and Disk

This task describes how to write the information stored in the file system memory buffer (cache) to the disk immediately rather than waiting for the system to write it at a specified interval. This action synchronizes the information in the file system cache with the information stored on the disk.

To Write Information in the File System Memory to Disk

On the file system for which you want to search for files, perform the following steps.

1 Access the FILESYSTEM application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Maintenance => Sync.

PATROL writes the memory buffer to disk.

Chapter 8 Monitoring and Managing File Systems

149

Managing File Systems Using PATROL

Searching for Files

This task describes how to find files based upon different file characteristics. With this FILESYSTEM application command, you determine where the search begins in the file system and to define search criteria. It then builds the search expression. This command returns the search expression and a list of all the files that match the expression.

Before You Begin

s

You can search for files using a different user name and password from the one under which PATROL is running. You can search for files based upon naming convention owner group change date

You can restrict searches to one or multiple physical disks a location or directory within a file system

To Search for Files

On the file system for which you want to search for files, perform the following steps.

1 Access the FILESYSTEM application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Maintenance => Find Files Like.

PATROL prompts you for the user name and password of an account that can read the files that match the criteria.

3 Type the user name and password and click OK. If you want to use the account
under which the PATROL Agent is currently running, leave both fields blank. PATROL displays the Find Files Like dialog box (Figure 22 on page 151).

150

PATROL Knowledge Module for Unix User Guide

Managing File Systems Using PATROL

Figure 22

Find Files Like Dialog Box

4 Define the search criteria. The search is a logical AND operation that returns only
files that match all the criteria defined by the following fields. Field
File Name

Description the regular expression pattern for one or more files For more information, see CRegular Expressions on page 407. Leave this field blank to search for any file name.

File Owner File Group

the owner of the file(s) the group to which the owner of the file(s) belongs current date

File Updated in Last n Days the period in days between the change date and the Limit to one physical disk

restrict the operation to the disk drive on which the file system is mounted

5 Click Find.
PATROL builds a search expression, searches the file system for files that match the criteria, and writes the results to a PATROL task object (Find Files Like) in the FILESYSTEM container.

Chapter 8 Monitoring and Managing File Systems

151

Managing File Systems Using PATROL

6 Access the Find Files Like task object as described in Accessing KM Commands
and InfoBoxes on page 402 and view the results. Your results should resemble this example:
Begin: find /home/Begin: find /home/3Supvr/patrol_console_cache -name disc* -xdev -exec ls -ld {} \; -rw-r--r-1 Begin: find /home/3Supvr/patrol_console_cache -name disc* -xdev -exec ls -ld {} \; -rw-r--r-1 Begin: find /home/3Supvr/patrol_console_cache -name disc* -xdev -exec ls -ld {} \; -rw-r--r-1 Begin: find /home/3Supvr/patrol_console_cache -name disc* -xdev -exec ls -ld {} \; -rw-r--r-1 3Supvr dev 927 May 4 14:39 /home/3Supvr/patrol_console_cache/disc-1.idx -rw-r--r-1 3Supvr dev 648 May 4 14:39 /home/3Supvr/patrol_console_cache/disc-1.dat -rw-r--r-1 3Supvr dev 829 May 3 13:54 /home/3Supvr/patrol_console_cache/disc_idx-1.idx -rw-r--r-1 3Supvr dev 0 May 3 13:54 /home/3Supvr/patrol_console_cache/disc_idx-1.dat -rw-r--r-1 3Supvr dev 12 May 3 13:55 /home/3Supvr/patrol_console_cache/disc_idx-1

Output Format
The output has the following format. Table 31 defines each expression.
Begin: find /path -name <text*?> -xdev/-mount -exec <ls> -ld {} \; -rw-r--r-- # <username> <group> ### MMM DD hh:mm <path>/<file_name.ext> -------------------- >>>>> DONE <<<<< --------------------

Table 31 Output

Find Files Like Command Output Format Description find command starting point for search file name pattern search for files that reside only on the same file system as path; xdev works only on Solaris and some BSD systems list the contents of the specified directory options for the ls command: -l long format listing, -d list only the directory names and not its contents argument that substitutes current file

Search Expression Begin: find /path - name <text?*> - xdev/-mount -exec ls -ld {}

152

PATROL Knowledge Module for Unix User Guide

Managing File Systems Using PATROL

Table 31 Output

Find Files Like Command Output Format Description file permissions number of links to the file owner of the file group of owner size in bytes date-time stamp path and file name end of file list

List of Files -rw-r--r-# username dev ### MMM DD hh:mm location -- >>> DONE <<< --

Removing Files
This task describes how to delete files based upon different file characteristics. With this FILESYSTEM application command, you specify where the file is located in the file system and to define file search criteria. It then builds the search and delete expression. This command returns the search and remove expression.

Before You Begin

s

You can search for files using a different user name and password from the one under which PATROL is running. You can search for and remove files based upon naming convention owner group change date

You can restrict search and remove operations to one or multiple physical disks a location/directory within a file system

Chapter 8 Monitoring and Managing File Systems

153

Managing File Systems Using PATROL

To Remove Files
On the file system for which you want to remove files, perform the following steps.

1 Access the FILESYSTEM application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Maintenance => Remove Files Like.

PATROL prompts you for the user name and password of an account that can read and write to the files that match the criteria.

3 Type the user name and password and click OK. If you want to use the account
under which the PATROL Agent is currently running, leave both fields blank. PATROL displays the Remove Files Like dialog box (Figure 23). Figure 23 Remove Files Like Dialog Box

154

PATROL Knowledge Module for Unix User Guide

Managing File Systems Using PATROL

4 Define the search and remove criteria. The search and remove feature is a logical
AND operation that returns only files that match all the criteria defined by the following fields. Field Name
File Name

Description the regular expression pattern for one or more files For more information, see CRegular Expressions on page 407. Leave this field blank to search for any file name.

File Owner File Group

the owner of the file(s) the group to which the owner of the file(s) belongs current date

File Updated in Last n Days the period in days between the change date and the Limit to one physical disk

restrict the search to the disk drive on which the file system is mounted

5 Click Find.
PATROL builds a search and remove expression, searches the file system for files that match the criteria, removes the files, and writes the search and remove expression to a PATROL task object (Remove Files Like) in the FILESYSTEM container.

6 Access the Remove Files Like task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the expression. Your result should resemble this example:
Begin: find /oracle/bin -name Payroll_2000* -xdev -exec rm {} \; -------------------- >>>>> DONE <<<<< --------------------

Chapter 8 Monitoring and Managing File Systems

155

Managing File Systems Using PATROL

Output Format
The output has the following format. Table 32 defines each expression.
Begin: find </path> -name <text?*> -xdev -exec rm {} \; -------------------- >>>>> DONE <<<<< --------------------

Table 32 Output

Remove Files Like Command Output Format Description find command starting point for search file name pattern search for files that reside only on the same file system as path; xdev works only on Solaris and some BSD systems run the Unix OS command remove options for the ls command: -l long format listing, -d list only the directory names and not its contents argument that substitutes current file end of search and remove output

Search Expression Begin: find /path - name <text?*> - xdev/-mount -exec rm -ld {} -- >>> DONE <<< --

156

PATROL Knowledge Module for Unix User Guide

Chapter

9
158 158 159 159 160

Monitoring Overall System Health

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor the overall health of the system. It provides immediate yes-or-no style answers to questions about overall CPU, file system, and swap space usage. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HEALTH AT A GLANCE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Overall CPU, File System, and Swap Usage . . . . . . . . . . . . . . . . . . . Viewing the Overall Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 9 Monitoring Overall System Health

157

Introduction

Introduction
The HEALTH AT A GLANCE application monitors the
s s s

overall CPU usage of the system overall file system usage of the system overall virtual memory of the system

Object Hierarchy
The HEALTH AT A GLANCE application class contains parameters on system-wide CPU, file system, and virtual memory usage. Figure 24 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 24 HEALTH AT A GLANCE Application Object Hierarchy

OS container
UNIX OS

Application class

Parameters

158

PATROL Knowledge Module for Unix User Guide

HEALTH AT A GLANCE

HEALTH AT A GLANCE
The HEALTH AT A GLANCE application class enables you to monitor the overall health of the system. By reviewing the information provided by the HAAG parameters, you should be able to determine if your system is healthy or not. The HEALTH AT A GLANCE tasks are described in detail on the following pages: Task Viewing the Overall CPU, File System, and Swap Usage Viewing the Overall Health Page 159 160

Viewing the Overall CPU, File System, and Swap Usage

This task describes how to view the systems overall statistics on CPU, file system and swap space usage and determine if the system is generally healthy.

To View the Systems Overall CPU, File System, and Swap Usage 1 Access the HEALTH AT A GLANCE application class so that you can view its
parameters.

2 Open the parameter that records the type of system information you want to view.
Field Name HAAGCpuUsage HAAGFileSystemUsage HAAGSwapUsage Description displays whether or not the systems overall CPU usage is satisfactory displays whether or not the systems overall file system usage is satisfactory displays whether or not the systems overall swap space usage is satisfactory

PATROL displays a graph that shows the requested information over time.

Chapter 9 Monitoring Overall System Health

159

HEALTH AT A GLANCE

Viewing the Overall Health

This task describes how to view a textual summary of the systems general health.

To View the Systems Overall Health 1 Access the HEALTH AT A GLANCE application class so that you can view its
parameters.

2 Open the HAAGHealthSummary parameter.

PATROL displays the information in a text parameter. Your result should resemble this example:
Day Mon DD HH:MM:SS CCYY CPU Utilization Total CPU Utilization OK User Percentage OK System Percentage OK Filesystem Usage proc home/dd apps/ fs13 Swap Usage Summary Swap Percentage Used OK

Used Used Used Used

Space Space Space Space

Percentage Percentage Percentage Percentage

OK OK WARN ALARM

160

PATROL Knowledge Module for Unix User Guide

Chapter

10
162 163 163 164 166 167 167 168 170 171 171 172 172

10

Monitoring and Managing Memory

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor and manage memory on a system. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy for MEMORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shared Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Shared-Memory Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Shared-Memory Segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Semaphores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Semaphore Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Semaphore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy for SWAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Virtual Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Amount of Swap Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Swap Partition Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Percentage of Used Swap Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 10 Monitoring and Managing Memory

161

Introduction

Introduction
The PATROL Knowledge Module for Unix and Linux monitors and manages memory through two applications: MEMORY and SWAP.
s

The MEMORY application monitors such memory activity as number of active virtual pages number of pages anticipated number of pages freed per second number of pages in and out per second number of pages scanned per second

The MEMORY application also manages shared-memory segments and semaphores. The SWAP application monitors virtual memory. It provides the following information for the entire system and for each swap area. amount of disk space reserved for swap amount of swap space used percentage of swap space used

162

PATROL Knowledge Module for Unix User Guide

Object Hierarchy for MEMORY

Object Hierarchy for MEMORY

The MEMORY application class contains several parameters with information about memory usage and management. Figure 25 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 25 MEMORY Application Object Hierarchy

OS container
UNIX OS

Application class Parameters

Shared Memory
The MEMORY application class enables you to manage shared-memory segments. With this feature, you can view which segments are shared and who owns them. You can also delete shared segments. The tasks related to shared memory are described in detail on the following pages: Task Viewing Shared-Memory Segments Deleting a Shared-Memory Segment Page 164 166

Chapter 10 Monitoring and Managing Memory

163

Shared Memory

Viewing Shared-Memory Segments

This task describes how to view information about the shared-memory segments. The PATROL KM for Unix lists the type of memory, permissions, owner, and group.

To View Shared-Memory Segments 1 Access the MEMORY application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select List Shared Memory Segments.

PATROL writes the information to a PATROL task object (List Shared Memory Segments) in the MEMORY container.

3 Access the List Shared Memory Segments task object as described in Accessing
KM Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
IPC status from <running system> as of Tue May 8 HH:MM:SS CDT CCYY T ID KEY MODE OWNER GROUP Shared Memory: m 101 0x1038f476 --rw-rw-rw3supr admin m 102 0x4130f478 --rw-rw-rwNight1 shift m 7703 0x4140f47b --rw-rw-rwbsmith HR_VA m 104 0x4109f479 --rw-rw-rwljohns ACCTG m 205 0x4111f37a --rw-rw-rwvpatel IS_TX m 106 0x4122f52c --rw-rw-rwvpatel IS_TX m 207 0x4103f47e --rw-rw-rwjocruz admin m 908 0x4777f47f --rw-rw-rwodella DISTB -------------------- >>>>> DONE <<<<< --------------------

Output Format
The output has the following format. Table 33 on page 165 defines each expression.
IPC status from <running system> as of Tue May 8 HH:MM:SS CDT CCYY T ID KEY MODE OWNER GROUP Shared Memory: m 101 0x1038f476 --rw-rw-rwuname gname -------------------- >>>>> DONE <<<<< --------------------

164

PATROL Knowledge Module for Unix User Guide

Shared Memory

Table 33 Field T

List of Shared-Segment Command Output Format Description type of facility qmessage queue mshared-memory segment ssemaphore

ID KEY MODE

the facility entry identifier the key that is used as an argument by various functions [msgget( ), semget( ), and shmget( )] to create the facility entry the facility accesses modes and flags It consists of 11 characters. The first 2 characters are Ra process is waiting on msgrc(2) Sa process is waiting on a msgsnd(2) Dthe shared-memory segment is scheduled to be removed (it will disappear when the last process attached to it detaches) Cshared-memory segment is cleared when the first attach is executed - flag not set The next 9 characters are Unix permissions: owner (first 3), group (second 3), other (third 3). Permissions are rread wwrite aalter - no permission granted

OWNER

the login name of the owner of the facility entry This name identifies the owner of the process that allocated the shared-memory segment.

GROUP

the group name of the owner

Chapter 10 Monitoring and Managing Memory

165

Shared Memory

Deleting a Shared-Memory Segment

This task describes how to delete a specified shared-memory segment based on its numeric ID, which can be discovered through the task, Viewing Shared-Memory Segments on page 164.

Before You Begin

To delete a shared-memory segment, you must own the segment or have an account with privileges to delete it. If you attempt to remove a segment that you do not own, PATROL issues a permission-denied error message.

To Delete Shared-memory Segments 1 Access the MEMORY application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Remove Shared Memory Segments.

PATROL displays the Remove Shared Memory Segment dialog box (Figure 26). Figure 26 Remove Shared Memory Segment Dialog Box

3 Type the numeric ID in the Memory Segment ID field and click either Apply or
Apply To Selected.

PATROL prompts you for a user name and password (Figure 27). Figure 27 Remove Shared Memory Segment Confirmation Dialog Box

166

PATROL Knowledge Module for Unix User Guide

Semaphores

4 Enter the user name and password of the owner of the shared segment or an
account with privileges to delete it and click OK. PATROL writes a confirmation message to a task window.
Remove shared memory segments --------------- >>>>> DONE <<<<< ---------------

Semaphores
The MEMORY application class enables you to manage semaphores in the same way you manage shared-memory segments. With this feature, you can view which semaphores have been created and who owns them. You can also delete semaphores. The semaphore management tasks are described in detail on the following pages: Task Viewing Semaphore Information Deleting a Semaphore Page 167 168

Viewing Semaphore Information

This task describes how to view information about the semaphores. The task lists type of memory, permissions, owner, and group.

To View Semaphores 1 Access the MEMORY application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select List Semaphores.

PATROL writes the information to a PATROL task object (List Semaphores) in the MEMORY container.

Chapter 10 Monitoring and Managing Memory

167

Semaphores

3 Access the List Semaphores task object as described in Accessing KM Commands

and InfoBoxes on page 402 and view the results. Your results should resemble this example:
IPC status from <running system> as of DAY MON DD HH:MM:SS CDT CCYY T KEY MODE OWNER GROUP Semaphores: s 0x1038f476 --ra-ra-ra3supr admin s 0x4130f478 --ra-ra-raNight1 shift s 0x4140f47b --ra-ra-rabsmith HR_VA s 0x4109f479 --ra-ra-raljohns ACCTG s 0x4111f37a --ra-ra-ravpatel IS_TX s 0x4122f52c --ra-ra-ravpatel IS_TX s 0x4103f47e --ra-ra-rajocruz admin s 0x4777f47f --ra-ra-raodella DISTB -------------------- >>>>> DONE <<<<< --------------------

Output Format
The output has the following format. For a description of the content, see Table 33 on page 165.
IPC status from <running system> as of DAY MON DD HH:MM:SS CDT CCYY T KEY MODE OWNER GROUP Semaphores: s 0x1038f476 --ra-ra-rauname gname -------------------- >>>>> DONE <<<<< --------------------

Deleting a Semaphore
This task describes how to delete a specified semaphore based on its numeric ID, which can be discovered through the task, Viewing Semaphore Information on page 167.

Before You Begin

To delete a semaphore, you must own the segment or have an account with privileges to delete it. If you attempt to remove a segment that you do not own, PATROL issues a permission-denied error message.

168

PATROL Knowledge Module for Unix User Guide

Semaphores

To Delete Semaphores 1 Access the MEMORY application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Remove Semaphores.

PATROL displays the Remove Semaphores dialog box (Figure 28). Figure 28 Remove Semaphores Dialog Box

3 Type the numeric ID in the Semaphore ID field and click either Apply or Apply To
Selected.

PATROL prompts you for a user name and password (Figure 29). Figure 29 Remove Semaphores Confirmation Dialog Box

4 Enter the user name and password of the owner of the semaphore or an account
with privileges to delete it and click OK. PATROL writes a confirmation message to a task window.
Remove semaphores --------------- >>>>> DONE <<<<< ---------------

Chapter 10 Monitoring and Managing Memory

169

Object Hierarchy for SWAP

Object Hierarchy for SWAP

The SWAP application class contains several parameters with information about partition size, free swap space available for the entire system, and the percentage of swap space used. Figure 30 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 30 SWAP Application Object Hierarchy

170

PATROL Knowledge Module for Unix User Guide

Virtual Memory

Virtual Memory
The SWAP application class enables you to monitor virtual memory. With this feature, you can view how much disk space is available for swapping and how much is currently used for swapping. The SWAP tasks are described in detail on the following pages: Task Viewing the Amount of Swap Space Viewing the Swap Partition Size Viewing the Percentage of Used Swap Space Page 171 172 172

Viewing the Amount of Swap Space

This task describes how to discover how much swap space is available for either a swap area or an entire system.

To View the Amount of Swap Space in a Swap Area 1 Access the swap area instance in the SWAP application class so that you can view
its parameters.

2 Open the SWPSwapFreeSpace parameter.

PATROL displays a graph that shows the amount of available swap space in the designated swap area over time.

To View the Amount of Total Swap Space on the System 1 Access the Summary instance in the SWAP application class so that you can view
its parameters.

2 Open the SWPToTSwapFreeSpace parameter.

PATROL displays a graph that shows the total amount of available swap space on the system over time.

Chapter 10 Monitoring and Managing Memory

171

Virtual Memory

Viewing the Swap Partition Size

This task describes how to discover the current size of the swap partition for a swap area.

To View the Swap Partition Size in a Swap Area 1 Access the swap area instance in the SWAP application class so that you can view
its parameters.

2 Open the SWPSwapSize parameter.

PATROL displays a graph that shows the amount of available swap space in the designated swap area over time.

Viewing the Percentage of Used Swap Space

This task describes how to discover the current amount of swap space being used as a percentage of the total amount available for either the entire system or a particular swap area.

To View the Percentage of Swap Space Used in a Swap Area 1 Access the swap area instance in the SWAP application class so that you can view
its parameters.

2 Open the SWPSwapUsedPercen parameter.

PATROL displays a gauge that shows the amount of swap space currently being used as a percentage of the total swap space available in the designated swap area.

To View the Percentage of the Total Swap Space Used on the System 1 Access the Summary instance in the SWAP application class so that you can view
its parameters.

2 Open the SWPToTSwapUsedPercen parameter.

PATROL displays a gauge that shows the total amount of available swap space currently being used as a percentage of the total swap space available on the system.

172

PATROL Knowledge Module for Unix User Guide

Chapter

11

11

Monitoring and Managing Networks

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor the Unix network. The following sections appear in this chapter.

NOTE
At installation, the NETWORK application class is set to inactive. To enable it to collect data for its parameters, see Activating the NETWORK Application Class on page 176. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network Activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Activating the NETWORK Application Class. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deactivating the NETWORK Application Class. . . . . . . . . . . . . . . . . . . . . . . . . . . Network Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing a Network Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discovering the Name of the Network Interface . . . . . . . . . . . . . . . . . . . . . . . . . . Establishing Behavior When Unavailable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Memory Buffer Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network Addresses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Listing Host and IP Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Listing Protocols Running on the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Protocol Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determining the Network Internet Protocol Address . . . . . . . . . . . . . . . . . . . . . . Network Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Routing Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determining the Number of Packets Entering and Exiting the Network . . . . . . Determining the Number of Packet Errors Entering and Exiting the Network. Network Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Listing TCP/IP Services Running on the Network . . . . . . . . . . . . . . . . . . . . . . . . Listing Trusted Hosts on the Network. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Copying or Transferring Files by Using FTP. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Checking for a Network Resource by Using ping . . . . . . . . . . . . . . . . . . . . . . . . . 174 175 176 176 177 177 178 179 180 181 184 184 187 189 189 191 191 192 192 194 194 196 196 198 198 199

Chapter 11

Monitoring and Managing Networks

173

Introduction

Logging In Remotely by Using rlogin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Querying the Name Server for the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Address Resolution Protocol (ARP). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Viewing the Address Resolution Protocol (ARP) Cache . . . . . . . . . . . . . . . . . . . . 202 Deleting a Host from the ARP Cache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Adding a Host to the ARP Cache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Unix to Unix Communication Protocol (UUCP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Checking for UUCP Support Files and Directories. . . . . . . . . . . . . . . . . . . . . . . . . 206 Cleaning Up UUCP Files in the Spool Directory. . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Copying Files Using UUCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 Listing All UUCP Servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Viewing the UUCP Log File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Viewing UUCP Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211

Introduction
The NETWORK application monitors the network for information on
s

interfaces names of all or individual networks memory buffer addresses local and remote addresses host name and corresponding IP Address protocols available protocols protocol statistics traffic routing tables number of incoming and outgoing packets percentage of incoming and outgoing packets with errors

This application also lets you manage the network by providing you with some standard network utilities, which are accessible from within PATROL. Such management tasks include
s s s s s

copying or transferring files using either ftp or UUCP logging into remote machines using rlogin checking the availability of network resources using ping querying the network name server adding and deleting hosts from the Address Resolution Protocol cache

174

PATROL Knowledge Module for Unix User Guide

Object Hierarchy

Object Hierarchy
The NETWORK application class contains the instances for networks that are being monitored. Each network instance contains parameters that gather information on packet traffic in and out of the network, on collisions, and on the percentage of errors in and out of the network. Figure 31 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 31 NETWORK Application Object Hierarchy

OS container
UNIX OS

Application class

Instances

Parameters

Chapter 11

Monitoring and Managing Networks

175

Network Activation

Network Activation
The NETWORK application and its corresponding collector, NETColl, are installed in an inactive state. They do not automatically gather information when loaded. The NETWORK tasks related to activating and deactivating the NETWORK application class and NETColl collector parameter are described in detail on the following pages: Task Activating the NETWORK Application Class Deactivating the NETWORK Application Class Page 176 177

WARNING
Use the same method to activate and deactivate application classes and their respective collectors. For example, if you use another method to deactivate the NETWORK application class (such as customizing it with a PATROL Console in developer mode), attempting to activate it with the menu command, Knowledge Module Admin => NETWORK Application => Activate Monitoring, will fail.

Activating the NETWORK Application Class

This task describes how activate the NETWORK application class and the NETColl collector parameter. This action causes PATROL to begin gathering information about the network and enables its network monitoring and managing features.

To Activate the NETWORK Application Class 1 Access the UNIX OS application KM menu commands as described in Accessing
KM Commands and InfoBoxes on page 402.

2 Select Knowledge Module Admin => NETWORK Application => Activate Monitoring.
PATROL begins collecting data for the NETWORK application class.

176

PATROL Knowledge Module for Unix User Guide

Network Interface

Deactivating the NETWORK Application Class

This task describes how deactivate the NETWORK application class and the NETColl collector parameter. This action causes PATROL to quit gathering information about the network and disables its network monitoring and managing features.

To Deactivate the NETWORK Application Class 1 Access the UNIX OS application KM menu commands as described in Accessing
KM Commands and InfoBoxes on page 402.

2 Select Knowledge Module Admin => NETWORK Application => Deactivate

Monitoring.

PATROL quits collecting data for the NETWORK application class.

Network Interface
The NETWORK application class enables you to view a list of network interfaces available to you, discover the name of individual network interfaces, and get a report on the networks memory buffer. You also can define the behavior of PATROL when a network interface is unavailable. The NETWORK tasks related to the network interface are described in detail on the following pages: Task Viewing a Network Interface Discovering the Name of the Network Interface Establishing Behavior When Unavailable Viewing Memory Buffer Information Page 178 179 180 181

Chapter 11

Monitoring and Managing Networks

177

Network Interface

Viewing a Network Interface

This task describes how to view the state of an automatically configured interface in a task window.

To View a Network Interface 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select List Network Interface.

PATROL writes the results to a PATROL task object, List Network Interface, in the NETWORK container.

3 Access the List Network Interface task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ########## <NOT INTERACTIVE> List network interface Name Mtu Net/Dest lo0 8232 loop hme0 1500 xyz.com -------------------Address localhost xyz.com >>>>> DONE Ipkts Ierrs Opkts Oerrs 73190 0 73190 0 386019 5200 14659 0 <<<<< -------------------Collis 0 1810 Queue 0 0

178

PATROL Knowledge Module for Unix User Guide

Network Interface

Output Format
The output has the following format. Table 34 defines each field.
Name Mtu Net/Dest Address Ipkts Ierrs Opkts Oerrs Collis Queue

-------------------- >>>>> DONE <<<<< --------------------

Table 34 Field Name Mtu Net/Dest

List Network Interface Command Output Format Description name of the interface maximum transmission unit also known as the maximum packet size network to which the interface is attached For Point-to-Point interfaces, this is the address on the other side of the link. address for each interface number of incoming packets number of incoming packets with errors number of outgoing packets number of outgoing packets with errors number of collisions length of the queue

Address Ipkts Ierrs Opkts Oerrs Collis Queue

Discovering the Name of the Network Interface

This task describes how to determine the name of the network interface.

To Discover the Name of the Network Interface

Access the NETWORK application instance InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the NETWORK InfoBox (Figure 32 on page 180), which provides the name of the network interface in the Network Interface field.

Chapter 11

Monitoring and Managing Networks

179

Network Interface

Figure 32

NETWORK InfoBox

Establishing Behavior When Unavailable

This task describes how to define the behavior of PATROL when the network goes down. You can instruct PATROL to place the NETWORK instance into a state of alarm or to destroy the icon of the NETWORK instance when the network interface goes down.

180

PATROL Knowledge Module for Unix User Guide

Network Interface

To Establish Network Interface Behavior 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Toggle Alarm/Destroy if Down.

PATROL writes the results to the PATROL standard output window. Depending upon the NETWORK instances current setting, PATROL writes one of the following messages:

Set Alarm
Set Alarm if the network is down

Destroy the Instance and Remove the Icon

Destroy the network instance if the network is down

NOTE
To reset the NETWORK to its previous behavior, repeat this procedure.

Viewing Memory Buffer Information

This task describes how to display memory management statistics in a task window. This information includes the current, maximum, and total amounts for streams, queues, mblk, dblk, linkblk, strevent, synq, and qband. It also provides the number of allocation failures for each category.

To View Memory Buffer Information 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select List Memory Buffer.

PATROL writes the results to a PATROL task object (List Memory Buffer) in the NETWORK container.

Chapter 11

Monitoring and Managing Networks

181

Network Interface

3 Access the List Memory Buffer task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example: # PSL Serial No.: ########## <NOT INTERACTIVE> List network memory buffer streams allocation: streams queues mblk dblk linkblk strevent syncq qband current 291 824 1725 1722 12 10 25 2 maximum 320 858 3683 3760 169 169 50 127 cumulative total 333447 869663 448272 74204701 55 554502 745 2 allocation failures 0 0 0 0 0 0 0 0

1439 Kbytes allocated for streams data -------------------- >>>>> DONE <<<<< --------------------

182

PATROL Knowledge Module for Unix User Guide

Network Interface

Output Format
The output has the following format. Table 35 defines each field. streams 291 320 333447 0 queues 824 858 869663 0 mblk 1725 3683 448272 0 dblk 1722 3760 74204701 0 linkblk 12 169 55 0 strevent 10 169 554502 0 syncq 25 50 745 0 qband 2 127 2 0 -------------------- >>>>> DONE <<<<< -------------------Table 35 Field streams queues mblk dblk linkblk strevent syncq qband List Memory Buffer Command Output Format Description streams queues allocated streams message block headers data blocks link blocks structures stream events number of messages number of streams queue flow control information structures

Chapter 11

Monitoring and Managing Networks

183

Network Addresses

Network Addresses
The NETWORK application class enables you to view both the local and remote addresses for networked resources. It also provides a list of host names for each resource and its corresponding IP address. The NETWORK tasks related to network addresses are described in detail on the following pages: Task Viewing Network Addresses Listing Host and IP Address Page 184 187

Viewing Network Addresses

This task describes how to view all the network addresses. Along with the local address, it provides the protocol version in use, the remote address, the sizes of the send window and send queue, the sizes of the receive window and receive queue, and the internal state of the address.

To View Network Addresses 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select List Network Addresses.

PATROL writes the results to a PATROL task object, List Network Addresses, in the NETWORK container.

184

PATROL Knowledge Module for Unix User Guide

Network Addresses

3 Access the List Network Addresses task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ########## <NOT INTERACTIVE> List network address TCP: IPv4 Local Address -------------------172.19.203.130.32785 127.0.0.1.32788 127.0.0.1.32783 127.0.0.1.32791

Remote Address -------------------172.19.203.130.32774 127.0.0.1.32783 127.0.0.1.32788 127.0.0.1.32790

Swind Send-Q Rwind Recv-Q ----- ------ ----- -----32768 0 32768 0 32768 0 32768 0 32768 0 32768 0 32768 0 32768 0

State ------CLOSE_WAIT ESTABLISHED ESTABLISHED ESTABLISHED

Active UNIX domain sockets Address Type Vnode Conn Local Addr Remote Addr 30000b8fa08 stream-ord 30000a3d6e8 00000000 /tmp/.X11-unix/X0 30000b8fba8 stream-ord 00000000 00000000 30000b8fd48 stream-ord 30000b5a7f8 00000000 /tmp/jd_sockV6

Output Format
The output has the following format. Table 36 defines each field.
TCP: IPv4 Local Address Remote Address ... Active UNIX domain sockets Address Type Vnode Conn

Swind Send-Q Rwind Recv-Q

State

Local Addr

Remote Addr

Table 36 Field TCP

List Network Addresses Command Output Format (Part 1 of 3) Description the version of TCP/IP local address expressed as a number remote address expressed as a number size of the send window in bytes the count of bytes not acknowledged by the remote host size of the receive window in bytes the count of bytes not copied by the user program connected to this socket
Chapter 11 Monitoring and Managing Networks 185

Local Address Remote Address Swind Send-Q Rwind Recv-Q

Network Addresses

Table 36 Field State

List Network Addresses Command Output Format (Part 2 of 3) Description internal state of the socket ESTABLISHEDestablished connection SYN_SENTactively attempting to establish a connection SYN_RECVbeing initialized. FIN_WAIT1closed, and the connection is shutting down FIN_WAIT2closed, and the socket is waiting for a shutdown from the remote end TIME_WAITwaiting after close for remote shutdown retransmission CLOSEDnot being used CLOSE_WAITthe remote end has shut down, waiting for the socket to close LAST_ACKthe remote end shut down, the socket is closed, and waiting for acknowledgement LISTENlistening for incoming connections CLOSINGboth sockets are shut down but all the data sent has not been sent UNKNOWNstate is unknown

Active Unix Domain Sockets Address address of the socket

186

PATROL Knowledge Module for Unix User Guide

Network Addresses

Table 36 Field Type

List Network Addresses Command Output Format (Part 3 of 3) Description types of socket (values vary based on platform for instance stream may be represented as SOCKET_STREAM on Linux and Stream_ord on Solaris)
s

s s s s s s

dgramthe socket is used in Datagram (connectionless) mode streama stream (connection) socket rawa raw socket rdmserves reliably-delivered messages seqpacketa sequential packet socket packeta RAW interface access socket unknownunable to identify socket type

Vnode Conn Local Addr Remote Addr

virtual node connection type the local hostname and port number of the socket remote address

Listing Host and IP Address

This task describes how to view the name of a host and its corresponding IP address in the network.

To List Host and IP Addresses 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Reports => List Hosts.

PATROL writes the results to a PATROL task object (List Hosts) in the NETWORK container.

Chapter 11

Monitoring and Managing Networks

187

Network Addresses

3 Access the List Hosts task object as described in Accessing KM Commands and
InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### dyn2280234 1D IN dyn2280235 1D IN dyn2280236 1D IN dyn2280237 1D IN dyn2280238 1D IN --------------- >>>>> DONE <<<<<

A A A A A ---------------

172.19.89.227 172.19.89.228 172.19.89.229 172.19.89.230 172.19.89.231

Output Format
The output has the following format. Table 37 defines each field.
hostname 1D IN A -------------------- >>>>> DONE <<<<< -------------------IP Address

Table 37 Field hostname 1D IN A

List Hosts Command Output Format Description the hosts alias information from the NIS map ip address

IP Address

188

PATROL Knowledge Module for Unix User Guide

Network Protocol

Network Protocol
The NETWORK application class enables you to discover which protocols are currently running on the network and view protocol-specific statistics, such as the number of incoming datagrams for UDP, the number of failed attempts for TCP, and incoming deliveries for IPV4. It also provides you the ability to view the IP Address of an individual host. The NETWORK tasks related to network protocols are described in detail on the following pages: Task Listing Protocols Running on the Network Viewing Protocol Statistics Determining the Network Internet Protocol Address Page 189 191 191

Listing Protocols Running on the Network

This task describes how to list which network protocols are running on the network.

To List Network Protocols 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Reports => List Protocols.

PATROL writes the results to a PATROL task object (List Protocols) in the NETWORK container.

Chapter 11

Monitoring and Managing Networks

189

Network Protocol

3 Access the List Protocols task object as described in Accessing KM Commands

and InfoBoxes on page 402 and view the results. Your results should resemble this example: # PSL Serial No.: ########## <NOT INTERACTIVE> List TCP/IP protocols NIS is detected rdp 27 xns-idp 22 hmp 20 udp 17 pup 12 egp 8 tcp 6 ggp 3 igmp 2 icmp 1 ip 0

RDP XNS-IDP HMP UDP PUP EGP TCP GGP IGMP ICMP IP

# # # # # # # # # # #

"reliable datagram" protocol Xerox NS IDP host monitoring protocol user datagram protocol PARC universal packet protocol exterior gateway protocol transmission control protocol gateway-gateway protocol internet group management protocol internet control message protocol internet protocol, pseudo protocol number

-------------------- >>>>> DONE <<<<< --------------------

Output Format
The output has the following format. Table 38 defines each field.
pt port PT comment -------------------- >>>>> DONE <<<<< --------------------

Table 38 Field pt port PT comment

List Protocols Command Output Format Description protocol the port number on which the protocol is running protocol description of the protocol

190

PATROL Knowledge Module for Unix User Guide

Network Protocol

Viewing Protocol Statistics

This task displays the statistics for each protocol layer.

To View Protocol Statistics 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Show Protocol Statistics.

PATROL writes the results to a PATROL task object (Show Protocol Statistics) in the NETWORK container.

3 Access the Show Protocol Statistics task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should list each protocol (RAWIP, UDP, TCP, IPV4, and so on) and provide per-protocol statistics.

Determining the Network Internet Protocol Address

This task describes how to determine the Network IP address.

To Determine the Network IP Address

Access the NETWORK application instance InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the NETWORK InfoBox (Figure 32 on page 180), which provides the name to the network interface protocol address in the IP Address field.

Chapter 11

Monitoring and Managing Networks

191

Network Traffic

Network Traffic
The NETWORK application class enables you to view the network routing table. It also provides statistics on the volume of packet traffic going into and coming out of the network and gives the percentage of those packets that contain errors. The NETWORK tasks related to network traffic are described in detail on the following pages: Task Viewing the Routing Table Determining the Number of Packet Errors Entering and Exiting the Network Page 192 194

Determining the Number of Packets Entering and Exiting the Network 194

Viewing the Routing Table

This task displays the routing table and statistics for the network. The information provided by the table includes a list of destination networks, the gateway for each, the router status and route designation, the number of active uses per route, the number of packets sent per route, and the network interface used by the route.

To View Routing Table Information 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select List Routing Table.

PATROL writes the results to a PATROL task object (List Routing Table) in the NETWORK container.

192

PATROL Knowledge Module for Unix User Guide

Network Traffic

3 Access the List Routing Table task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example: # PSL Serial No.: ########## <NOT INTERACTIVE> List network routing table Routing Table: IPv4 Destination bmt_r_and_s BASE-ADDRESS.M CAST.NET default localhost Gateway spyro spyro gate19 localhost Flags U U UG UH Ref 1 1 1 27 Use 1386 0 1523 125281 Interface hme0 hme0

lo0

-------------------- >>>>> DONE <<<<< --------------------

Output Format
The output has the following format. Table 39 defines each field.
Routing Table: IPv4 Destination Gateway Flags Ref Use -------------------------------------- ---- ----###.###.###.### xyz.com UGH 11 11 -------------------- >>>>> DONE <<<<< --------------------

Interface --------hme0

Table 39 Field

List Routing Table Command Output Format Description the destination network the router through which packets are forwarded Uthe router is up Gthe route is to a gateway Hthe destination is a host rather than a network number of active uses per route number of packets sent per route the network interface that the route uses

Destination Gateway Flags

Ref Use Interface

Chapter 11

Monitoring and Managing Networks

193

Network Traffic

Determining the Number of Packets Entering and Exiting the Network

This task describes how to view the total number of packets going into and coming out of the network. These statistics reflect all packet traffic since the system was last rebooted.

To View the Number of Packets at the Moment

Access the NETWORK application instance InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the NETWORK InfoBox (Figure 32 on page 180), which provides the statistics of incoming and outgoing packets in the Total Packets In and Total Packets Out fields.

To View the Number of Packets Over Time 1 Access a NETWORK application instance so that you can view its parameters. 2 Open the parameter that records the type of packet information that you want to
view. Parameter NETPacketsIn NETPacketsOut Description displays the total number of incoming packets during the sample interval displays the total number of outgoing packets during the sample interval

PATROL displays a graph that shows the requested information over time.

Determining the Number of Packet Errors Entering and Exiting the Network
This task describes how to view the percentage of packets that contain errors. The information is separated into packets coming into and going out of the network. These statistics reflect all packet traffic errors since the system was last rebooted.

194

PATROL Knowledge Module for Unix User Guide

Network Traffic

To View the Number of Packet Errors at the Moment

Access the NETWORK application instance InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the NETWORK InfoBox (Figure 32 on page 180), which provides the statistics of incoming and outgoing packets in the Total Outgoing Errors and Total Incoming Errors fields.

To View the Number of Packet Errors Over Time 1 Access a NETWORK application instance so that you can view its parameters. 2 Open the parameter that records the type of packet information that you want to
view. Parameter NETInErrPrc NETOutErrPrc Description displays the percentage of incoming data packets that contain packet format errors displays the percentage of outgoing data packets that contain packet format errors

PATROL displays a graph that shows the requested information over time.

Chapter 11

Monitoring and Managing Networks

195

Network Management

Network Management
The NETWORK application class enables you to review some TCP/IP services running on the network, determine which hosts are trusted, and perform some standard network management tasks from within PATROL. These tasks include copying files to remote hosts, checking the availability of resources using ping, logging in remotely, and querying the networks name server. The NETWORK tasks related to network management are described in detail on the following pages: Task Listing TCP/IP Services Running on the Network Listing Trusted Hosts on the Network Copying or Transferring Files by Using FTP Checking for a Network Resource by Using ping Logging In Remotely by Using rlogin Querying the Name Server for the Network Page 196 198 198 199 200 201

Listing TCP/IP Services Running on the Network

This task describes how to create a list of all TCP/IP Services currently running on the network. It provides the name of the service, the port and protocol that the service uses, and any comments that were added when the service was registered.

To List Services Running on the Network 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Reports => List Services.

PATROL writes the results to a PATROL task object (List Services) in the NETWORK container.

196

PATROL Knowledge Module for Unix User Guide

Network Management

3 Access the List Services task object as described in Accessing KM Commands and
InfoBoxes on page 402 and view the results. Your results should resemble this example: # PSL Serial No.: ########## <NOT INTERACTIVE> List TCP/IP services NIS is detected example nbiff qip_qdhcp DAServer pmdmgr new-rwho nntp link ftp uucp name 2010/tcp 1950/udp 1097/tcp 987/tcp 696/tcp 550/udp 119/tcp 87/tcp 21/tcp 540/tcp 42/tcp # test port, Frank Ginac #QIP DHCP Lease Update daemon # SQL distributed access # OpenView Postmaster Manager readnews untp ttylink uucpd nameserver # Network News Transfer Protocol # File Transfer Protocol (Control) # uucp daemon

-------------------- >>>>> DONE <<<<< --------------------

Output Format
The output has the following format. Table 40 defines each field.
service port/prtcl # comments -------------------- >>>>> DONE <<<<< --------------------

Table 40 Field service port/prtcl comment

List Services Command Output Format Description the name of the service the assigned port number and communication protocol, which is either UDP or TCP a description of the service or a reference to the individual or group that registered the service

Chapter 11

Monitoring and Managing Networks

197

Network Management

Listing Trusted Hosts on the Network

This task describes how to create a list of all trusted hosts currently running on the network. A trusted host is defined as a computer system that cannot be accessed illegally. These criteria are established by the National Computer Security Center (NCSC).

To List Services Running on the Network 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Reports => List Trusted Hosts.

PATROL writes the results to a PATROL task object (List Trusted Hosts) in the NETWORK container.

3 Access the List Trusted Hosts task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results.

Copying or Transferring Files by Using FTP

This task describes how to use the File Transfer Protocol utility to copy files from one system to another within PATROL.

Before You Begin

PATROL supports the following FTP commands, which are grouped by type. FTP Command open, close quit, bye help bin, ASCII get, mget put, mput ls, dir cd, lcd mkdir, rmdir pwd Description connect to and disconnect from a remote computer log off list all the supported ftp functions binary and ASCII transfer formats retrieve one or multiple files from the remote system place one or multiple files on the remote system list file names only or with file property information change directories within the network or only locally make or remove directory print working directory

198

PATROL Knowledge Module for Unix User Guide

Network Management

To Transfer Files to a Remote Computer 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Utilities => Ftp.

PATROL opens the PATROL Ftp dialog box, which provides a command prompt.

3 Type open hostname to connect to a remote host.

The ftp utility prompts you for a username and password.

4 Set the data transfer format by typing either ASCII or bin. 5 Navigate to the desired directory (cd, pwd, ls) and transfer files to (put, mput) or
from (get, mget) the remote computer. The ftp utility provides feedback on the success of each command that you enter.

6 When you are done, type close to disconnect, and then type bye or quit to log off.
PATROL closes the PATROL Ftp dialog box.

Checking for a Network Resource by Using ping

This task describes how to use the ping utility to check if a network resource is available.

To Ping a Remote Computer on the Network 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Utilities => Ping.

PATROL displays the Ping dialog box (Figure 33). Figure 33 Ping Dialog Box

Chapter 11

Monitoring and Managing Networks

199

Network Management

3 Type the computer name in the Enter machine name field and click Apply.
PATROL writes the results to a PATROL task object (Ping) in the NETWORK container.

4 Access the Ping task object as described in Accessing KM Commands and

InfoBoxes on page 402 and view the results. Your results should resemble this example:
Ping to Paris_database_server Paris_database_server is alive -------------------- >>>>> DONE <<<<< --------------------

Logging In Remotely by Using rlogin

This task describes how to use the rlogin utility within PATROL to logon a remote computer.

To Logon a Remote Computer on the Network 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Utilities => Rlogin.

PATROL displays the Rlogin dialog box (Figure 34). Figure 34 Rlogin Dialog Box

3 Type the computer name in the Enter machine name field and click Apply.
PATROL writes the results to a PATROL task object (Rlogin) in the NETWORK container.

4 Access the Rlogin task object as described in Accessing KM Commands and

InfoBoxes on page 402 and view the results.

200

PATROL Knowledge Module for Unix User Guide

Address Resolution Protocol (ARP)

Querying the Name Server for the Network

This task describes how to query the name server from within PATROL.

To Query the Name Server for the Network 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Utilities => Query Name Server.

PATROL displays the PATROL Query Name Server dialog box with a command prompt.

Address Resolution Protocol (ARP)

The Address Resolution Protocol (ARP) is a TCP/IP protocol that is used to obtain a network nodes physical address. ARP operates at layers 2 (Data Link) and 3 (Network) of the Open Systems Interconnection (OSI) model. The physical address, also referred to as the Machine Access Control (MAC) address, is the unique serial number that identifies one network card from another. The network maintains a list of devices, IP addresses, and physical addresses in the ARP cache. With the NETWORK KM, you can view the ARP cache and add or delete resources (computers, printers, routers, anything with a network card) from it. The NETWORK tasks related to Address Resolution Protocol (ARP) are described in detail on the following pages: Task Viewing the Address Resolution Protocol (ARP) Cache Deleting a Host from the ARP Cache Adding a Host to the ARP Cache Page 202 203 205

Chapter 11

Monitoring and Managing Networks

201

Address Resolution Protocol (ARP)

Viewing the Address Resolution Protocol (ARP) Cache

This task describes how to view the Address Resolution Protocol (ARP) cache, which consists of the Internet-to-Ethernet address translation tables. ARP is a TCP/IP protocol used to obtain a network nodes physical address (the Machine Access Control address). The ARP cache is a list of each network device, its IP address, mask, and physical address.

To View the ARP Cache 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Utilities => ARP => Show ARP cache.
PATROL writes the results to a PATROL task object (Show ARP Cache) in the NETWORK container.

3 Access the Show ARP Cache task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example: # PSL Serial No.: ########## <NOT INTERACTIVE> Net to Media Table: IPv4 Device hme0 hme0 hme0 hme0 hme0 hme0 hme0 hme0 hme0 hme0 IP Address starbelly01 catnhat05 sneetch spot javaweb gate19 bluefish redfish greenegg BASE-ADDRESS.MCAST.NET Mask 255.255.255.255 255.255.255.255 255.255.255.255 255.255.255.255 255.255.255.255 255.255.255.255 255.255.255.255 255.255.255.255 255.255.255.255 240.0.0.0 Flags Phys Addr 00:10:4b:39:a5:d3 00:06:5b:ee:23:c8 00:03:47:06:dd:1f 02:60:6d:21:12:8c 00:50:8b:f4:fd:66 00:00:0c:07:ac:13 08:00:20:a2:90:68 00:10:4b:39:40:77 00:d0:b7:b1:21:59 01:00:5e:00:00:00

SM

-------------------- >>>>> DONE <<<<< --------------------

202

PATROL Knowledge Module for Unix User Guide

Address Resolution Protocol (ARP)

Output Format
The output has the following format. Table 41 defines each field.
Device IP Address Mask ----- -------------------- --------------hme0 meridian.xyz.com 255.255.255.255 Flags ----p Phys Addr --------------01:00:b2:83:8c:5b

Table 41 Field Device

Show ARP Cache Command Output Format Description device name assigned to the physical address the numerical address or its text string alias subnet mask Ppublish, includes IP address for the machine and the addresses that have been explicitly added using the Add Host to Cache (arp -s) option Sstatic, not learned for the ARP protocol Uunresolved, waiting for ARP response Mmapping, only used for multicast entry for 224.0.0.0

IPAddress Mask Flags

Phys Address

the physical address, also known as the Machine Access Control (MAC) address

Deleting a Host from the ARP Cache

This task describes how to delete a host (resource) from the Address Resolution Protocol (ARP) cache.

Before You Begin

You must be root to run this command.

Chapter 11

Monitoring and Managing Networks

203

Address Resolution Protocol (ARP)

To Delete a Host from the ARP Cache 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Utilities => ARP => Delete Host From Cache.
PATROL displays the Delete Host From Cache dialog box (Figure 35). Figure 35 Delete Host from Cache Dialog Box

3 Type the computer name in the Enter host name field and click Apply.
PATROL prompts you for the password for the root account (Figure 36). Figure 36 Delete Host From Cache Confirmation Dialog Box

4 Type the root password and click OK.

PATROL writes the results to the PATROL standard output window. Your results should resemble this example:
CCYYMMDDHHMMSS Dallas_email_server (255.255.255.255) deleted

204

PATROL Knowledge Module for Unix User Guide

Address Resolution Protocol (ARP)

Adding a Host to the ARP Cache

This task describes how to add a host (resource) to the Address Resolution Protocol (ARP) cache.

To Add a Host to the ARP Cache 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select TCP/IP => Utilities => ARP => Add Host To Cache.
PATROL displays the Add Host From Cache dialog box (Figure 37). Figure 37 Add Host To Cache Dialog Box

3 Type the computer name in the Enter host name field and click Apply.
PATROL writes the results to the PATROL standard output window. Your results should resemble this example:
CCYYMMDDHHMMSS Add Dallas_email_server to ARP Cache

Chapter 11

Monitoring and Managing Networks

205

Unix to Unix Communication Protocol (UUCP)

Unix to Unix Communication Protocol (UUCP)

Unix to Unix Communication Protocol (UUCP) is a Unix utility that copies a file from one computer to another. Unlike TCP/IP, which is a routable communications protocol, UUCP provides a point-to-point transmission where a user at one Unix computer dials up and establishes a session with another Unix computer. UUCP can be used to send e-mail using a bang path, which specifies the entire route to the destination computer and uses exclamation points (!), also known as bangs, as delimiters. The bang path format is generally: host_n!host_2!host_1!user_account. The NETWORK tasks related to UUCP are described in detail on the following pages: Task Checking for UUCP Support Files and Directories Cleaning Up UUCP Files in the Spool Directory Copying Files Using UUCP Listing All UUCP Servers Viewing the UUCP Log File Viewing UUCP Statistics Page 206 207 208 210 210 211

Checking for UUCP Support Files and Directories

This task describes how to verify the presence of the files and directories that UUCP requires.

Before You Begin

You must be root to run this command.

To Check for UUCP Support Files and Directories 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select UUCP => Utilities => UUCP Check.

PATROL prompts you for the password for the root account.

206

PATROL Knowledge Module for Unix User Guide

Unix to Unix Communication Protocol (UUCP)

3 Type the root password and click OK.

PATROL writes the results to a PATROL task object (UUCP Check) in the NETWORK container.

4 Access the UUCP Check task object as described in Accessing KM Commands

and InfoBoxes on page 402 and view the results.

Cleaning Up UUCP Files in the Spool Directory

This task describes how to delete all the files in the spool directory.

Before You Begin

You must be root to run this command.

NOTE
This task is not supported on SOLARIS systems.

To Delete Files in the Spool Directory 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select UUCP => Utilities => UUCP Clean.

PATROL prompts you for the password for the root account.

3 Type the root password and click OK.

PATROL writes the results to a PATROL task object (UUCP Clean) in the NETWORK container.

4 Access the UUCP Clean task object as described in Accessing KM Commands and
InfoBoxes on page 402 and view the results.

Chapter 11

Monitoring and Managing Networks

207

Unix to Unix Communication Protocol (UUCP)

Copying Files Using UUCP

This task describes how to copy files using UUCP.

Before You Begin

You must be root to run this command.

To Copy Files 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select UUCP => Utilities => UUCP.

PATROL displays the UUCP dialog box (Figure 38). Figure 38 UUCP Dialog Box

3 Specify what file you want to copy, how you want it copied, and where you want it
copied.
Enter optionsthe copy options control how the copy/transfer is performed, provide for user notification via mail, and provide job statistics as well as debug information

Options Description -c -C -d -f do not copy local file to the spool directory for transfer to a remote machine (default) copy local file to spool directory for transfer make all necessary directories for the file copy (default) do not make intermediate directories for the file copy

208

PATROL Knowledge Module for Unix User Guide

Unix to Unix Communication Protocol (UUCP)

Options Description -gx assign a service grade, x, which is the priority of the data transfer You must use the uuglist command to determine the appropriate format (single letter, number, or alphanumeric string) for the service grade -j -m -nuser -r -sfile -xn print the uucp job identification string on standard output notify, via mail, the user who initiated the transfer that it is complete notify, via mail, user when the file transfer is complete place the job in the file transfer queue but do not start the transfer send transfer status to file (full pathname) set debug level n (0-9); the higher the number the more detail debug information is provided

Enter sourcethe full pathname of the file to be copied Enter destinationthe full pathname to where you want the file copied; it can

be a file or a directory

4 Click Apply.
PATROL prompts you for the password for the root account.

5 Type the root password and click OK.

PATROL writes the results to a PATROL task object (UUCP) in the NETWORK container.

6 Access the UUCP task object as described in Accessing KM Commands and

InfoBoxes on page 402 and view the results.

Chapter 11

Monitoring and Managing Networks

209

Unix to Unix Communication Protocol (UUCP)

Listing All UUCP Servers

This task describes how to generate a list of all the hosts known to UUCP.

To List All UUCP Servers 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select UUCP => Reports => List Server Name.

PATROL writes the results to a PATROL task object (List Server Name) in the NETWORK container.

3 Access the List Server Name task object as described in Accessing KM Commands
and InfoBoxes on page 402 and view the results.

Viewing the UUCP Log File

This task describes how to view the UUCP log file.

To Display the UUCP Log File 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select UUCP => Reports => Show Log.

PATROL writes the results to a PATROL task object (Show Log) in the NETWORK container.

3 Access the Show Log task object as described in Accessing KM Commands and
InfoBoxes on page 402 and view the results.

210

PATROL Knowledge Module for Unix User Guide

Unix to Unix Communication Protocol (UUCP)

Viewing UUCP Statistics

This task describes how to display information about the UUCP requests.

To Display UUCP Statistics 1 Access the NETWORK application instance menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select UUCP => Reports => Show Statistics.

PATROL writes the results to a PATROL task object (Show Statistics) in the NETWORK container.

3 Access the Show Statistics task object as described in Accessing KM Commands

and InfoBoxes on page 402 and view the results.

Chapter 11

Monitoring and Managing Networks

211

Unix to Unix Communication Protocol (UUCP)

212

PATROL Knowledge Module for Unix User Guide

Chapter

12
214 214 216 216 217 218 218 219 220

12

Monitoring Network File Systems

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor various network components. The Network File System (NFS) provides you with an abstract view of the entire file system, regardless of the various systems architectures and operating systems. With the Network Information Service (NIS), you can look up resources distributed throughout the network. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network File System (NFS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network File System (NFS) Servers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Resetting Network File System Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network Information System (NIS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network Information System Accounts. . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network Information System Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Network Information System Servers . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 12 Monitoring Network File Systems

213

Introduction

Introduction
Together with TCP/IP, the Network File System (NFS) and the Network Information Service (NIS) construct the network on Unix.
s

In a heterogeneous environment, with NFS you can bind various file systems together, regardless of architecture and operating system. It presents the file system to the network as a uniform file hierarchy. Formerly known as Yellow Pages, NIS provides a distributed lookup service that you can use to locate resources anywhere in the network.

The NFS application class monitors these services and provides a means to gather information about which resources (servers, users, and groups) use or are used by these networking components.

Object Hierarchy
The NFS application class contains several parameters with information about network file system activity. Figure 39 on page 215 illustrates the basic hierarchy but does not necessarily display all instances and parameters.

214

PATROL Knowledge Module for Unix User Guide

Object Hierarchy

Figure 39

NFS Application Object Hierarchy

Chapter 12 Monitoring Network File Systems

215

Network File System (NFS)

Network File System (NFS)

NFS works by imposing an abstract model of a file system on all file systems. For each operating system, the NFS model is translated into the local file system commands. For example, a list directory contents command, ls, issued from a Unix machine with a DOS file system mounted on it, is translated to a dir command on the DOS node. With the NFS application class, you can monitor which servers are shared and can be mounted by other machines. You also can manage the NFS counters, resetting them to zero when necessary. The NFS tasks that relate to the network file system are described in detail on the following pages: Task Viewing Network File System (NFS) Servers Resetting Network File System Counters Page 216 217

Viewing Network File System (NFS) Servers

This task describes how to list which servers are Network File System servers. An NFS server is a server that makes its resources available for mounting by remote machines.

To View NFS Servers 1 Access the NFS application menu as described in Accessing KM Commands and
InfoBoxes on page 402.

2 Select NFS => Reports => List Servers.

PATROL writes the list of servers to a PATROL task object, List Servers NFS.

3 Access the List Servers NFS task object as described in Accessing KM Commands
and InfoBoxes on page 402 and view the results. Your results should resemble this example:
List NFS Server Houston_iv Florida_iii --------------- >>>>> DONE <<<<< ---------------

216

PATROL Knowledge Module for Unix User Guide

Network File System (NFS)

Resetting Network File System Counters

This task describes how to reset or zero out the Network File System counters.

Before You Begin

You must be root to run this command.

To View NFS Servers 1 Access the NFS application menu as described in Accessing KM Commands and
InfoBoxes on page 402.

2 Select NFS => Zero NFS Counters.

PATROL prompts you for the password for the root account. Figure 40 NFS User Name Dialog Box

3 Supply the required information and click OK.

PATROL resets the NFS counters and issues an nfs_stat command. It writes the commands output to the PATROL task object, Zero NFS Counters.

4 Access the Zero NFS Counters task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results.

Chapter 12 Monitoring Network File Systems

217

Network Information System (NIS)

Network Information System (NIS)

The Network Information System (NIS) works by distributing administrative databases, referred to as maps, throughout a network. Each domain has at least one map. These maps identify resources throughout the network and establish the resources level of accessibility. With the NFS application class, you can determine which servers store NIS maps. You also can view which users and groups are available via NIS. The NFS tasks that relate to NIS are described in detail on the following pages: Task Viewing Network Information System Accounts Viewing Network Information System Groups Viewing Network Information System Servers Page 218 219 220

Viewing Network Information System Accounts

This task describes how to list Network Information System accounts.

To View NIS Accounts 1 Access the NFS application menu as described in Accessing KM Commands and
InfoBoxes on page 402.

2 Select NIS => Reports => List Accounts.

PATROL writes the list of accounts to the PATROL task object, List Accounts.

3 Access the List Accounts task object as described in Accessing KM Commands

and InfoBoxes on page 402 and view the results. Your results should resemble this example:
tjsmith:X7kxaNoUE5vlc:25507:180:T. John Smith,Krakow, Request by Tammy Rea,TERMINATED:/home/trea:/bin/false psellars:H0zXwymSHE4gQ:58196:501:Peter Sellars, Full, paris, NO_EMPNUM:/home/award:/bin/ksh lpaulson:DcLKu6HBnjBi.:44444:111:Linus Paulson, , , :/home/lpaulson:/bin/ksh edgas:YmlBKzznaP82s:2005:105:Edgar Gas,,:/home/edgas:/bin/ksh --------------- >>>>> DONE <<<<< ---------------

218

PATROL Knowledge Module for Unix User Guide

Network Information System (NIS)

Output Format
The output has the following format. Table 42 defines each field. user name:password:accountid:seq:comments
-------------------- >>>>> DONE <<<<< --------------------

Table 42 Field password accountid seq

List of NIS Accounts Command Output Format Description encrypted password the account identifier, which has a maximum length of eight characters the index of the account id in the group of valid accounts for the username Zero (0) indicates that the account id is the default account for the user.

user name the login name of the user

comments text comments added during the account creation process

Viewing Network Information System Groups

This task describes how to list Network Information System groups.

To View NIS Groups 1 Access the NFS application menu as described in Accessing KM Commands and
InfoBoxes on page 402.

2 Select NIS => Reports => List Groups.

PATROL writes the list of accounts to the PATROL task object, List Groups.

3 Access the List Groups task object as described in Accessing KM Commands and
InfoBoxes on page 402 and view the results, which should resemble this example:
houston:*:212:aabaal,mhartma dallas:*:709:ccu,wfictes portland:*:4001:bwinsier,jguaglia montreal:*:1717:attack9,nmelcamp,vware,sgd,bbucca,package,skrishna katmandu:*:111:cstevens bayport:*:222:bayport singapore:*:333:njudd,lvandros,jdean,spamar paris:*:4444:oppp,mduchamp,mmarceau --------------- >>>>> DONE <<<<< ---------------

Chapter 12 Monitoring Network File Systems

219

Network Information System (NIS)

Output Format
The output has the following format. Table 43 defines each field.
group_name:password:###:member_1,member_2,member_X -------------------- >>>>> DONE <<<<< --------------------

Table 43 Field

List of NIS Groups Command Output Format Description name of the NIS group encrypted password The field is represented by an asterisk (*) and is usually left blank. numeric group ID (gid) list of user names of the members granted the privilege of group Members are separated by a comma (,).

group_name password ### member_X

Viewing Network Information System Servers

This task describes how to list Network Information System servers, which is a server that maintains a set of NIS Maps for other servers to query.

To View NIS Servers 1 Access the NFS application menu as described in Accessing KM Commands and
InfoBoxes on page 402.

2 Select NIS => Reports => List Servers.

PATROL writes the list of accounts to the PATROL task object, List Servers.

3 Access the List Servers task object as described in Accessing KM Commands and
InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> Print NIS servers zenith zenith titan titan xanadu xanadu patois patois --------------- >>>>> DONE <<<<< ---------------

220

PATROL Knowledge Module for Unix User Guide

Network Information System (NIS)

Output Format
The output has the following format. Table 44 defines each field.
server_name NIS_map_key -------------------- >>>>> DONE <<<<< --------------------

Table 44 Field

List of NIS Groups Command Output Format Description name of the NIS master server or NIS slave server the keys for those maps in which the values are null or the key is not part of the value

server_name NIS_map_key

Chapter 12 Monitoring Network File Systems

221

Network Information System (NIS)

222

PATROL Knowledge Module for Unix User Guide

Chapter

13

13

Monitoring and Managing Printers

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor and manage printers, print queues, and print jobs. The following sections appear in this chapter.

NOTE
At installation, the PRINTER application class is set to inactive. To activate the PRINTER application class and collect data for its parameters, see Activating the PRINTER Application Class on page 225. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Printer Activation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Activating the PRINTER Application Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deactivating the PRINTER Application Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . Printers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Excluding a Printer from Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selecting Printers to Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Number of Print Jobs in a Printers Queue . . . . . . . . . . . . . . . . . . . . Stopping a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Print Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing All Print Jobs Assigned to a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing All Print Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Print Job . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting All Print Jobs from a Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Moving a Print Job to Another Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Moving All Print Jobs to Another Printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Print Queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stopping a Print Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting a Print Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 224 225 225 226 226 227 227 228 228 229 231 231 232 233 234 235 236 237 238 239

Chapter 13

Monitoring and Managing Printers

223

Introduction

Introduction
The PRINTER application monitors the printers on the system to determine how many jobs are waiting to be processed. It also lets you manage printing functions from within PATROL. Some of the tasks that you can perform are
s s s s

moving print jobs deleting print jobs starting and stopping printers starting and stopping print queues

Object Hierarchy
The PRINTER application class contains an instance for each printer that is being monitored. Each printer instance contains a parameter that monitors the length of its respective print queue. Figure 41 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 41 PRINTERS Application Object Hierarchy

224

PATROL Knowledge Module for Unix User Guide

Printer Activation

Printer Activation
The PRINTER application and its corresponding collector, PRINTERColl, are installed in an inactive state. They do not automatically gather information when loaded. The PRINTER tasks related to activating and deactivating the PRINTER application class and PRINTERColl collector parameter are described in detail on the following pages: Task Activating the PRINTER Application Class Deactivating the PRINTER Application Class Page 225 226

WARNING
Use the same method to activate and deactivate application classes and their respective collectors. For example, if you use another method to deactivate the PRINTER application class (such as customizing it with a PATROL Console in developer mode), attempting to activate it with the menu command, Knowledge Module Admin => PRINTER Application => Activate Monitoring, will fail.

Activating the PRINTER Application Class

This task describes how to activate the PRINTER application class and the PRINTERColl collector parameter. This action causes PATROL to begin gathering information about the network and enables its network monitoring and managing features.

To Activate the PRINTER Application Class 1 Access the UNIX OS application KM menu commands as described in Accessing
KM Commands and InfoBoxes on page 402.

2 Select Knowledge Module Admin => PRINTER Application => Activate Monitoring.
PATROL begins collecting data for the PRINTER application class.

Chapter 13

Monitoring and Managing Printers

225

Printers

Deactivating the PRINTER Application Class

This task describes how to deactivate the PRINTER application class and the PRINTERColl collector parameter. This action causes PATROL to quit gathering information about the network and disables its network monitoring and managing features.

To Deactivate the PRINTER Application Class 1 Access the UNIX OS application KM menu commands as described in Accessing
KM Commands and InfoBoxes on page 402.

2 Select Knowledge Module Admin => PRINTER Application => Deactivate Monitoring.
PATROL quits collecting data for the PRINTER application class.

Printers
The PRINTER application class enables you to monitor and manage individual printers. By default, PATROL monitors all printers on the system. The PRINTER tasks related to printers are described in detail on the following pages: Task Excluding a Printer from Monitoring Selecting Printers to Monitor Viewing the Number of Print Jobs in a Printers Queue Stopping a Printer Starting a Printer Page 227 227 228 228 229

226

PATROL Knowledge Module for Unix User Guide

Printers

Excluding a Printer from Monitoring

This task describes how to stop monitoring a single printer.

Before You Begin

To perform this task, you must be in developer mode. If you attempt to stop monitoring a printer when in operator mode, PATROL displays the following error message in the system output window.
PRINTER: Only Development Console can filter an instance

To Stop Monitoring a Printer

On the printer that you do not want to monitor, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Stop Monitoring.

PATROL removes the printer icon from the PATROL Object Hierarchy and stops monitoring the selected printer. To resume monitoring of the printer, see Selecting Printers to Monitor on page 227.

Selecting Printers to Monitor

This task describes how to determine which printers PATROL excludes from monitoring and how to resume monitoring printers that have been excluded from monitoring.

Before You Begin

s

By default, the PRINTER application class is inactive. Once you activate the PRINTER application class and collector, PATROL monitors all printers. The exclusion list feature supports regular expressions. To stop monitoring large numbers of printers that have uniform names, use regular expressions. To perform this task, you must be in developer mode. If you attempt to stop monitoring a printer when in operator mode, PATROL displays the following error message in the system output window.

PRINTER: Only Development Console can filter an instance

Chapter 13

Monitoring and Managing Printers

227

Printers

To Stop Monitoring Printers

On any printer instance, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Edit PRINTER Exclusion List.

PATROL opens the PRINTER Exclusion List dialog box, which displays a list of monitored printers and a list of excluded printers.

3 In the Monitored Printers list box, select the name of the printer that you want to
exclude from monitoring.

4 Click the right arrow to move the printer into the Exclued Printers list box 5 Click Apply.
PATROL adds the printer to the list of excluded printers and then removes the printer icon from the PATROL Object Hierarchy.

6 Repeat step 3 through step 5 for each printer that you want to exclude.

Viewing the Number of Print Jobs in a Printers Queue

This task describes how to view the total number of print jobs in a printers queue on the system.

To View the Total Number of Print Jobs in a Printer Queue 1 Access the PRINTER application so that you can view its parameters. 2 Open the PRNQLength parameter.
PATROL displays a gauge that shows the number of print jobs assigned to the printers queue.

Stopping a Printer
This task describes how to disable a selected printer so that the printer cannot process print jobs.

228

PATROL Knowledge Module for Unix User Guide

Printers

Before You Begin

s

You must be root to run this command. Because the print manager will continue to add print jobs to the print queue while the printer is unavailable, use this command only if the printer will be unavailable for a short time. If the printer will be unavailable for long periods of time, stop the printers queue. For more information on this task, see Stopping a Print Queue on page 238 command.

To Stop a Printer
On the printer that you want to stop so that it cannot process any more print jobs, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Stop Printer.

PATROL prompts you for the password for the root account.

3 Type the password and click OK.

s

If the print queue is enabled when you stop the printer, PATROL changes the state of the PRINTER instance to ALARM. If the print queue is disabled or stopped when you stop the printer, PATROL changes the state of the PRINTER instance to OFFLINE.

PATROL changes the printer icon to OFFLINE. To restart the printer from within PATROL, see Starting a Printer on page 229.

Starting a Printer
This task describes how to enable a selected printer so that the printer can process print jobs.

Before You Begin

s s

You must be root to run this command. Make sure that the print queue associated with this printer is running.

Chapter 13

Monitoring and Managing Printers

229

Printers

To Start a Printer
On the printer that you want to start, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Start Printer.

PATROL prompts you for the password for the root account.

3 Type the password and click OK.

s

If the print queue is enabled or started when you start the printer, PATROL changes the state of the PRINTER instance to OK. If the print queue is disabled or stopped when you start the printer, PATROL changes the state of the PRINTER instance to WARN.

PATROL changes the printer icon to OK and the printer begins processing any jobs that have collected in its queue. To stop the printer from within PATROL, see Stopping a Printer on page 228.

230

PATROL Knowledge Module for Unix User Guide

Print Jobs

Print Jobs
The PRINTER application class enables you to monitor and manage print jobs. With this feature, you can delete and move print jobs from busy or out-of-service printers. The tasks related to print jobs are described in detail on the following pages: Task Viewing All Print Jobs Assigned to a Printer Viewing All Print Jobs Deleting a Print Job Deleting All Print Jobs from a Printer Moving a Print Job to Another Printer Moving All Print Jobs to Another Printer Page 231 232 233 234 235 236

Viewing All Print Jobs Assigned to a Printer

This task describes how to view a list of all the print jobs assigned to a specific printer.

Before You Begin

s

To view a list of all print jobs owned by a particular user, see Viewing Print Jobs that Belong to a User on page 375. To view a list of all print jobs assigned to a printer, see Viewing All Print Jobs on page 232

To View a List of Print Jobs Assigned to a Specific Printer

On the printer of whose jobs you want a list, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select List Print Jobs.

PATROL writes the information to a PATROL task object (List Print Jobs) in the UNIX OS container.

3 Access the List Print Jobs task object as described in Accessing KM Commands
and InfoBoxes on page 402 and view the results.
Chapter 13 Monitoring and Managing Printers 231

Print Jobs

Viewing All Print Jobs

This task describes how to view a list of all the print jobs on the system.

Before You Begin

s

To view a list of all print jobs owned by a particular user, see Accessing KM Commands and InfoBoxes on page 402. To view a list of all print jobs assigned to a printer, see Viewing All Print Jobs Assigned to a Printer on page 231

To View a List of All Print Jobs

On any of the printers, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select List All Print Jobs.

PATROL writes the information to a PATROL task object (List All Print Jobs) in the UNIX OS container.

3 Access the List All Print Jobs task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
Display printer job list for all printer(s) lp4-0 lysander!root 182 lp4-3 wenchell!tmavis 4182 Epson2-1 logger!dpatel 3750 Epson2-3 sandhill!whill 005 Epson3-5 hp-dev!jjjones 7431 --------------------- >>>>> DONE <<<<< MMM DD MM:SS MMM DD MM:SS MMM DD MM:SS MMM DD MM:SS MMM DD MM:SS ---------------------

Output Format
The output has the following format. Table 45 on page 233 defines each field.
printer-# system_name!user job_# MMM DD MM:SS

-------------------- >>>>> DONE <<<<< --------------------

232

PATROL Knowledge Module for Unix User Guide

Print Jobs

Table 45 Field printer-#

View All Print Jobs Command Output Format Description the printer to which the job is assigned and its position in the print queue

system_name!user the name of the system from which the print job originated and the user who created it The exclamation point (!) is a delimiter. job_# MMM DD MM:SS the print job ID number the time at which the print job was created

Deleting a Print Job

This task describes how to delete a print job from a print queue.

Before You Begin

s

You must be root to run this command. To perform this task, you must be in developer mode. If you attempt to stop monitoring a printer when in operator mode, PATROL displays the following error message in the system output window.

PRINTER: Only Development Console can filter an instance

To Delete a Print Job from a Printer

On the printer whose queue contains the print job that you want to delete, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Delete Job.

PATROL displays the Delete Print Job dialog box (Figure 42). Figure 42 Delete Print Job Dialog Box

Chapter 13

Monitoring and Managing Printers

233

Print Jobs

3 Type the print job number in the Enter Job ID field and click on Apply. To learn the
number of a print job, see Viewing All Print Jobs on page 232. PATROL prompts you for the password for the root account.

4 Type the password and click OK.

PATROL deletes the print job.

Deleting All Print Jobs from a Printer

This task describes how to delete all print jobs from a printers queue.

Before You Begin

You must be root to run this command.

To Delete All Print Jobs from a Printer

On the printer from which you want to delete all jobs, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Delete All Jobs.

PATROL prompts you for the password for the root account (Figure 43). Figure 43 Delete All Jobs Dialog Box

3 Type the password and click OK.

PATROL deletes all print jobs.

234

PATROL Knowledge Module for Unix User Guide

Print Jobs

Moving a Print Job to Another Printer

This task describes how to move a print job from one printer to another. This feature is useful when a printer is busy and you have a job that needs to run immediately.

Before You Begin

You must be root to run this command.

To Move a Print Job from One Printer to Another

On the printer whose queue contains the print job that you want to move, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Move Job to Another Printer.

PATROL displays the Move Job to Another Printer dialog box (Figure 44). Figure 44 Move Job to Another Printer Dialog Box

3 Type the print job number in the Enter Job Id field. To learn the number of a print
job, see Viewing All Print Jobs on page 232.

Chapter 13

Monitoring and Managing Printers

235

Print Jobs

4 Type the name of the printer to which you want to move the job and click Apply.
PATROL prompts you for the password for the root account (Figure 45). Figure 45 Move Jobs To Another Printer Confirmation Dialog Box

5 Type the password and click OK.

PATROL moves the print job from the current printer to the designated printer.

Moving All Print Jobs to Another Printer

This task describes how to move all print jobs assigned to one printer to another printer. This feature is useful when a printer is offline.

Before You Begin

You must be root to run this command.

To Move All Print Jobs from One Printer to Another

On the printer whose jobs you want to move, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Move All Jobs to Another Printer.

PATROL displays the Move All Jobs to Another Printer dialog box (Figure 46). Figure 46 Move All Jobs to Another Printer Dialog Box

236

PATROL Knowledge Module for Unix User Guide

Print Queues

3 Type in the Enter New Destination field the name of the printer to which you want
to move the job and click Apply. PATROL prompts you for the password for the root account (Figure 47). Figure 47 Move Jobs To Another Printer Confirmation Dialog Box

4 Type the password and click OK.

PATROL moves all print jobs from the current printer to the designated printer.

Print Queues
The PRINTER application class enables you to monitor and manage print queues. With this feature, you can start and stop queues so that the system print manager will not assign print jobs to printers that are down. The PRINTER tasks related to print queues are described in detail on the following pages: Task Stopping a Print Queue Starting a Print Queue Page 238 239

Chapter 13

Monitoring and Managing Printers

237

Print Queues

Stopping a Print Queue

This task describes how to disable a selected print queue so that the print manager does not assign any more jobs to it.

Before You Begin

s

You must be root to run this command. Use this command if the printer will be unavailable for long periods of time. If the printer will only be unavailable for a short period of time, stop the printer. For more information on this task, see Stopping a Printer on page 228 command. You can also disable spooling to this printer.

To Stop a Print Queue

On the printer whose queue you want to stop, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Stop Queue.

PATROL prompts you for the password for the root account.

3 Type the password and click on OK.

s

If the printer is enabled or online when you stop the print queue, PATROL changes the state of the PRINTER instance to WARN. If the printer is disabled or stopped when you stop the print queue, PATROL changes the state of the PRINTER instance to OFFLINE.

To restart the print queue from within PATROL, see Starting a Print Queue on page 239.

238

PATROL Knowledge Module for Unix User Guide

Print Queues

Starting a Print Queue

This task describes how to enable a selected print queue so that the print manager can assign print jobs to it.

Before You Begin

You must be root to run this command.

To Start a Print Queue

On the printer whose queue you want to start, perform the following steps.

1 Access the PRINTER application menu as described in Accessing KM Commands

and InfoBoxes on page 402.

2 Select Start Queue.

PATROL prompts you for the password for the root account.

3 Type the password and click on OK.

s

If the printer is enabled or online when you start the print queue, PATROL changes the state of the PRINTER instance to OK. If the printer is disabled or stopped when you start the print queue, PATROL changes the state of the PRINTER instance to ALARM.

To stop a print queue from within PATROL, see Stopping a Print Queue on page 238.

Chapter 13

Monitoring and Managing Printers

239

Print Queues

240

PATROL Knowledge Module for Unix User Guide

Chapter

14

14

Monitoring Processes
The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor and manage processes on a system. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 Configuring Multiple Instances of a Process for Monitoring . . . . . . . . . . . . . . . . 247 Configuring a Process to Alarm When the Number of Running Instances Exceeds a Set Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 Configuring a Process to Alarm When the Number of Running Instances Falls Below a Set Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Configuring a Process to Restart When PATROL Detects It Has Stopped. . . . . 256 Configuring a Process to Stop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Configuring a Process to Alarm When Its Parent Process ID (PPID) is Not 1 . . 262 Configuring a Process to Alarm When It is Owned by an Unauthorized User . 265 Configuring a Process to Suspend Its Alert for N Cycles if Patrol Detects It is Not Running . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 Modifying the Configuration of a Monitored Process . . . . . . . . . . . . . . . . . . . . . . 270 Monitoring Process Presence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 Deleting a Monitored Process from the Monitoring Queue . . . . . . . . . . . . . . . . . 278 Specifying a Flag File Directory for Process Presence Blackout Periods. . . . . . . 278 Suspending Process Monitoring for a Process Instance . . . . . . . . . . . . . . . . . . . . 280 Resuming Process Monitoring for a Process Instance . . . . . . . . . . . . . . . . . . . . . . 280 Reporting on a Selected Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 Process Host Groups in a Cluster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Rules for Creating Process Host Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Creating a Host Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 Configuring Communication Attributes for a Process Host Group . . . . . . . . . . 284 Modifying Host Attributes within a Process Group . . . . . . . . . . . . . . . . . . . . . . . 285 Removing a Host from a Host Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 Querying a Selected Host within a Process Host Group . . . . . . . . . . . . . . . . . . . . 286 Synchronizing the Local Hosts List to a Selected Hosts List for a Process Host Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 Synchronizing a Selected Hosts List to the Local Host List for a Process Host Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 Viewing the Process List for a Selected Host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Chapter 14 Monitoring Processes 241

Synchronizing the Local Processs List with a Selected Process's List . . . . . . . . . 289 Synchronizing a Selected Process's List with the Local Process's List . . . . . . . . . 290 Viewing the Process Presence List for a Selected Process . . . . . . . . . . . . . . . . . . . 291 Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Viewing the Number of Zombie Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Viewing Zombie Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 Viewing the Total Number of Processes Running . . . . . . . . . . . . . . . . . . . . . . . . . 295 Viewing Processes that Use Excessive CPU. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Viewing the Number of Processes Waiting on Resources . . . . . . . . . . . . . . . . . . . 298 Viewing the Total Number of User Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Viewing the Average Number of User Processes . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Priority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Reprioritizing a Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 Reprioritizing All Processes in a Process Group . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Reprioritizing All Processes Owned by a User . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 Plug-in Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 Configuring the Process KM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

242

PATROL Knowledge Module for Unix User Guide

Introduction

Introduction
The PATROL Knowledge Module for Unix and Linux monitors and manages processes through the PROCESS and PROCESS_PRESENCE applications and the Process KM Configuration Plug-in for the PATROL Configuration Manager (PCM). With these application classes, you can
s s

designate specific processes for monitoring review the overall status of processes on the system such as total number of processes total number of zombie processes list of all zombie processes list of all processes consuming excessive CPU total number of processes waiting for resources total number of user (nonroot) processes average number of processes per user reprioritize process by individual process user ownership process group membership

You can use the PROCESS_PRESENCE application class to monitor

s

a single process on a single computer a local process across multiple computers within a process host group (this feature is helpful when you are monitoring processes in a clustered environment) several running instances of a process on a single computer several running instances of a process across multiple computers within a process host group (this feature is helpful when you are monitoring processes in a clustered environment)

You can manage a process by way of its PROCESS_PRESENCE instance to

s

restart a process that has stopped stop a process alarm when an unauthorized User ID owns a process alarm when the number of processes running in a host group drops below a specified minimum number
Chapter 14 Monitoring Processes 243

Object Hierarchy

alarm when the number of processes running in a host group exceeds a specified maximum number alarm when a process does not have a parent process ID (PPID) of 1 (a PPID of 1 indicates that a process is owned by init)

You can use the plug-in for PCM to configure process monitoring without using a console. See Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux on page 304 for details.

Object Hierarchy
The PROCESS application class contains the PROCESS_PRESENCE application class and process parameters. The PROCESS_PRESENCE application class contains instances for individual processes that are being monitored at your request. Each instance contains parameters with information on the individual processs memory and CPU usage. Figure 48 on page 245 illustrates the basic hierarchy but does not necessarily display all instances and parameters.

244

PATROL Knowledge Module for Unix User Guide

Object Hierarchy

Figure 48

PROCESS Application Object Hierarchy

OS container
UNIX OS

Application class Parameters and PROCESS_PRESENCE class

PROCESS_PRESENCE Instances

PROCESS_PRESENCE Parameters

Chapter 14

Monitoring Processes

245

Processes

Processes
The PROCESS application class monitors processes on the operating system level by default and provides general information, such as averages and totals, for all processes running on a managed system. This feature enables you to designate specific processes for monitoring. The PROCESS tasks that relate to individual processes are described in detail on the following pages: Task Configuring Multiple Instances of a Process for Monitoring Configuring a Process to Alarm When the Number of Running Instances Exceeds a Set Limit Configuring a Process to Alarm When the Number of Running Instances Falls Below a Set Limit Configuring a Process to Restart When PATROL Detects It Has Stopped Configuring a Process to Stop Configuring a Process to Alarm When It is Owned by an Unauthorized User Page 247 250 253 256 259 265

Configuring a Process to Alarm When Its Parent Process ID (PPID) is Not 1 262

Configuring a Process to Suspend Its Alert for N Cycles if Patrol Detects It is 268 Not Running Modifying the Configuration of a Monitored Process Monitoring Process Presence Deleting a Monitored Process from the Monitoring Queue Specifying a Flag File Directory for Process Presence Blackout Periods Suspending Process Monitoring for a Process Instance Resuming Process Monitoring for a Process Instance Reporting on a Selected Process 270 274 278 278 280 280 280

246

PATROL Knowledge Module for Unix User Guide

Processes

Configuring Multiple Instances of a Process for Monitoring

This task describes how to configure PATROL to monitor multiple instances of a process.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Process Instance Label field, type a name for the monitored process icon. 6 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

7 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. To monitor multiple instances of the same process, this value must be set to 2 or above. If the number of running process instances falls below the value set in this field, PATROL will issue an alert.

8 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

Chapter 14

Monitoring Processes

247

Processes

step 9 through step 18 are optional when you are configuring multiple instances of a process for monitoring. If you do not want to set values for the fields as indicated in these steps, you do not have to.

9 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

10 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler.

11 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

12 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

13 In the Start Command field, type the command string that will start the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Start command.

14 In the Stop Command field, type the command string that will stop the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Start command.

15 In the Command Execution User Name field, type the user ID under which the
command will be executed.
248 PATROL Knowledge Module for Unix User Guide

Processes

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

16 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

18 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

19 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer, or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

20 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

Chapter 14

Monitoring Processes

249

Processes

Configuring a Process to Alarm When the Number of Running Instances Exceeds a Set Limit
This task describes how to configure PATROL to alarm when the number of running instances for a process exceeds a set limit.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Blackout field, enter a blackout beginning and ending time (use 24-hour
time).

6 In the Process Instance Label field, type a name for the monitored process icon. 7 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous.

8 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. This field is optional.

9 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. You must type a value of 1 or greater in this field in order for PATROL to alarm when the process instance count is exceeded. When the number of running process instances exceeds the value you type in this field, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

250

PATROL Knowledge Module for Unix User Guide

Processes

step 10 through step 19 are optional when you are configuring a process to alarm if the number of running instances exceeds a set limit. If you do not want to set values for the fields as indicated in these steps, you do not have to.

10 In the Acceptable Process Owners field, type the user IDs for the accounts that can
own the process. Separate multiple user IDs with spaces.

11 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler.

12 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in this dialog box for PATROL to automatically restart a process.

13 In the Execution Attempts field, type a value to set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

14 In the Start Command field, type the command string that will start the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Start command.

15 In the Stop Command field, type the command string that will stop the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Stop command.

16 In the Command Execution User Name field, type the user ID under which the
command will be executed.

Chapter 14

Monitoring Processes

251

Processes

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

18 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

19 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

20 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

21 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

252

PATROL Knowledge Module for Unix User Guide

Processes

Configuring a Process to Alarm When the Number of Running Instances Falls Below a Set Limit
This task describes how to configure PATROL to alarm when the number of running instances for a process falls below a set limit.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Process Instance Label field, type a name for the monitored process icon. 6 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

7 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. If the number of running process instances falls below the value set in this field, PATROL will issue an alert. step 8 through step 18 are optional if you are configuring a process to alarm when the number of running instances falls below a set limit. If you do not want to set values for the fields as indicated in these steps, you do not have to.

Chapter 14

Monitoring Processes

253

Processes

8 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

9 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

10 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler.

11 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

12 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

13 In the Start Command field, type the command string that will start the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Start command.

14 In the Stop Command field, type the command string that will stop the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Start command.

254

PATROL Knowledge Module for Unix User Guide

Processes

15 In the Command Execution User Name field, type the user ID under which the
command will be executed.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

16 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

18 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

19 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer, or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

20 Click OK.

Chapter 14

Monitoring Processes

255

Processes

The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

Configuring a Process to Restart When PATROL Detects It Has Stopped

This task describes how to configure a process to restart if PATROL detects that it has stopped.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Process Instance Label field, type a name for the monitored process icon. 6 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

step 7 through step 10 are optional when you are configuring PATROL to restart a process. If you do not want to set values for the fields as indicated in these steps, you do not have to.

256

PATROL Knowledge Module for Unix User Guide

Processes

7 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. If the number of running process instances falls below the value set in this field, PATROL will issue an alert.

8 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

9 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

10 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler.

11 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

12 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

13 In the Start Command field, type the command string that will start the process
instance.

14 In the Stop Command field, type the command string that will stop the process
instance. This field is optional if you are configuring PATROL to restart a process that has stopped.

15 In the Command Execution User Name field, type the user ID under which the
command will be executed.

Chapter 14

Monitoring Processes

257

Processes

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

16 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

18 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

19 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer, or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

20 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

258

PATROL Knowledge Module for Unix User Guide

Processes

Configuring a Process to Stop

This task describes how to configure PATROL to stop a process. If the PATROL KM for Unix detects that the number of running process instances exceeds the value set in the Maximum Count field, and you have provided a command to stop the process in the Stop Command field, the KM will attempt to stop process instances until the number of running instances falls within the set threshold.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Process Instance Label field, type a name for the monitored process icon. 6 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

step 7 through step 13 are optional when you are configuring PATROL to stop a process. If you do not want to set values for the fields as indicated in these steps, you do not have to.

7 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. If the number of running process instances falls below the value set in this field, PATROL will issue an alert.

Chapter 14

Monitoring Processes

259

Processes

8 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

9 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

10 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler.

11 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

12 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

13 Leave the Start Command field blank. 14 In the Stop Command field, type the command string that will stop the process
instance.

15 In the Command Execution User Name field, type the user ID under which the
command will be executed.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

16 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

260

PATROL Knowledge Module for Unix User Guide

Processes

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

step 17 and step 18 are optional if you are configuring PATROL to stop a process. If you do not want to set values for the fields as indicated in these steps, you do not have to.

17 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

18 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

19 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer, or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

20 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

Chapter 14

Monitoring Processes

261

Processes

Configuring a Process to Alarm When Its Parent Process ID (PPID) is Not 1

This task describes how to configure PATROL to alarm when the parent process ID (PPID) for a process is not 1. A parent process ID of 1 indicates that a process is owned by init, or the Unix scheduler.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Process Instance Label field, type a name for the monitored process icon. 6 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

step 7 through step 9 are optional when you are configuring PATROL to alarm when the PPID for a process is not 1. If you do not want to set values for the fields as indicated in these steps, you do not have to.

7 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. If the number of running process instances falls below the value set in this field, PATROL will issue an alert.

262

PATROL Knowledge Module for Unix User Guide

Processes

8 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

9 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

10 Select the Parent Process ID Must Be 1 option button.

Step 11 through step 18 are optional when you are configuring PATROL to alarm when the PPID for a process is not 1. If you do not want to set values for the fields as indicated in these steps, you do not have to.

11 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

12 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

13 In the Start Command field, type the command string that will start the process
instance.

14 In the Stop Command field, type the command string that will stop the process
instance.

15 In the Command Execution User Name field, type the user ID under which the
command will be executed.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

Chapter 14

Monitoring Processes

263

Processes

16 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

18 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

19 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer, or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

20 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

264

PATROL Knowledge Module for Unix User Guide

Processes

Configuring a Process to Alarm When It is Owned by an Unauthorized User

This task describes how to configure PATROL to alarm when a process is owned by an unauthorized user.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Process Instance Label field, type a name for the monitored process icon. 6 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

step 7 and step 8 are optional when you are configuring PATROL to alarm when a process is owned by an unauthorized user. If you do not want to set values for the fields as indicated in these steps, you do not have to.

7 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. If the number of running process instances falls below the value set in this field, PATROL will issue an alert.

Chapter 14

Monitoring Processes

265

Processes

8 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

9 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

10 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler. Step 11 through step 18 are optional when you are configuring PATROL to alarm when a process is owned by an unauthorized user. If you do not want to set values for the fields as indicated in these steps, you do not have to.

11 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

12 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

13 In the Start Command field, type the command string that will start the process
instance.

14 In the Stop Command field, type the command string that will stop the process
instance.

15 In the Command Execution User Name field, type the user ID under which the
command will be executed.

266

PATROL Knowledge Module for Unix User Guide

Processes

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

16 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

18 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

19 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer, or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

20 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

Chapter 14

Monitoring Processes

267

Processes

Configuring a Process to Suspend Its Alert for N Cycles if Patrol Detects It is Not Running
This task describes how to configure a process to suspend its alert for a designated number of cycles when PATROL detects that the process is not running. This feature gives the system time to detect that a process has died and restart the process before PATROL issues an alarm. Delaying an alert also helps by giving the host group time to restart a process that died on an alternate host.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Actions group box, select Add New Process, and click OK.
The Add New Process dialog box is displayed.

5 In the Process Instance Label field, type a name for the monitored process icon. 6 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

step 7 through step 17 are optional when you are configuring PATROL to delay an alert for a process that has died. If you do not want to set values for the fields as indicated in these steps, you do not have to.

7 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. If the number of running process instances falls below the value set in this field, PATROL will issue an alert.

268

PATROL Knowledge Module for Unix User Guide

Processes

8 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

9 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

10 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler.

11 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

12 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

13 In the Start Command field, type the command string that will start the process
instance.

14 In the Stop Command field, type the command string that will stop the process
instance.

15 In the Command Execution User Name field, type the user ID under which the
command will be executed.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

16 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.
Chapter 14 Monitoring Processes 269

Processes

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts.

18 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

19 Select one of the following option buttons to indicate whether PATROL should
monitor the process on one host computer, or across a group of hosts: Option Monitor Only On This Host Monitor Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

20 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

Modifying the Configuration of a Monitored Process

This task describes how to modify the configuration of a monitored process.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

270

PATROL Knowledge Module for Unix User Guide

Processes

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Process List group box, select the process you want to modify. 5 In the Actions group box, select Modify Selected Process, and click OK.
The Modify Selected Process dialog box is displayed. Use the instructions in step 6 through step 19 to modify the monitoring attributes you would like to change.

6 In the Process Instance Label field, type a name for the monitored process icon. 7 In the Monitored Process String field, type the command or a unique command
string for the process that is being monitored.

NOTE
s

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous. You must specify a single process if you want to use the PATROL feature that allows you to restart a process automatically.

8 In the Minimum Count field, type a value to set the minimum number of process
instances that must be running on the local computer or in the host group. For example, if you want at least three instances of the process to run at all times, type 3 in this field. If the number of running process instances falls below this value, PATROL will issue an alert.

9 In the Maximum Count field, type a value to set the maximum number of process
instances that may be running on the local computer or in the host group. If the number of running process instances exceeds this value, PATROL will issue an alert. The value of the Maximum Count field must be greater than the value of the Minimum Count field.

10 In the Acceptable Process Owners field, type the user IDs for the accounts that may
own the process. Separate multiple user IDs with spaces.

11 Select the Parent Process ID Must Be 1 option button if the parent process ID (PPID)
must be 1. A process with a PPID of 1 is owned by init, or the Unix scheduler.

Chapter 14

Monitoring Processes

271

Processes

12 Select the Restart Automatically option button if you want the KM to automatically
restart a process when it detects that the process count is less than the set minimum. If you select this option button, the KM will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process.

NOTE
You must specify a start command and a command execution user name in the appropriate fields on this dialog box if you want PATROL to automatically restart a process.

13 In the Execution Attempts field, type a value set the number of times the host will
attempt to run a Start Process or Stop Process command before it stops trying to run the command.

14 In the Start Command field, type the command string that will start the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Start command.

15 In the Stop Command field, type the command string that will stop the process
instance.

NOTE
You must specify a command execution user account and password if you want to use the Start command.

16 In the Command Execution User Name field, type the user ID under which the
command will be executed.

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

17 In the Command Execution Password field, type the password for the user ID under
which the command will be executed. You can leave this field blank. The KM will accept a NULL value.

272

PATROL Knowledge Module for Unix User Guide

Processes

NOTE
You must specify a command execution user account and password if you want to use the Start and Stop commands.

18 In the Alert Delay Count field, type a value to set the number of collection intervals
that this host will defer an alert while it waits for the process count to be reestablished on a host or across a group of hosts. If you delay the alert, the system has time to detect that a process has died and restart it automatically before PATROL issues an alarm.

19 Click the drop-down list button in the Alert State field and select the state change
that will occur when the process count either falls below the minimum threshold or exceeds the maximum threshold and the alert delay count reaches 0. The state change will apply to the following parameters:
s s s

PROCPPCountCheck PROCPPOwnerCheck PROCPPParentID1

20 From the Monitoring Scope drop-down list, select whether you want PATROL to
monitor the process on one host computer or across a group of hosts: Option Only On This Host Across Host Group Description PATROL will only monitor the process on the selected local host computer. PATROL will monitor the process across a group of host computers. Select this option if you want to monitor a process in a clustered environment.

21 In the Blackout Period fields, enter a beginning and ending blackout time (use
24-hour clock).

22 Click OK.
The dialog box closes, and the Manage List of Monitored Processes dialog box is displayed. The process that you just added to the monitored queue is listed in the Process List group box.

Chapter 14

Monitoring Processes

273

Processes

Monitoring Process Presence

Process Presence monitoring is only available when using the Data Collection Manager (DCM) application and Perform Kernel Collector data collection.

To define Process Presence entries for PATROL for Unix and Linux 1 Drill down into the PROCESS container instance. 2 Select the PROCESS_PRESENCE icon and right-click to open its menu. 3 Select Manage List of Monitored Processes from the menu. NOTE
If the DCM collector is in use, it can take a while to create the PROCESS instance after which there may be another wait for the PROCESS_PRESENCE container instance to be created. If the PATROL Script Language (PSL) data collector is in use, the PROCESS_PRESENCE container will not be created.

Using Regular Expressions to Monitor Process Presence

You can use regular expressions defined by the REGEXP standard (REGEX API) to monitor process presence. However, because regular expression matching requires significant overhead, you should minimize its use. For example, if you want to match CRM80DM without getting CRM89DMO, use CRM89DM[^O] for the definition.

274

PATROL Knowledge Module for Unix User Guide

Processes

Agent Configuration Variables for Process Presence Monitoring

Table 46 contains descriptions of the variables used to monitor Process Presence. Table 46 Variable Variables for Monitoring Process Presence (Part 1 of 2) Description

/PUK/PROCCONT/daemonObjectList Created automatically. If deleted, it will be automatically recreated Internal use only. PCM configuration is not recommended. /PUK/PROCPRES/INSTANCE/info Defines individual processes to be monitored. The following instances are created by default: ==> amd, defined by the REGEXP standard (REGEX, bind, cron, sendmail, and ypbind. Mirrors all hosts in the host group if the monitoring scope is across all hosts. When a new instance definition is discovered, a new PROCPRES instance is created and monitoring begins. Before using PCM to apply a configuration change to an existing definition, restart the Patrol Agent to reflect the change. One per instance. PCM configuration is allowed. PUK/Process/Default/Created Created automatically. Internal use only. PCM configuration is not recommended. /UNIX/PROCCONT/processStamp Automatically created when you add a new process using the Manage List of Monitored Processes menu. If deleted, it is not recreated until the next use of Manage List of Monitored Processes to add another new process. Internal use only. PCM configuration is not recommended.

Chapter 14

Monitoring Processes

275

Processes

Table 46 Variable

Variables for Monitoring Process Presence (Part 2 of 2) Description Use of this option involves synchronization of several hosts and incorporates an encrypted password. Created when
s

/UNIX/PROCCONT/nodeList

you choose the Manage Host Group menu option. Defines list of hosts in host group. a host group is established and mirrored on all hosts added to the host group. Must be consistent across all hosts in the Group Host List. If this variable is deleted from one or more of the hosts: It is not automatically recreated. The host will not respond to requests from any of the other hosts in the host group. Attempts to use Synchronize=>Host Group will result in No Response. When a new process is added to one of the other nodes in the Group Host List that has Monitoring Scope set to Across All Hosts, the process definition is mirrored on the host. However, the group host list is not automatically mirrored.

PCM configuration is not recommended. /UNIX/PROCCONT/nodeStamp Created when the Manage Host Group menu option is used to define a host group. Internal use only. PCM configuration not recommended.

Using the /PUK/PROCPRES/INSTANCE/info Variable

Field definitions for the /PUK/PROCPRES/INSTANCE/info variable are closely aligned with the Add New Process dialog of Manage List of Monitored Processes menu option. Process Instance Label Monitored Process String Minimum Count label command string 1

276

PATROL Knowledge Module for Unix User Guide

Processes

Maximum Count Acceptable Process Owners Parent Process ID Must Be 1 Restart Automatically Start Command Stop Command Command Execution User Name Alert Delay Count Alert State Monitoring Scope Blackout Period start Blackout Period end

1 blank NO (YES = 1, NO = 2) NO (YES = 1, NO = 2) blank blank blank

Command Execution Attempts 5

Command Execution Password blank 1 ALARM (WARN = 1, ALARM = 2) Only on This Host (Only on This Host = 1, Across Host Group = 2) 0 (Hours) 0 (Hours) 0 (Minutes) 0 (Minutes)

You can apply the following example ruleset can by using Patrol Configuration Manager or pconfig to define two processes (newproc1 and newproc2). PATROL_CONFIG "/PUK/PROCPRES/newproc1/info" = { REPLACE = "newproc1^Bnewproc1^B1^B1^B^B2^B2^B5^B^B^B^B^B1^B2^B1^B0^B0^B0^B0" }, "/PUK/PROCPRES/newproc2/info" = { REPLACE = "newproc2^Bnewproc2^B1^B1^B^B2^B2^B5^B^B^B^B^B1^B2^B1^B0^B0^B0^B0" } The field separator is CTRL-B. On a Unix system, use the vi editor to enter the CTRL-B character. Be careful when you manually edit configuration entries. You may want to start with a configuration that was generated by KMs with Process Presence for PATROL for Unix. For example, use a ruleset captured from an existing configuration generated by using Process Presence to monitor KMs.

Using the /UNIX/PROCCONT/nodeList Variable

These are the field definitions for /UNIX/PROCCONT/nodeList variable. Do not use PCM or pconfig to apply this variable. The node list consists of one or more nodes separated by the @ character. Each node specification consists of the hostname, port, username, and password fields separated by an * asterisk.

Chapter 14

Monitoring Processes

277

Processes

host1*portnumber*username*password@host2*portnumber*username*password@host3*portnum ber*username*pasword@. The host1 value must be the local host which runs the Patrol Agent.

Deleting a Monitored Process from the Monitoring Queue

This task describes how to stop monitoring a process.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Process List group box, select the process you want to delete. 5 In the Actions group box, select Delete Selected Process, and click OK.
A confirmation message appears.

6 Click Yes to close the confirmation message.

The Manage List of Monitored Processes dialog box refreshes, and the process is removed from the list of monitored processes.

Specifying a Flag File Directory for Process Presence Blackout Periods

This task describes how to specify the directory that contains process presence flag files, and how to use those files as a resource external to the PATROL Agent for disabling the generation of state change alerts for the following parameters: PROCPPCountCheck, PROCPPOwnerCheck, PROCPParentPID1.

278

PATROL Knowledge Module for Unix User Guide

Processes

You can use flag files to create the equivalent of a PATROL blackout for specific process presence instances, suspending warnings and alarms during regular system maintenance. To create a blackout period for a process presence instance, move a flag file for the process presence instance into the designated flag file directory. When PATROL finds a flag file in the specified directory, it will ignore any alarms or warnings generated for the process.

Before You Begin

Before you identify the directory location, ensure that the following requirements are met for the flag file directory and the flag files. The flag file directory must conform to the following standards:
s s

the directory must exist on the system the PATROL Agent must have read permissions on the flag directory

Flag files must conform to the following standards:

s s

flag files must reside in the designated flag file directory flag files must be named ignore_ProcessName, where ProcessName is the name assigned to the monitored process

NOTE
Flag files do not expire after a specified time period. You must manually remove flag files from the specified flag file directory before PATROL will resume monitoring for a process presence instance.

To Specify a Flag File Directory 1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Identify Flag Directory from the pop-up menu. 4 Type the full path of the flag file directory in the Enter Full Path field. NOTE
Do not store flag files in the /tmp directory; many Unix systems automatically clear this directory upon system reboot.

5 Click Apply.

Chapter 14

Monitoring Processes

279

Processes

Suspending Process Monitoring for a Process Instance

This task describes how to suspend process monitoring for a process instance. Create a file ignore_ProcessName and copy it to the designated flag file directory. Flag files must be named ignore_ProcessName, where ProcessName is the monitored process name. When the flag file is found in the designated directory, PATROL ignores alerts and events that are generated for the monitored process.

NOTE
Flag files do not expire after a specified time period. You must manually remove flag files from the specified flag file directory before PATROL will resume monitoring for a process presence instance.

Resuming Process Monitoring for a Process Instance

This task describes how to resume monitoring of a process instance after process monitoring has been suspended. To start monitoring the process again, remove the ignore_ProcessName file from the flag file directory.

Reporting on a Selected Process

This task describes how to run a report on a selected process.

To Report on a Selected Process 1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage List of Monitored Processes from the pop-up menu.

The Manage List of Monitored Processes dialog box is displayed.

4 In the Process List group box, highlight the process for which you want to run a
report.

280

PATROL Knowledge Module for Unix User Guide

Processes

5 In the Actions group box, select the Report on Selected Process option button, and
click OK. The Report on Process Name dialog box is displayed. This dialog box displays the monitoring attributes that were set for the process. In addition, the dialog box displays the number of process instances running on the local host, and the total number of process instances running within the host group.

6 Click Close.
The Report on Process Name dialog box closes, and the Manage List of Monitored Processes dialog box is displayed.

Chapter 14

Monitoring Processes

281

Process Host Groups in a Cluster

Process Host Groups in a Cluster

The PROCESS application class monitors processes in a clustered environment. PATROL allows you to define each cluster as a process host group. This feature enables you to monitor and manage a specific process within a group. For information on how to define a process host group, see Rules for Creating Process Host Groups on page 282. The PROCESS tasks that relate to creating and managing Process Host Groups are described in detail on the following pages: Task Creating a Host Group Configuring Communication Attributes for a Process Host Group Modifying Host Attributes within a Process Group Removing a Host from a Host Group Querying a Selected Host within a Process Host Group Page 283 284 285 286 286

Synchronizing the Local Hosts List to a Selected Hosts List for a Process 288 Host Group Synchronizing a Selected Hosts List to the Local Host List for a Process Host Group Viewing the Process List for a Selected Host Synchronizing the Local Processs List with a Selected Process's List Synchronizing a Selected Process's List with the Local Process's List Viewing the Process Presence List for a Selected Process 288 289 289 290 291

Rules for Creating Process Host Groups

You must follow these rules when you create a process host group:
s

The local host from which you are creating the group must be the first host added to the group. In other words, the PATROL Agent for the local host computer must be the first PATROL Agent added to the group. Hosts may only be added to the group by a member of the host group. Any host within the group can add another member; you do not have to add all hosts from the original host. A host computer may only belong to one group at a time.

282

PATROL Knowledge Module for Unix User Guide

Process Host Groups in a Cluster

Creating a Host Group

This task describes how to create a process host group.

Before You Begin

You must comply with the rules for creating process host groups.

To Create a Process Host Group 1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage Host Group from the pop-up menu.

The Manage Host Group dialog box is displayed.

4 In the Actions group box, select the Add option button and click OK.
The Add Host to Watched Hosts dialog box is displayed.

5 In the Host Name field, type the name of the host on which the PATROL Agent is
installed and running.

6 In the Agent Port field, type the port number of the PATROL Agent that is installed
on the host.

7 In the User Name field, type the user name for the PATROL Agent account that is
installed on the host.

8 In the Password field, type the password for the PATROL Agent that is installed on
the host.

9 Click Add.
The dialog box closes, and the Manage Host Group dialog box is displayed. The new host is listed in the Host Group Members group box.

10 Click Close to close the Manage Host Group dialog box.

Chapter 14

Monitoring Processes

283

Process Host Groups in a Cluster

Configuring Communication Attributes for a Process Host Group

This task describes how to configure the communication attributes for a process host group.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Synchronize => Group List from the pop-up menu.

The Synchronize Host List dialog box is displayed.

4 In the Summary of Group Lists group box, highlight a host. 5 Click the Configure Group Communication Options for host option button, and click
OK.

The Group Process Monitor Configuration dialog box is displayed.

6 In the Timeout Group Communications After ___ Seconds field, type a value in the
field to specify the number of seconds that the local host will wait for a response from a remote host. The valid range for this field is 1-60 seconds. The default value is 10 seconds.

7 In the Check Response of Group Every ___ Intervals field, type a value in the field to
indicate the number of polling cycles that this host will allow to lapse between attempts to contact the PATROL Agent for each host within the group. The valid range for this field is 0-99 intervals. The default value for this field is 0, which means that the host is not performing the check.

NOTE
In larger host groups, this check can consume a significant amount of system resources. Do not turn this feature on unless you need to use it.

8 Select the Send Local Process Changes to Group option button to allow the PATROL
Agent to propagate process presence instance changes for the local host to hosts within the group that are set up to receive local host changes from other members of the group.

284

PATROL Knowledge Module for Unix User Guide

Process Host Groups in a Cluster

9 Select the Accept Local Process Changes from Group option button to allow the
PATROL Agent to receive process presence instance changes marked local from other hosts within the group that are set up to propagate changes to members of the host group.

10 Click OK.
The dialog box closes, and the communication setting for the group are changed.

Modifying Host Attributes within a Process Group

This task describes how to modify a host that is part of a process group.

Before You Begin

s

You can modify the following information about a host: Agent port number User name Password

You must comply with the rules for creating process host groups.

To Modify a Host within a Process Group 1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage Host Group from the pop-up menu.

The Manage Host Group dialog box is displayed.

4 In the Host Group Members group box, highlight the host you want to modify. 5 In the Actions group box, select the Modify Selected Host option button and click
OK.

The Modify Watched Hosts List dialog box is displayed.

6 In the Agent Port field, type the port number of the PATROL Agent that is installed
on the host.

Chapter 14

Monitoring Processes

285

Process Host Groups in a Cluster

7 In the User Name field, type the user name for the PATROL Agent account that is
installed on the host.

8 In the Password field, type the password for the PATROL Agent that is installed on
the host.

9 Click Modify.
The dialog box closes, and the Manage Host Group dialog box is displayed. Changes to the PATROL Agent port number and account user name are displayed in the Host Group Members group box. PATROL does not display the password for the PATROL Agent.

10 Click Close to close the Manage Host Group dialog box.

Removing a Host from a Host Group

This task describes how to delete a host from a process host group.

To Delete a Host from a Process Group 1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage Host Group from the pop-up menu.

The Manage Host Group dialog box is displayed.

4 In the Host Group Members group box, highlight the host you want to delete. 5 In the Actions group box, select the Delete Selected Host option button and click OK.
The dialog box refreshes, and the selected host is deleted from the Host Group
Members group box.

Querying a Selected Host within a Process Host Group

This task describes how to run a query on a selected host within a process group.

286

PATROL Knowledge Module for Unix User Guide

Process Host Groups in a Cluster

To Query a Host within a Process Host Group 1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Manage Host Group from the pop-up menu.

The Manage Host Group dialog box is displayed.

4 In the Host Group Members group box, highlight the host you want to query. 5 In the Actions group box, select the Query Selected Host option button and click OK.
The Status group box will display the following information:
s s s s s

local host name the last host that sent an update within the group the date and time that the last update was sent to the group whether or not the selected host responded when PATROL tried to query it the following diagnostic messages No sessiondisplays when the host was unable to open a session on the remote PATROL Agent Event faileddisplays when the session was opened on the remote PATROL Agent, but sending an event to the remote PATROL Agent failed No responsedisplays when the PATROL Agent timed out while waiting for a response to the event that was sent to the remote PATROL Agent Bad return datadisplays when the message that is received seems to be from a PATROL Agent other than the one with which the session was opened Unknown OS typedisplays when the PATROL Agent did not return a known OS type string Non-UNIX OSdisplays when the remote PATROL Agent is not running on a supported Unix operating system OKdisplays when the PATROL Agent responded correctly

Chapter 14

Monitoring Processes

287

Process Host Groups in a Cluster

Synchronizing the Local Hosts List to a Selected Hosts List for a Process Host Group
This task describes how to replace the local host list with the host list from the selected host. You can use this option to copy host group settings from the selected host to your local host.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Synchronize => Group List from the pop-up menu.

The Synchronize Host List dialog box is displayed.

4 In the Summary of Group Lists group box, highlight a host. 5 Select the Replace Local Host List with Selected option button and click OK.
The dialog box refreshes, and the local host list is replaced with the information from the selected host list.

Synchronizing a Selected Hosts List to the Local Host List for a Process Host Group
This task describes how to replace the selected host list with the host list from the local host. You can use this option to copy host group settings from the local host to the selected host.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Synchronize => Group List from the pop-up menu.

The Synchronize Host List dialog box is displayed.

4 In the Summary of Group Lists group box, highlight a host. 5 Select the Replace Selected Host List with Local option button and click OK.

288

PATROL Knowledge Module for Unix User Guide

Process Host Groups in a Cluster

The dialog box refreshes, and the selected host list is replaced with the information from the local host list.

Viewing the Process List for a Selected Host

This task describes how to view the list of hosts that are part of a selected host group.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Synchronize => Group List from the pop-up menu.

The Synchronize Host List dialog box is displayed.

4 In the Summary of Group Lists group box, highlight a host. 5 Select the View Host List from Selected Host option button and click OK.
The Process Presence Host List for Host dialog box is displayed.

6 Click Close.

Synchronizing the Local Processs List with a Selected Process's List

This task describes how to replace the process list for a local host with the process list on the selected host.

Before You Begin

If you want to receive changes that are marked as local and group, you must set up the local host to accept local process changes from members of the host group.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Synchronize => Process List from the pop-up menu.

Chapter 14

Monitoring Processes

289

Process Host Groups in a Cluster

The Synchronize Process List dialog box is displayed.

4 In the Summary of Process Lists group box, select a host name. 5 Click the Replace Local Process List with Selected option button and click OK.
The dialog box refreshes, and the process list for the local host is replaced by the process list for the selected host.

Synchronizing a Selected Process's List with the Local Process's List

This task describes how to replace the process list for a selected host with the process list on the selected host.

Before You Begin

If you want to receive changes that are marked as local and group, you must set up the selected host to accept local process changes from the group.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Synchronize => Process List from the pop-up menu.

The Synchronize Process List dialog box is displayed.

4 In the Summary of Process Lists group box, select a host name. 5 Click the Replace Selected Process List with Local option button and click OK.
The dialog box refreshes, and the process list for the local host is replaced by the process list for the selected host.

290

PATROL Knowledge Module for Unix User Guide

Process Host Groups in a Cluster

Viewing the Process Presence List for a Selected Process

This task describes how to view the list of processes that are being monitored on a host group.

1 Double-click the PROCESS application container. 2 Access the PROCESS_PRESENCE application menu as described in Accessing
KM Commands and InfoBoxes on page 402.

3 Select Synchronize => Process List from the pop-up menu.

The Synchronize Host List dialog box is displayed.

4 In the Summary of Process Lists group box, highlight a host. 5 Select the View Process List from Selected Host option button and click OK.
The Process Presence Process List for Host dialog box is displayed.

6 Click Close.

Chapter 14

Monitoring Processes

291

Status

Status
The PROCESS application class enables you to monitor processes based on their status. This feature allows you to view which and how many processes are running on the system. It also enables you to view processes based on characteristics such as state and CPU usage. The PROCESS tasks related to status are described in detail on the following pages: Task Viewing the Number of Zombie Processes Viewing Zombie Processes Viewing the Total Number of Processes Running Viewing Processes that Use Excessive CPU Viewing the Number of Processes Waiting on Resources Viewing the Total Number of User Processes Viewing the Average Number of User Processes Page 292 293 295 295 298 298 299

Viewing the Number of Zombie Processes

This task describes how to view the total number of processes in Zombie state on the system at the present time or in the last couple of hours. PATROL identifies a zombie process as an ended process whose entry remains in the process table but is not assigned to a user or kernel space and does not consume any other resources. Processes in this state are also referred to as defunct processes.

To View the Number of Processes Now in Zombie Mode

Access the PROCESS application InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the PROCESS InfoBox (Figure 49 on page 293), which provides the Number of processes in a Zombie state.

292

PATROL Knowledge Module for Unix User Guide

Status

Figure 49

PROCESS InfoBox

To View the Number of Processes Recently in Zombie Mode 1 Access the PROCESS application class so that you can view its parameters. 2 Open the PROCNoZombies parameter.
PATROL displays a graph that shows the number of Zombie processes that occurred over the last couple of hours.

Viewing Zombie Processes

This task displays information about the zombie processes that persist on the system. PATROL identifies a zombie process as an ended process whose entry remains in the process table but is not assigned to a user or kernel space and does not consume any other resources. Processes in this state are also referred to as defunct processes.

To View Zombie Processes 1 Access the PROCESS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select List Zombie Processes.

PATROL writes the information to a PATROL task object, List Zombie Processes, in the UNIX OS container.

Chapter 14

Monitoring Processes

293

Status

3 Access the List Zombie Processes task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example: # PSL Serial No.: ######### <NOT INTERACTIVE> List zombie process 6933 6932 3Supvr Z 0 0 <defunct> ??? ---------------- >>>>> DONE <<<<< ----------------

Output Format
The output has the following format. Table 47 defines each field. pidppiduserssizecputcmd namecmd line

Table 47 Field pid ppid user s

List Zombie Process Command Output Format Description the process ID of the process the process ID of the parent process the name of the user that owns the process the status of the process Iintermediate 0nonexistent Rrunning Ssleeping Tstopped Wwaiting Xgrowing Zzombie, terminated

size cmd name cmd line

the total size of the process in virtual memory, in blocks the first word of the command line that started the process the complete command line (command and arguments up to 80 characters) that started the process; note that the command line can be modified during the process execution

294

PATROL Knowledge Module for Unix User Guide

Status

Viewing the Total Number of Processes Running

This task describes how to view the total number of processes running on the system at the present time or in the last couple of hours.

To View the Total Number of Processes Now Running

Access the PROCESS application InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the PROCESS InfoBox, which provides the total number of processes running on the system at the present time.

To View the Total Number of Processes Recently Running 1 Access the PROCESS application class so that you can view its parameters. 2 Open the PROCNumProcs parameter.
PATROL displays a graph that shows the total number of processes running on the system over the last couple of hours.

Viewing Processes that Use Excessive CPU

This task describes how to view the processes that use the most CPU. These processes are often referred to as CPU hogs. PATROL provides two methods to obtain this information.

Before You Begin

PATROL allows you to discover these CPU hogs by using a menu command or a parameter.
s

The menu command allows you to define excessive CPU utilization through a dialog box when you execute the command. The parameter compares each processes CPU utilization against the other processes and lists the processes that consume the most CPU. The cputhreshold value has a default value of 80 percent.

Chapter 14

Monitoring Processes

295

Status

To View the Processes that Use Excessive CPU as Defined by You 1 Access the PROCESS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select CPU Hogs.

PATROL displays the CPU Hogs dialog box (Figure 50). Figure 50 CPU Hogs Dialog Box

3 Select the threshold by moving the slider bar to the right (maximum value of 100)
or the left (minimum value of 0) and click Accept. PATROL surveys the system for 30 seconds and then writes the information to a PATROL task object, CPU Hogs Process, in the UNIX OS container.

4 Access the CPU Hogs Process task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example: # PSL Serial No.: ######### <NOT INTERACTIVE> CPU Hogs collecting data for apprx. 30 seconds... Processes consuming at least 50% of CPU during this interval (30 sec.): PID CPU% USER CMD 29111 73 3Suprv audit_script -I i18nInstance -M CPU Hogs complete ---------------- >>>>> DONE <<<<< ----------------

296

PATROL Knowledge Module for Unix User Guide

Status

Output Format
The output has the following format. Table 48 defines each field. pid cpu% user cmd line

Table 48 Field pid cpu% user cmd line

CPU Hogs Command Output Format Description the process ID of the process the percentage of CPU that the process consumed during the survey interval the name of the user that owns the process the complete command line (command and arguments up to 80 characters) that started the process; note that the command line can be modified during the process execution

To View the Processes that Consume the Most CPU

This parameter displays the 10 processes that consume the most CPU.

1 Access the PROCESS application class so that you can view its parameters. 2 Open the PROCTopProcs parameter.
PATROL displays the information in a text parameter. Your result should resemble this example: # PSL Serial No.: ######### %TIME PID PPID TOT SECS 0.00 0 0 18 0.00 1 0 1 0.00 2 0 0 0.00 3 0 588 0.00 315 1 0

SIZE 0 96 0 0 217

UID COMMAND root sched root /etc/init root pageout root fsflush root /usr/lib/saf/sac

Last update: Thu Jun 14 12:50:27 CCYY Sample Period: 30 secs.

Chapter 14

Monitoring Processes

297

Status

Text Parameter Format

The output has the following format. Table 49 defines each field. %time pid ppid tot secs size uid command

Table 49 Field %time pid ppid tot secs size

PROCTopProcs Text Parameter Format Description the percentage of cpu time that the process consumes the process ID of the process the process ID of the parent process the total number of seconds the process consumed CPU resources the total size of the process in virtual memory, in blocks

uid command the complete command line (command and arguments up to 80 characters) that started the process; note that the command line can be modified during the process execution

Viewing the Number of Processes Waiting on Resources

This task describes how to view the total number of processes on the system waiting for resources.

To View the Number of Processes Waiting for Resources 1 Access the PROCESS application class so that you can view its parameters. 2 Open the PROCProcWait parameter.
PATROL displays a graph that shows the number of processes that are currently waiting for resources and have been over the last couple of hours.

Viewing the Total Number of User Processes

This task describes how to view the total number of user (non-root) processes running on the system.

298

PATROL Knowledge Module for Unix User Guide

Priority

To View the Total Number of User (non-root) Processes Running on the System 1 Access the PROCESS application class so that you can view its parameters. 2 Open the PROCUserProcs parameter.
PATROL displays a graph that shows the number of user (non-root) processes that are running on the system.

Viewing the Average Number of User Processes

This task describes how to view the average number of user (non-root) processes running on the system.

To View the Average Number of Processes Per User 1 Access the PROCESS application class so that you can view its parameters. 2 Open the PROCAvgUsrProcs parameter.
PATROL displays a graph that shows the average number of processes running on the system per user.

Priority
The PROCESS application class enables you to manage process execution priority based on the individual process, the owner of the process, and the group to which the process belongs. From within PATROL, it enables you to decrease a processs priority. The PROCESS tasks are described in detail on the following pages: Task Reprioritizing a Process Reprioritizing All Processes in a Process Group Reprioritizing All Processes Owned by a User Page 300 301 302

Chapter 14

Monitoring Processes

299

Priority

Reprioritizing a Process
This task describes how to reprioritize an individual process. Reprioritizing a processs execution is commonly referred to as renicing. Priorities, or nice values, range from -20 (highest) to 20 (lowest: process will run when no other process attempts to run). To lower a processs priority, you must raise its nice value. To raise a processs priority you must lower its nice value.

Before You Begin

s

At a Unix command prompt, use the Unix list active processes command with the long option (ps -l) to find the process ID and current priority of the process whose execution priority you want to change. The execution priority, also referred to as the nice value, is stored in the NI column. Only privilege users can decrease the nice value. Other users can only increase the nice value within the range 0 to 20.

To Reprioritize a Process 1 Access the PROCESS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Renice Process.

PATROL displays the Renice Process dialog box (Figure 51). Figure 51 Renice Process Dialog Box

3 Type the Process ID number in the Enter Process ID field. 4 Type an integer value for the amount you want to lower or raise the priority in the
Enter Priority field of the dialog box. Negative values are valid. The larger the number that you enter is, the lower the resulting priority of the process; the smaller the number is, the higher the priority.

EXAMPLE
To raise the priority from 8 to 6, type a value of -2 in the Enter Priority field.

300

PATROL Knowledge Module for Unix User Guide

Priority

5 Click Apply.
PATROL prompts you for a user name and password (Figure 52). Figure 52 Renice Process Confirmation Dialog Box

6 Enter the user name and password of the owner of the process or an account with
privileges to reprioritize it, and click OK. PATROL writes a confirmation message to the system output window. Your result should resemble this example: CCYY0614153103 PROCESS: Renice process 333 to priority: 6

Reprioritizing All Processes in a Process Group

This task describes how to reprioritize all processes that belong to a process group. Reprioritizing a processs execution is commonly referred to as renicing. Priorities, or nice values, range from -20 (highest) to 20 (lowest: process will run when no other process attempts to run). To lower a process groups priority, you must raise its nice value. To raise a process groups priority you must lower its nice value.

Before You Begin

s

At a Unix command prompt, use the Unix list active processes command with the long j option (ps -lj) to find the group process ID and current priorities of the processes that belong to the group. The group ID is stored in the PGID column. The execution priority, also referred to as the nice value, is stored in the NI column. Only privileged users can decrease the nice value. Other users can only increase the nice value within the range 0 to 20.

Chapter 14

Monitoring Processes

301

Priority

To Reprioritize a Process 1 Access the PROCESS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Renice Group.

PATROL displays the Renice Group dialog box.

3 Type the Group ID number in the Enter Group ID field. 4 Type an integer value for the amount you want to lower or raise the priority in the
Enter Priority field of the dialog box. Negative values are valid. The larger the number that you enter is, the lower the resulting priority of the process; the smaller the number is, the higher the priority.

EXAMPLE
To lower the priority from 8 to 10, type a value of 2 in the Enter Priority field.

5 Click Apply.
PATROL prompts you for a user name and password.

6 Enter the user name and password of the owner of the group or an account with
privileges to reprioritize it, and click OK. PATROL writes a confirmation message to the system output window. Your result should resemble this example: CCYY0614153103 PROCESS: Renice group suprvizr to priority: 10

Reprioritizing All Processes Owned by a User

This task describes how to reprioritize all processes that belong to a specific user. This task only decreases the execution priority of these processes. Reprioritizing a processs execution is commonly referred to as renicing. Priorities, or nice values, range from -20 (highest) to 20 (lowest: process will run when no other process attempts to run). To lower a users process priority, you must raise its nice value. To raise a users process priority you must lower its nice value.

302

PATROL Knowledge Module for Unix User Guide

Priority

Before You Begin

s

At a Unix command prompt, use the Unix list active processes command with the long option (ps -l) to find the processes that belong to a specific user, indicated by user ID, and the current priorities of the processes that belong to the user. The user ID is stored in the UID column. The execution priority, also referred to as the nice value, is stored in the NI column. Only privileged users can decrease the nice value. Other users can only increase the nice value within the range 0 to 20.

To Reprioritize a Process 1 Access the PROCESS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Renice User.

PATROL displays the Renice User dialog box.

3 Type the user ID number in the Enter User ID field. 4 Type an integer value for the amount you want to lower or raise the priority in the
Enter Priority field of the dialog box. Negative values are valid. The larger the number that you enter is, the lower the resulting priority of the process; the smaller the number is, the higher the priority.

EXAMPLE
To lower the priority from 10 to 15, type a value of 5 in the Enter Priority field.

5 Click Apply.
PATROL prompts you for a user name and password.

6 Enter the user name and password of the user or an account with privileges to
reprioritize the process owned by the user, and click OK. PATROL writes a confirmation message to the system output window. Your result should resemble this example: CCYY0614153103 PROCESS: Renice user 3Supvr to priority: 15

Chapter 14

Monitoring Processes

303

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux
The PATROL Configuration Manager includes the Process KM Configuration Plug-in that allows you to configure the Process KM to monitor specific processes. The plug-in provides three tabs. Each tab enables you to specify certain monitoring criteria.
s

The Process tab designates a process or processes to be monitored by the PATROL Knowledge Module for Unix. The Start/Restart Options tab controls the number of instances of a process that is being monitored. You can configure the knowledge module to automatically stop an instance of a monitored process if it exceeds the maximum process count or automatically start a process if the number of process instances falls below the minimum count. The Alert and Monitor Options tab specifies what type of alert is generated by the knowledge module and the time frame in which the alert is generated when the monitored process falls below the minimum count or rises above the maximum.

Plug-in Actions
The plug-ins interact with PATROL Configuration Manager based upon the task that you are performing. When adding rulesets to establish and configure monitoring, you use the Agent tree view pane. When updating or deleting rulesets, you use the RuleSets tree view pane.

Adding Object Instances to Monitor

Configuring PATROL to monitor some aspect of an application involves creating an instance for the object being monitored such as a process and establishing a ruleset. This procedure is initiated in the Agent tree view pane.

To Add an Object Instance 1 In the Agent tree view pane, select an agent by clicking on it. 2 Right-click the agent and select the plug-in and its option such as either Process KM
Configuration => Processes.

The configuration manager displays a configuration dialog box.

304

PATROL Knowledge Module for Unix User Guide

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux

3 Click Add.
The plug-in displays a dialog box for creating instances.

4 Type the required information in the appropriate fields, which are described in
Table 50. Table 50
Fields Process Instance New Instance Label Monitored Process Type the name of the process as you want it to appear in PATROL Type the process command of the process that you want to monitor

New Instance Dialog Boxes and Fields

Description

5 Click OK. 6 Select each tab and provide the relevant information in the configuration dialog
box. For more information about the Process KM Configuration dialog box, see Configuring the Process KM on page 307.

7 Click either Apply to save your changes and leave the dialog box open or OK to
save the changes and close the dialog box.

Updating Monitored Object Instances

Changing how PATROL monitors some aspects of an application involves editing a ruleset. This procedure is initiated in the RuleSet tree view pane.

1 In the RuleSet tree view pane, expand the RuleSet folder. 2 Navigate to the backup ruleset of the agent whose object instance monitoring you
want to update. Expand PCM => backup => backup_container => agent => ccyymmdd-hhmmss_ruleset.

3 Right-click the ruleset and select the plug-in and its option such as either Process
KM Configuration => Processes.

NOTE
If no object instances have been added for monitoring, the menu command will be inactive. For more information about adding an object instance for monitoring, see Adding Object Instances to Monitor on page 304.

Chapter 14

Monitoring Processes

305

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux

The configuration manager displays a configuration dialog box.

4 Select one instance from the instance list and click Update. 5 Select the appropriate tab and edit the relevant information in the configuration
dialog box. For more information about the Process KM Configuration dialog box, see Configuring the Process KM on page 307.

6 Click either Apply to save your changes and leave the dialog box open or OK to
save the changes and close the dialog box.

Deleting Monitored Object Instances

Stopping PATROL from monitoring some aspects of an application involves deleting a ruleset. This procedure is initiated in the RuleSet tree view pane.

1 In the RuleSet tree view pane, expand the RuleSet folder. 2 Navigate to the backup ruleset of the agent whose object instance monitoring that
you want to update. Expand PCM => backup => backup_container => agent => ccyymmdd-hhmmss_ruleset.

3 Right-click the ruleset and select the plug-in and its option such as either Process
KM Configuration => Processes.

NOTE
If no object instances have been added for monitoring, the menu command will be inactive. For more information about adding an object instance for monitoring, see Adding Object Instances to Monitor on page 304.

The configuration manager displays a configuration dialog box.

4 Select one instance from the instance list. 5 Click Delete. 6 Click either Apply to save the deletion and leave the dialog box open or OK to save
the deletion and close the dialog box.

306

PATROL Knowledge Module for Unix User Guide

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux

Configuring the Process KM

Before You Begin
You must have setup a PATROL object to monitor a process.

To Configure Process Monitoring from PATROL Configuration Manager 1 Add a new PATROL object instance to monitor a process as described in Adding
Object Instances to Monitor on page 304 or select a ruleset created to monitor a process as described in Updating Monitored Object Instances on page 305.

2 Select an Instance from the Process Instance List. 3 Select the Process tab and specify how many process instances should run
simultaneously and who can own the process. Table 51 describes the process properties used to define the criteria. Table 51 Field
Process Instance List

Process tab properties (Part 1 of 2)

Description Lists the processes being monitored; provides both the process name and the user-defined label by which PATROL applications refer to it Type a name for the monitored process instance icon in this field. Type the command or unique command string for the process that is being monitored

Process Instance Label Monitored Process String

PATROL monitors all processes that match the string you type in this field. When you enter text in this field, you may end up monitoring multiple processes. For example, if you type vi in this field, PATROL will monitor processes named vi, view, and previous.
Minimum Count Use the spinner buttons, or type a value in this field to set the minimum number of process instances that may be running on the local computer or in the host group. Use the spinner buttons, or type a value in this field to set the maximum number of process instances that may be running on the local computer or in the host group. The value in the Maximum Count field must be equal to or greater than the value in the Minimum Count field.

Maximum Count

Chapter 14

Monitoring Processes

307

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux

Table 51 Field

Process tab properties (Part 2 of 2)

Description Type the user IDs of all acceptable process owners in this field; separate multiple user IDs with spaces. From the drop list, select one of the following options: Yes - indicates that the parent process ID (PPID) for the process must be 1. (Processes with a PPID of 1 are owned by init, or the Unix scheduler). No indicates that the parent process ID can be a number other than 1.

Acceptable Process Owners Parent Process ID Must Be 1

4 Select the Start/Restart Options tab and specify if a process should be restarted
automatically, commands for starting and stopping a process, and credentials for accounts to start and stop passwords. Table 52 describes the process properties used to define the criteria. Table 52
Field Restart Automatically

Start/Restart tab properties (Part 1 of 2)

Description From the drop list, select one of the following options: Yesindicates that the process will be restarted automatically if PATROL detects it has stopped. If you select this option, the PATROL KM for Unix will attempt to restart the process when it detects that the process instance count has fallen below the specified minimum threshold. The KM uses the value in the Command Execution Attempts field to determine how many times it will try to restart a process. Note: To restart a process automatically, you must provide a start command and a command execution account user name and password in the appropriate fields on this dialog box. Noindicates that the process must be restarted manually if it stops.

Command Execution Attempts

Type a value in this field to set the number of times the host will attempt to run a Start Process or Stop Process command before it stops trying to run the command. The value that you enter in this field must be a 1 or greater.

Start Command

Type the command string that will start the process instance. You must specify a command execution user account and password if you want to use the Start command.

308

PATROL Knowledge Module for Unix User Guide

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux

Table 52
Field

Start/Restart tab properties (Part 2 of 2)

Description Type the command string that will stop the process instance. You must specify a command execution user account and password if you want to use the Stop command.

Stop Command

Command Execution User Name

Type the user ID under which the command will be executed. You must specify a command execution user account and password if you want to use the Start and Stop commands.

Command Execution Password

Type the password for the user ID under which the command will be executed. You can leave this field blank; the KM will accept a null password. You must specify a command execution user account and password if you want to use the Start and Stop commands.

5 Select the Alert and Monitor Options tab and specify the location of the processes
running and the conditions under which an alert is issued. Table 53 describes the process properties used to define the criteria. Table 53
Field Alert Delay Count

Alert and Monitor (Part 1 of 2)

Description Select a value in this field to set the number of collection intervals that this computer will defer an alert while it waits for the process count to be reestablished across the host or group. Delaying the initial alert allows the system to correct a situation without generating an event to notify you that there is a problem.

Alert State

From the drop list select a state (ALARM or WARNING) change for the following parameters: PROCPPCountCheckPROCPPCountCheck_parameter PROCPPOwnerCheckPROCPPOwnerCheck_parameter PROCPPParentPID1PROCPPParentPID1_parameter PATROL sets these parameter to the selected state when the Action Deferral Count reaches zero and the process count across the range is out of range.

Chapter 14

Monitoring Processes

309

Using the PATROL Configuration Manager to Configure the PATROL KM for Unix and Linux

Table 53
Field Scope

Alert and Monitor (Part 2 of 2)

Description From the drop list, select one of the following options to specify whether the process should be monitored on one computer or across a computer host group: Only on this Hostindicates that PATROL should only monitor the process on the local host computer Across Host Groupindicates that PATROL should monitor the process across the entire host group

Blackout Period

You have the option of setting a blackout period during which you do not want to monitor the selected process. To set the blackout period: In the Start drop boxes, enter the time (in 24-hour clock format) to begin the blackout period in hours (first drop list) and minutes (second drop list). In the End drop boxes, enter the time that you want the blackout period to end in hours (first drop list) and minutes (second drop list). For example, if you wanted the blackout period to occur between 11:00 PM and 5:00 AM, in the Start fields you would enter 23 and 00. In the End fields you would enter 05 and 00.

6 Click Apply to save the settings. 7 Click OK.

PATROL Configuration Manager saves your changes and closes the dialog box.

8 Apply the configuration changes.

s

If you are adding a new object instance for monitoring (working in the Agent tree view pane), click the Apply button to apply the ruleset and begin monitoring. If you are updating an object instance for monitoring, (working in the RuleSet tree view pane) assign the updated rulesets to the desired agent(s), and then click the Apply button to apply the ruleset and begin monitoring with the new settings.

310

PATROL Knowledge Module for Unix User Guide

Chapter

15
312 312 313 313 315 318 321 321 322 326 326 327 330 331 334 335 337

15

Scheduling Tasks
The PATROL Knowledge Module for Unix and Linux gives you the ability to schedule processes and to schedule monitoring blackout periods in your environment.The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Scheduling a Task for Execution via the Operating System Command Line Processor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Scheduling a Task for Execution Using a PATROL Command Processor . . . . . Scheduling a PSL Task for Execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modifying an Existing Scheduled Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Removing a Scheduled Task from the Queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Collecting Debug Data for the Scheduler Application . . . . . . . . . . . . . . . . . . . . . Blackouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Specifying a Default Directory for Blackout Scripts. . . . . . . . . . . . . . . . . . . . . . . . Creating a Monitoring Blackout Script. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Information about a Blackout Script. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modifying an Existing Blackout Script. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting a Blackout Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Scheduling a Blackout Script for Execution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Canceling a Scheduled Blackout Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 15

Scheduling Tasks

311

Introduction

Introduction
The PATROL Knowledge Module for Unix and Linux schedules processes and establishes monitoring blackout periods in your environment through the SCHEDULER application.

NOTE
You can change settings in the PATROL SCHEDULER application class using the PATROL Operator Console. Set up access control lists (ACLs) to prevent users from making changes to SCHEDULER settings through a PATROL Operator Console. With this application class, you can
s

write a script to perform a task or job and then schedule that task to run at once or at regular intervals suspend monitoring of applications during routine maintenance or other activities that generate false alarms

Object Hierarchy
The SCHEDULER application class contains SCHEDULER parameters. You can use the SCHEDULER application class to schedule processes and programs to run at specified times. In addition, you can use the SCHEDULER application class to suspend PATROL event generation during specified time periods. Figure 53 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 53 SCHEDULER Application Object Hierarchy
OS container

Application class

Parameter

312

PATROL Knowledge Module for Unix User Guide

Tasks

Tasks
The SCHEDULER application class gives you the ability to write a script to perform a task or job and then schedule that task to run at regular intervals. This feature enables you to designate which script runs when. The SCHEDULER tasks that relate to scheduling scripts are described in detail on the following pages: Task Scheduling a Task for Execution via the Operating System Command Line Processor Scheduling a PSL Task for Execution Modifying an Existing Scheduled Script Removing a Scheduled Task from the Queue Collecting Debug Data for the Scheduler Application Page 313

Scheduling a Task for Execution Using a PATROL Command Processor 315 318 321 321 322

Scheduling a Task for Execution via the Operating System Command Line Processor
This task describes how to schedule a script for execution via the operating system command line.

Before You Begin

Ensure that the following requirements are met:
s

The PATROL Agent must have read and execute permissions for the script that is being scheduled for execution. The PATROL Agent must have write permissions for the output file, if you are creating a script with an output file.

To Schedule a Task for Execution Using the Command Line Processor 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Tasks from the pop-up menu.

Chapter 15

Scheduling Tasks

313

Tasks

The Schedule Tasks dialog box is displayed.

3 In the Actions group box, select Schedule New Task and click Accept.
The Schedule New Task dialog box is displayed.

4 Select the Schedule an Operating System Script option button and click Accept.
The Schedule an Operating Script dialog box is displayed.

5 In the Execute Script at: group box, use the spinner buttons or type a value in the
following fields to schedule the script: Field Month Day Year hh Description indicate the month in which the script should execute indicate the date on which the script should execute indicate the year in which the script should execute indicate at which hour of the day the script should execute Enter hours in military time. For example, if you want to schedule a script to execute at 4 P.M., enter 16 in this field. mm ss indicate at which minute the script should execute indicate at which second the script should execute

6 If you want to set the script to occur at regularly scheduled intervals, use the
spinner buttons or type values in the following fields of the Repeat Execution: group box: Field Days hh mm ss Description indicate the number of days that should pass between script executions indicate the number of hours that should pass between script executions indicate the number of minutes that should pass between script executions indicate the number of seconds that should pass between script executions

7 Select the Execute Once Immediately option button if you want the script to execute
now.

314

PATROL Knowledge Module for Unix User Guide

Tasks

NOTE
If you do not select this option and you are scheduling a script that was supposed to run in the past, it will not run again until the repeated execution runtime. For example, if you have a backup script that is supposed to run every Sunday evening, but you fill out this dialog box on a Monday morning, PATROL will not run the script until the next scheduled execution. You must toggle the Execute Once Immediately button on to run the script now.

8 In the Script File Name field, type the full path and file name for the script. NOTE
The PATROL Agent account must have read and execute permissions for the script. The PATROL Agent account must also have read permissions for the directory in which the script resides.

9 In the Execution Instance field, type the name of the PATROL instance against
which the command should be executed. This field is optional.

10 In the Write Output To field, type the full path and file name for the output file. This
field is optional.

NOTE
If you specify an output file name in this field, the PATROL Agent must have write permissions for the output file.

11 Click Accept.
The dialog box closes, and PATROL schedules the script for execution.

Scheduling a Task for Execution Using a PATROL Command Processor

This task describes how to schedule a task for execution using a command interpreter. You can schedule an existing script for execution, or schedule a script that does not yet exist for execution.

Before You Begin

Ensure that the following requirements are met:
s

The PATROL Agent must have read and execute permissions for the script that is being scheduled for execution.

Chapter 15

Scheduling Tasks

315

Tasks

The PATROL Agent must have write permissions for the output file, if you are creating a script with an output file.

WARNING
If you schedule a script with incorrect information (for example, a non-existent script, or one that resides in a directory that is different from the one you specified) for execution, PATROL will attempt to execute it at the scheduled time. However, the execution will fail, and the script will not run.

To Schedule a Task for Execution Via an Interpreter 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Tasks from the pop-up menu.

The Schedule Tasks dialog box is displayed.

3 In the Actions group box, select Schedule New Task and click Accept.
The Schedule New Task dialog box is displayed.

4 Select the Schedule a PATROL Command Processor Script option button.

The Schedule a PATROL Command Processor Script dialog box is displayed.

5 In the Execute Script at: group box, use the spinner buttons or type a value in the
following fields to schedule the blackout script: Field Month Day Year hh Description indicate the month in which the script should execute indicate the date on which the script should execute indicate the year in which the script should execute indicate at which hour of the day the script should execute Enter hours in military time. For example, if you want to schedule a script to execute at 4 P.M., enter 16 in this field. mm ss indicate at which minute the script should execute indicate at which second the script should execute

6 If you want to set the script to occur at regularly scheduled intervals, use the
spinner buttons or type values in the following fields of the Repeat Execution: group box:

316

PATROL Knowledge Module for Unix User Guide

Tasks

Field Days hh mm ss

Description indicate the number of days that should pass between script executions indicate the number of hours that should pass between script executions indicate the number of minutes that should pass between script executions indicate the number of seconds that should pass between script executions

7 Select the Execute Once Immediately option button if you want the script to execute
now.

NOTE
If you do not select this option and you are scheduling a script that was supposed to run in the past, it will not run again until the repeated execution runtime. For example, if you have a backup script that is supposed to run every Sunday evening, but you fill out this dialog box on a Monday morning, PATROL will not run the script until the next scheduled execution. You must toggle the Execute Once Immediately button on to run the script now.

8 In the PATROL Command Processor field, type the name of the program that should
be used to interpret the script you have written. The command processor must be one that is defined to the PATROL Agent on which you are running the script before you can use it. For more information on defining command interpreters to the PATROL Agent, look up the PSL built-in function named execute() in the PATROL Script Language Reference Manual Volume 2PSL Functions.

9 In the Script File Name field, type the full path and file name for the script. NOTE
The PATROL Agent account must have read and execute permissions for the script. The PATROL Agent account must also have read permissions for the directory in which the script resides.

10 In the Execution Instance field, type the name of the PATROL instance against
which the command should be executed. This field is optional.

11 In the Write Output To field, type the full path and file name for the output file. This
field is optional.

Chapter 15

Scheduling Tasks

317

Tasks

NOTE
If you specify an output file name in this field, the PATROL Agent must have write permissions for the output file.

12 Click Accept. NOTE

If you entered an incorrect full path and file name for the script, the PATROL Scheduler File Name dialog box is displayed. Use the following procedure to correctly fill out the dialog box. In the Script File Name field, type the full name and file path for the script and click Accept. If you entered the correct full path and file name correctly, the script is scheduled for execution.

Scheduling a PSL Task for Execution

This task describes how to schedule a script for execution via PATROL PSL. You can schedule an existing script for execution, or schedule a script that does not yet exist for execution.

Before You Begin

Ensure that the following requirements are met:
s

The PATROL Agent must have read and execute permissions for the script that is being scheduled for execution. The PATROL Agent must have write permissions for all output files, if you are creating a script with output files.

WARNING
If you schedule a script with incorrect information (for example, a non-existent script, or one that resides in a directory that is different from the one you specified) for execution, PATROL will attempt to execute it at the scheduled time. However, the execution will fail, and the script will not run.

318

PATROL Knowledge Module for Unix User Guide

Tasks

To Schedule a PSL Script for Execution 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Tasks from the pop-up menu.

The Schedule Tasks dialog box is displayed.

3 In the Actions group box, select Schedule New Task and click Accept.
The Schedule New Task dialog box is displayed.

4 Select the Schedule a PATROL PSL Script option button and click Accept. 5 In the Execute Script at group box, use the spinner buttons or type a value in the
following fields to schedule the script: Field Month Day Year hh Description indicate the month in which the script should execute indicate the date on which the script should execute indicate the year in which the script should execute indicate at which hour of the day the script should execute Enter hours in military time. For example, if you want to schedule a script to execute at 4 P.M., enter 16 in this field. mm ss indicate at which minute the script should execute indicate at which second the script should execute

6 If you want to set the script to occur at regularly scheduled intervals, use the
spinner buttons or type values in the following fields of the Repeat Execution group box: Field Days hh mm ss Description indicate the number of days that should pass between script executions indicate the number of hours that should pass between script executions indicate the number of minutes that should pass between script executions indicate the number of seconds that should pass between script executions

7 Select the Execute Once Immediately option button if you want the script to execute
now.

Chapter 15

Scheduling Tasks

319

Tasks

NOTE
If you do not select this option and you are scheduling a script that was supposed to run in the past, it will not run again until the repeated execution runtime. For example, if you have a backup script that is supposed to run every Sunday evening, but you fill out this dialog box on a Monday morning, PATROL will not run the script until the next scheduled execution. You must toggle the Execute Once Immediately button on to run the script now.

8 In the Internal PSL Process Name field, type the name of the PSL process that will be
run with the script. For more information on PSL process names, look up the PslEexecute() function in the PATROL Script Language Reference Manual Volume 2PSL Functions.

9 In the Script File Name field, type the full path and file name for the script. NOTE
If you enter an incorrect file name and path, the PATROL Scheduler File Name dialog box will display when you click Accept. This dialog box prompts you to correct the file name and path, or schedule the task as-is. If you schedule a script with incorrect information (for example, a non-existent script, or one that resides in a directory that is different from the one you specified) for execution, PATROL will attempt to execute it at the scheduled time. However, the execution will fail, and the script will not run.

10 In the Write Compile Errors To field, type the full path and file name for the error
output file. This field is optional.

NOTE
If you specify an error output file name in this field, the PATROL Agent must have write permissions for the output file.

11 In the Write Compile Warnings To field, type the full path and file name for the
warning output file. This field is optional.

NOTE
If you specify a warning output file name in this field, the PATROL Agent must have write permissions for the output file.

12 Click Accept.
The dialog box closes, and PATROL schedules the script for execution.

320

PATROL Knowledge Module for Unix User Guide

Tasks

Modifying an Existing Scheduled Script

This task describes how to modify an existing script for execution.

Before You Begin

Ensure that the following requirements are met:
s

The PATROL Agent must have read and execute permissions for the script that is being scheduled for execution. The PATROL Agent must have write permissions for all output files, if you are creating a script with output files.

To Modify an Existing Task 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Tasks from the pop-up menu.

The Schedule Tasks dialog box is displayed.

3 In the Scheduled Tasks group box, highlight the task you want to modify. 4 In the Actions group box, select Modify Selected Task and click OK.
The Select Function to Execute Script dialog box is displayed.

5 Finish the task by performing one of the following actions:

To Modify and Schedule this Type of Script Follow this Procedure operating system PATROL PSL PATROL Command Processor Scheduling a Task for Execution via the Operating System Command Line Processor, step 5 through step 11 Scheduling a PSL Task for Execution, step 5 through step 12 Scheduling a Task for Execution Using a PATROL Command Processor, step 5 through step 12 Page 313 318 315

Removing a Scheduled Task from the Queue

This topic provides information about removing a scheduled task from the queue.

Chapter 15

Scheduling Tasks

321

Tasks

To Remove a Scheduled Task from the Queue 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Tasks from the pop-up menu.

The Schedule Tasks dialog box is displayed.

3 In the Scheduled Tasks group box, highlight the task you want to modify. 4 In the Actions group box, select Remove Selected Task and click OK. 5 Click OK.
PATROL removes the scheduled task from the queue and refreshes the dialog box.

Collecting Debug Data for the Scheduler Application

This topic provides information about scheduling PSL debugger data collection. You can use this feature to turn PSL debugger data collection on and off for different parts of the Scheduler application. For more information on PSL debugger data, see the PATROL Script Language Reference Manual Volume I-PSL Essentials.

To Collect PSL Debugger Data for the Scheduler Application 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Scheduling Debug from the pop-up menu.

The PATROL Scheduler PSL Debugger dialog box is displayed.

3 In the Scheduling Debug dialog box, select the check box next to each selection that
you want to turn on. Debug data will be collected for that option.

4 For the Scheduler Daemon Debugger group box item, select the check box next to
each selection that you want to turn on for run-time checks and traces. When a debugging selection is turned on, debug data is collected for that option.

322

PATROL Knowledge Module for Unix User Guide

Tasks

Check Box Run-Time Checks Arithmetic operations with NULL strings Arithmetic operations with non-numeric operands

Description checks to ensure that numeric variables are explicitly initialized to 0, instead of defaulting to a NULL string. checks for arithmetic operations that involve non-numeric operands An expression that attempted to numerically add "mary" + "john" would generate this warning.

Illegal or undefined arithmetic operands

checks for illegal or undefined arithmetic operations An operation that attempted to divide by zero would generate this warning.

Uninitialized variables checks for variables that were not explicitly initialized If you attempt to use a variable before explicitly equating it to a value, this warning is generated. PSL Version 2.0 built-in allows you to suppress warning messages that were not functions suppressible in PSL Version 2.0 PSL Version 3.0 built-in allows you to perform an error check of built-in functions functions that were not in PATROL Version 2.0 This option detects the following errors: s passing non-numeric arguments when a numeric value is required s passing a bad argument s passing the wrong number of arguments to a printf() function Note: BMC Software recommends enabling PSL Version 2.0 and PSL Version 3.0 built in functions. Traces Function call tracing reports which functions are called, but does not return information about the arguments. Function call tracing captures both user-defined and built-in functions. reports the arguments passed to all user-defined or built-in functions Function call tracing must also be enabled if you want to use this debugger option. Function return value tracing Variable assignment tracing reports the value returned by calls to all user-defined or built-in functions reports the variable name (if available), and the value assigned to it

Function argument tracing

Chapter 15

Scheduling Tasks

323

Tasks

Check Box PSL errno tracking PSL lock tracing

Description reports any nonzero values stored in the PSL errno variable reports the interprocess actions that occur during lock() and unlock() function processing This option reports back the granting, denying, and releasing of locks.

5 For the Scheduler Facility Debugger group box item, select the check box next to
each selection that you want to turn on for run-time checks and traces. When a debugging selection is turned on, debug data is collected for that option. Check Box Run-Time Checks Arithmetic operations with NULL strings Arithmetic operations with non-numeric operands checks to ensure that numeric variables are explicitly initialized to 0, instead of defaulting to a NULL string. checks for arithmetic operations that involve non-numeric operands An expression that attempted to numerically add "mary" + "john" would generate this warning. Illegal or undefined arithmetic operands checks for illegal or undefined arithmetic operations An operation that attempted to divide by zero would generate this warning. Uninitialized variables checks for variables that were not explicitly initialized If you attempt to use a variable before explicitly equating it to a value, this warning is generated. PSL Version 2.0 built-in allows you to suppress warning messages that were not functions suppressible in PSL Version 2.0 PSL Version 3.0 built-in allows you to perform an error check of built-in functions functions that were not in PATROL Version 2.0 This option detects the following errors: s passing non-numeric arguments when a numeric value is required s passing a bad argument s passing the wrong number of arguments to a printf() function Note: BMC Software recommends enabling PSL Version 2.0 and PSL Version 3.0 built in functions. Description

324

PATROL Knowledge Module for Unix User Guide

Tasks

Check Box Traces Function call tracing

Description reports which functions are called, but does not return information about the arguments. Function call tracing captures both user-defined and built-in functions. reports the arguments passed to all user-defined or built-in functions Function call tracing must also be enabled if you want to use this debugger option.

Function argument tracing

Function return value tracing Variable assignment tracing PSL errno tracking PSL lock tracing

reports the value returned by calls to all user-defined or built-in functions reports the variable name (if available), and the value assigned to it reports any nonzero values stored in the PSL errno variable reports the interprocess actions that occur during lock() and unlock() function processing This option reports back the granting, denying, and releasing of locks.

6 To complete the PATROL Scheduler PSL Debugger dialog box, choose one of the
following options:
s

Click Accept to toggle the debugger data collection on for the items that you selected. Click Cancel to close the dialog box without saving your changes.

NOTE
s

The left side of the dialog box controls debug options for the scheduler daemon. The right side of the dialog box controls debug options for the graphical user interface (GUI). Debug information is only sent to the PATROL Agent Console window when the area being debugged is active. The PATROL KM for Unix does not generate debug data when the scheduler daemon is sleeping, or the GUI is not being used. In the left side PATROL begins collecting debug data for the options that you selected. Debug information is displayed in the System Output Window.

Chapter 15

Scheduling Tasks

325

Blackouts

Blackouts
The SCHEDULER application class gives you the ability to suspend monitoring of applications, which is also known as blacking out PATROL objects. This feature enables you to designate which instances are blacked out or hidden from monitoring and when they are. The SCHEDULER tasks that relate to blacking out instances from monitoring are described in detail on the following pages: Task Specifying a Default Directory for Blackout Scripts Creating a Monitoring Blackout Script Viewing Information about a Blackout Script Modifying an Existing Blackout Script Deleting a Blackout Script Scheduling a Blackout Script for Execution Canceling a Scheduled Blackout Script Page 326 327 330 331 334 335 337

Specifying a Default Directory for Blackout Scripts

This task describes how to specify a directory in which blackout scripts will be saved when the supplied path is a relative path name (a path name that does not begin with a slash / character).

Before You Begin

Before you identify the directory location, ensure that the blackout script directory meets the following requirements:
s

The blackout script directory must exist on the system. The PATROL default account must have read and execute permissions for the blackout script directory.

NOTE
If you do not specify a default directory for blackout scripts, blackout scripts will be saved in the PATROL_HOME directory (the directory specified in the PATROL Agent /home variable).

326

PATROL Knowledge Module for Unix User Guide

Blackouts

To Specify the Blackout Script Directory 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Set Default Blackout Script Directory from the pop-up menu.
The Identify Default Blackout Script Directory dialog box is displayed.

3 Type the full path of the blackout script directory in the Enter Full Path field and
click Apply.

NOTE
Do not store blackout script files in the /tmp directory; many Unix systems automatically clear this directory upon system reboot.

Creating a Monitoring Blackout Script

This task describes how to create a blackout script that suspends propagation of state changes (for example, alarms and warnings) on selected application classes and instances.

Before You Begin

The PATROL KM for Unix will create the blackout script for you; you do not have to write the script before you begin this procedure. However, you must ensure that the default PATROL account has read and execute permissions on the directory in which the blackout script resides.

To Create a Blackout Script for Immediate Execution 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Blackouts from the SCHEDULER application menu.

The Schedule Blackouts dialog box is displayed.

3 Select Add New Blackout Script and click OK.

The Blackout Script File Name dialog box is displayed.

Chapter 15

Scheduling Tasks

327

Blackouts

4 Type a file name (and path, if desired) in the Blackout Script Full File Path field and
click OK.

NOTE
The PATROL default account must have read and execute permissions on the directory in which the script is stored.

The Blackout Instance Selection dialog box is displayed.

5 Use the following table to perform the actions necessary to select application
instances for a blackout period. To Achieve this Result Perform the Following Action(s)

suspend monitoring for all Click the Select All Instances option button and application class instances on a click Select. computer suspend monitoring for all application instances within selected application classes 1. In the Application List group box, select all the application classes for which you want to suspend monitoring. 2. Click the Select All Instances in the Highlighted Classes option button and click Select. suspend monitoring for selected application instances 1. In the Application List group box, select the application class for which you want to suspend monitoring. 2. Click the Select Only the Highlighted Instances option button and click Select. The application class expands to list all instances within the application class. 3. Highlight the instances for which you want to suspend monitoring and click Select. clear all of your instance Click the Clear All Selections option button and selection choices and start over click Select.

NOTE
If you want to remove an application class or instance from the script, highlight it in the Selected Instances group box and click Select.

6 Click Done.
The Add New Blackout Script dialog box is displayed.

328

PATROL Knowledge Module for Unix User Guide

Blackouts

NOTE
The Blackout Script Status dialog box is also displayed. This dialog box provides information about the status of the blackout script; you cannot enter data into it.

7 Use the spinner buttons or type a value in the following fields:

Field Days Hrs Mins Description indicate how many days the blackout period should last indicate how many hours the blackout period should last indicate how many minutes the blackout period should last

8 Perform one of the following actions to set monitoring blackout periods: A If you want to set the same blackout period for all instances in the blackout
script, click the Apply period to all remaining instances button.

B If you want to set different blackout periods for each application instances, do
not select the Apply period to all remaining instances button. If this button is not selected, the Blackout Period dialog box will display for each instance in the script, allowing you to set different blackout time periods for each instance.

9 Click Apply.
The Add New Blackout Script Name Summary dialog box is displayed.

10 Click Continue.
A confirmation dialog box displays, asking if you want to schedule the script for execution now.

11 Perform one of the following actions: A Click Yes if you want to schedule the blackout script for execution now. B Click No if you do not want to schedule the blackout script for execution now. If
you click No, you must perform the steps in Scheduling a Blackout Script for Execution on page 335 before the blackout script will execute. If you clicked Yes, the Schedule PATROL PslExecute() Action dialog box is displayed. Proceed to step 14.

Chapter 15

Scheduling Tasks

329

Blackouts

12 In the Execute Script at: group box, use the spinner buttons or type a value in the
following fields to schedule the blackout script: Field Month Day Year hh Description indicate the month in which the blackout script should execute indicate the date on which the blackout script should execute indicate the year in which the blackout script should execute indicate at which hour of the day the blackout script should execute Enter hours in military time. For example, if you want to schedule a script to execute at 4 P.M., enter 16 in this field. mm ss indicate at which minute the blackout script should execute indicate at which second the blackout script should execute

13 If you want to set the blackout script to occur at regularly scheduled intervals, use
the spinner buttons or type values in the following fields of the Repeat Execution: group box: Field Days hh mm ss Description number of days that should pass between blackout script executions number of hours that should pass between blackout script executions number of minutes that should pass between blackout script executions number of seconds that should pass between blackout script executions

14 Click Accept.
The dialog box closes, and PATROL schedules the blackout script for execution.

Viewing Information about a Blackout Script

This task describes how to view information about a blackout script.

1 Access the SCHEDULER application menu as described in Accessing KM

Commands and InfoBoxes on page 402.

2 Select Schedule Blackouts from the SCHEDULER application menu. 3 Select a blackout script from the Blackout Script Files group box.

330

PATROL Knowledge Module for Unix User Guide

Blackouts

4 Select View Selected Script from the Actions group box and click OK.
The View Selected Script dialog box is displayed. This dialog box provides the following information about the selected blackout script:
s s s

full path and file name of the blackout script date and time that the blackout script was last modified instances that are included in the blackout script, and their scheduled blackout times

5 Click Done.
The dialog box closes.

Modifying an Existing Blackout Script

This task describes how to modify an existing blackout script.

Before You Begin

You can only use this procedure to modify blackout scripts that you created using the Scheduling Blackouts command. If you created a blackout script outside of PATROL, you cannot modify it using this procedure.

To Modify an Existing Blackout Script 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Blackouts from the SCHEDULER application menu.

The PATROL Blackout Scheduler dialog box is displayed.

3 Select a blackout script from the Blackout Script Files group box. 4 Select Modify Selected Script from the Actions group box and click OK.
The Blackout Instance Selection dialog box is displayed.

Chapter 15

Scheduling Tasks

331

Blackouts

5 Use the following table to perform the actions necessary to select application
instances for a blackout period. To Achieve this Result Perform the Following Action(s)

suspend monitoring for all Click the Select All Instances option button and click application class instances Select. on a computer suspend monitoring for all 1. In the Application List group box, select all the application instances application classes for which you want to suspend within selected application monitoring. classes 2. Click the Select All Instances in the Highlighted Classes option button and click Select. suspend monitoring for selected application instances 1. In the Application List group box, select the application class for which you want to suspend monitoring. 2. Click the Select Only the Highlighted Instances option button and click Select. The application class expands to list all instances within the application class. 3. Highlight the instances for which you want to suspend monitoring and click Select. clear all of your instance selection choices and start over Click the Clear All Selections option button and click Select.

NOTE
If you want to remove an application class or instance from the script, highlight it in the Selected Instances group box and click Select.

6 Click Done.
The Add New Blackout Script dialog box is displayed.

7 Use the spinner buttons or type a value in the following fields:

Field Days Hrs Mins Description indicate how many days the blackout period should last indicate how many hours the blackout period should last indicate how many minutes the blackout period should last

332

PATROL Knowledge Module for Unix User Guide

Blackouts

8 Perform one of the following actions to set monitoring blackout periods: A If you want to set the same blackout period for all instances in the blackout
script, click the Apply period to all remaining instances button.

B If you want to set different blackout periods for each application instances, do
not select the Apply period to all remaining instances button. If this button is not selected, the Blackout Period dialog box will display for each instance in the script, allowing you to set different blackout time periods for each instance.

9 Click Apply.
The Add New Blackout Script Name Summary dialog box is displayed.

10 Click Continue.
A confirmation dialog box displays, asking if you want to schedule the script for execution now.

11 Perform one of the following actions: A Click Yes if you want to schedule the blackout script for execution now. B Click No if you do not want to schedule the blackout script for execution now. If
you click No, you must perform the steps in Scheduling a Blackout Script for Execution before the blackout script will execute. If you clicked Yes, the Schedule PATROL PslExecute() Action dialog box is displayed. Proceed to step 15.

12 In the Execute Script at: group box, use the spinner buttons or type a value in the
following fields to schedule the blackout script: Field Month Day Year hh Description indicate the month in which the blackout script should execute indicate the date on which the blackout script should execute indicate the year in which the blackout script should execute indicate at which hour of the day the blackout script should execute Enter hours in military time. For example, if you want to schedule a script to execute at 4 P.M., enter 16 in this field. mm ss indicate at which minute the blackout script should execute indicate at which second the blackout script should execute

Chapter 15

Scheduling Tasks

333

Blackouts

13 If you want to set the blackout script to occur at regularly scheduled intervals, use
the spinner buttons or type values in the following fields of the Repeat Execution: group box: Field Days hh mm ss Description number of days that should pass between blackout script executions number of hours that should pass between blackout script executions number of minutes that should pass between blackout script executions number of seconds that should pass between blackout script executions

14 Select the Execute Once Immediately option button if you want the script to execute
now.

NOTE
If you do not select this option and you are scheduling a script that was supposed to run in the past, it will not run again until the repeated execution runtime. For example, if you have a blackout script that is supposed to run every Sunday evening, but you fill out this dialog box on a Monday morning, PATROL will not run the script until the next scheduled execution. You must toggle the Execute Once Immediately button on to run the script now.

15 Click Accept.
The dialog box closes, and PATROL schedules the blackout script for execution.

Deleting a Blackout Script

This task describes how to delete a blackout script. When you perform this task, PATROL removes the blackout script from the designated blackout script directory. If the blackout script is scheduled for execution, instances of the task are also removed from the SCHEDULER.

Before You Begin WARNING

Deleting a blackout script does not end blackouts that are currently in effect. To end a blackout that is current in effect, perform the procedure that is documented in Canceling a Scheduled Blackout Script on page 337.

334

PATROL Knowledge Module for Unix User Guide

Blackouts

To Delete a Blackout Script 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Blackouts from the SCHEDULER application menu. 3 Select a blackout script from the Blackout Script Files group box. 4 Select Delete Selected Script from the Action group box and click OK.
The blackout script is deleted.

Scheduling a Blackout Script for Execution

This task describes how to schedule a blackout script for execution. You can only use this procedure to schedule blackout scripts that were created using the PATROL for Unix blackout scheduling features.

Before You Begin

The blackout script must be one that was created using the blackout scheduling features provided by PATROL for Unix. See Creating a Monitoring Blackout Script for instructions about using PATROL to create a blackout script.

WARNING
If you have created other blackout scripts outside of PATROL, you must use the procedure documented in Scheduling a Task for Execution via the Operating System Command Line Processor to schedule them for execution.

To Schedule an Existing Blackout Script for Execution 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Scheduling Blackouts from the SCHEDULER application menu.

The PATROL Blackout Scheduler dialog box is displayed.

3 Highlight a script in the Blackout Script Files group box. 4 Select Schedule Selected Script from the Action group box and click OK.
The Schedule PATROL PslExecute() Action dialog box is displayed.
Chapter 15 Scheduling Tasks 335

Blackouts

5 In the Execute Script at: group box, use the spinner buttons or type a value in the
following fields to schedule the blackout script: Field Month Day Year hh Description indicate the month in which the blackout script should execute indicate the date on which the blackout script should execute indicate the year in which the blackout script should execute indicate at which hour of the day the blackout script should execute Enter hours in military time. For example, if you want to schedule a script to execute at 4 P.M., enter 16 in this field. mm ss indicate at which minute the blackout script should execute indicate at which second the blackout script should execute

6 If you want to set the blackout script to occur at regularly scheduled intervals, use
the spinner buttons or type values in the following fields of the Repeat Execution: group box: Field Days hh mm ss Description number of days that should pass between blackout script executions number of hours that should pass between blackout script executions number of minutes that should pass between blackout script executions number of seconds that should pass between blackout script executions

7 Select the Execute Once Immediately option button if you want the script to execute
now.

NOTE
If you do not select this option and you are scheduling a script that was supposed to run in the past, it will not run again until the repeated execution runtime. For example, if you have a blackout script that is supposed to run every Sunday evening, but you fill out this dialog box on a Monday morning, PATROL will not run the script until the next scheduled execution. You must toggle the Execute Once Immediately button on to run the script now.

8 Click Accept.
The dialog box closes, and PATROL schedules the blackout script for execution.

336

PATROL Knowledge Module for Unix User Guide

Blackouts

Canceling a Scheduled Blackout Script

This task describes how to cancel the blackout period for a script. When you perform this task, PATROL resumes propagating state changes (for example, alarms and warnings) for the instances that were part of the blackout script. PATROL does not delete the blackout script from its designated location.

To Cancel an Existing Blackout Script 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Schedule Blackouts from the SCHEDULER application menu. 3 Select a blackout script from the Blackout Script Files group box. 4 Select Cancel Existing Blackout (Selected Script) from the Actions group box and click
OK.

The blackout period is canceled, and PATROL resumes propagating state changes for the instances that were part of the script.

Chapter 15

Scheduling Tasks

337

Blackouts

338

PATROL Knowledge Module for Unix User Guide

Chapter

16
340 340 341 341 344 345 345 346 348

16

Reviewing Security
The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor various aspects of the Unix operating system security. With it, you determine which files were created by super users, which files have global write permissions, and which users have no passwords or have multiple user accounts. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Files with SUID or SGID Permissions . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Files with Global Write Privileges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . User Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Users that Failed to Execute su/msu Commands . . . . . . . . . . . . . . . . . Viewing Users without Passwords . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Users with Multiple Sessions Running . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 16 Reviewing Security

339

Introduction

Introduction
PATROL Knowledge Module for Unix and Linux monitors various aspects of the operating system that affect its security. The KM monitors characteristics of files such as whether or not the file has
s s

set user or group permissions global write access

The KM also monitors characteristics of user accounts and user activity that pose a potential threat to security such as which users
s s s

failed to execute a set user (su or msu) command have no or blank passwords are running multiple sessions

NOTE
This chapter discusses how PATROL monitors the Unix operating system. It does not discuss the security features employed by PATROL. For that information, see the getting started guide for your product and the PATROL Security User Guide.

Object Hierarchy
Figure 54

UNIX OS

The SECURITY application class does not possess any instances or contain any parameters. Figure 54 illustrates the basic hierarchy. SECURITY Application Object Hierarchy

OS container
UNIX OS

Application class

340

PATROL Knowledge Module for Unix User Guide

Files

Files
The SECURITY application class enables you to monitor files with potentially dangerous permissions and privileges. With this feature, you can determine if a file has set user or group permissions and if a file can be written to by any user. The SECURITY tasks related to files are described in detail on the following pages: Task Viewing Files with SUID or SGID Permissions Viewing Files with Global Write Privileges Page 341 344

Viewing Files with SUID or SGID Permissions

This task describes how to view a list of files that have the set user ID (SUID) or set group ID (SGID) permission set. With this permission, the file can set its own user ID or group ID, respectively, regardless of the owner or group to which the owner belongs.

Before You Begin

To perform this task, you must provide a user account and password. It does not have to be the root account.

To View a List of Files with SUID and/or SGID Permissions 1 Access the SECURITY application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Administration => List SUID and SGID Files

PATROL prompts you for a user account and password (Figure 55). Figure 55 SECURITY User Name and Password Dialog Box

Chapter 16 Reviewing Security

341

Files

3 Type the appropriate user name and password and click OK.
PATROL displays the Find SUID/SGID Files Like dialog box (Figure 56). Figure 56 Find SUID/SGID Files Like Dialog Box

4 Define the search criteria. The search is a logical AND operation that returns only
files that match all the criteria defined by the following fields. Field Start find at File Name Description the directory structure from which to begin the search. The default is root (/). the regular expression pattern for one or more files. For more information, see CRegular Expressions on page 407. Leave this field blank to search for any file name. the owner of the file(s) the group to which the owner of the file(s) belongs the period in days between the change date and the current date restricts the operation to the disk drive on which the directory defined in the Start find at field resides

File Owner File Group File Updated in n days ago Limit to one physical disk

5 Click Find.
PATROL builds a search expression, searches the system for files that match the criteria, and writes the results to a PATROL task object (List SUID and SGID Files) in the UNIX OS container.

342

PATROL Knowledge Module for Unix User Guide

Files

6 Access the List SUID and SGID Files task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> List SUID and SGID Files Begin: find /oracle/empdata/payroll \( -perm -004000 -o -perm -002000 \) -exec ls -ld {} \; -rwsr-sr-x 1 3Supvr admin 129 Jun 20 14:38 /oracle/empdata/payroll/may_2001 -------------------- >>>>> DONE <<<<< --------------------

Output Format
The output has the following format. Table 54 describes each expression.
Begin: find /path -name <text> \( -perm -004000 -o -perm -002000 \) -exec <ls> -ld {} \; -rwsr-sr-- # <username> <group> ### MMM DD hh:mm <path>/<file_name.ext> -------------------- >>>>> DONE <<<<< --------------------

Table 54

Find Files Like Command Output Format find command starting point for search file name pattern

Search Expression Begin: find /path - name <text?*>

- perm -00400 -o -perm defines the permissions of files that the find command looks for -002000 -exec ls -ld {} List of Files -rwsr-sr-# username dev ### MMM DD hh:mm location file permissions number of links to the file owner of the file group of owner size in bytes date-time stamp path and file name lists the contents of the specified directory options for the ls command: -l long format listing, -d list only the directory names and not its contents argument that substitutes current file

Chapter 16 Reviewing Security

343

Files

Viewing Files with Global Write Privileges

This task describes how to view all files that can be written to by any user on the system.

Before You Begin

To perform this task, you must provide a user account and password. It does not have to be the root account.

To View a List of Files that Any User Can Write To 1 Access the SECURITY application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Administration => List Files With Global Write.

PATROL prompts you for a user account and password.

3 Type the appropriate user name and password and click OK.
PATROL writes the results to a PATROL task object (List Files with Global Write) in the UNIX OS container.

4 Access the List Files with Global Write task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### /usr/openwin/lib/locale/libs /var/ntb/tmp /var/ntb/appcnfg/appmngr/System_Admin /etc/PERL /dev/stty

Output Format
The output has the following format. Table 55 describes each field.
directory/filename -------------------- >>>>> DONE <<<<< --------------------

Table 55 Field

List Files With Global Write Command Output Format Description

directory/filename path and filename of file or directory with global write permissions

344

PATROL Knowledge Module for Unix User Guide

User Activity

User Activity
The SECURITY application class enables you to monitor user activity that could compromise your systems security. With this feature, you can determine which users have no or blank passwords, which users are running multiple sessions, and which users attempted to perform a set user command and failed. The SECURITY tasks related to users are described in detail on the following pages: Task Viewing Users that Failed to Execute su/msu Commands Viewing Users without Passwords Viewing Users with Multiple Sessions Running Page 345 346 348

Viewing Users that Failed to Execute su/msu Commands

This task describes how to view a list of users who attempted and failed to execute a set user (su) or (msu) command.

Before You Begin

To perform this task, you must provide a user account and password. It does not have to be the root account.

To View a List of Users that Failed to Execute su/msu Commands 1 Access the SECURITY application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Administration => List Failed su/msu logins.

PATROL prompts you for a user account and password.

3 Type the appropriate user name and password and click OK.
PATROL writes the information to a PATROL task object (List Failed su/msu logins) in the UNIX OS container.

Chapter 16 Reviewing Security

345

User Activity

4 Access the List Failed su/msu logins task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> List Failed su/msu logins SU 01/02 15:56 - pts/8 dpallet-ssp SU 01/17 09:39 - pts/8 rabby-root SU 01/18 17:45 - pts/17 mmoulin-root SU 01/23 15:13 - pts/13 nedned-root

Output Format
The output has the following format. Table 56 describes each field.
cmd mm/dd hh:mm - pts/# acct_from-acct_to

Table 56 Field cmd

List Failed su/msu logins Command Output Format Description the unsuccessful command subecome super user (root) or other user msua variant of su that is no longer supported by platforms on which PATROL KM for Unix and Linux runs month and date on which the failed login attempt occurred time at which the failed login attempt occurred pseudo tty subsystem slave terminal name the user account from which the command was executed the user account to which the acct_from user attempted to switch

mm/dd hh:mm pts/# acct_from acct_to

Viewing Users without Passwords

This task describes how to view a list of users who either do not have a password or have a blank password.

Before You Begin

To perform this task, you must provide a user account and password. It does not have to be the root account.

346

PATROL Knowledge Module for Unix User Guide

User Activity

To View a List of Users that Do Not Have a Password or Have a Blank Password 1 Access the SECURITY application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Administration => List Users Without Passwords.

PATROL prompts you for a user account and password.

3 Type the appropriate user name and password and click OK.
PATROL writes the information to a PATROL task object (List Users Without Password) in the UNIX OS container.

4 Access the List Users Without Password task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results.Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> List users with-out password ------------------ >>>>> NIS LIST <<<<< -----------------guest2 admin_request larryking ADMIN-NEW ------------------ >>>>> LOCAL LIST <<<<< ----------------------------------- >>>>> DONE <<<<< --------------------

Output Format
The output has the following format. Table 57 describes each field.
------------------ >>>>> LIST SEPARATOR <<<<< -----------------username

Table 57 Field

List Users Without Passwords Command Output Format Description a line that separates and categorizes the accounts based on their location in the system NISNIS accounts without passwords LOCALaccounts on the local machine without passwords the user name of an account that either does not have a password or has a blank password

list separator

username

Chapter 16 Reviewing Security

347

User Activity

Viewing Users with Multiple Sessions Running

This task describes how to view a list of users who have duplicate user ID entries, a condition that results from running multiple sessions.

Before You Begin

To perform this task, you must provide a user account and password. It does not have to be the root account.

To View a List of Users with Multiple Sessions Running 1 Access the SECURITY application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Administration => List Duplicate User ID Entry.

PATROL prompts you for a user account and password.

3 Type the appropriate user name and password and click OK.
PATROL writes the information to a PATROL task object (List Duplicate User ID Entry) in the UNIX OS container.

4 Access the List Duplicate User ID Entry task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> List users with duplicate user id user id: 9351 -- vdolor, tg1_a, tg1_b, tg2_a, tg1_b, pdolor, user id: 4262 -- mhartman, ldap, user id: 923 -- 3Supvr, 4Supvr, user id: 1092 -- r45, r46, user id: 2784 -- nots, yesman, -------------------- >>>>> DONE <<<<< --------------------

348

PATROL Knowledge Module for Unix User Guide

User Activity

Output Format
The output has the following format. Table 58 describes each field.
user id: 4262 -- <username_#>, <username_#+1>,

Table 58 Field

List Duplicate User ID Entries Command Output Format Description the user ID the user name under which an additional session is run

user id: # # # # <username_# >

Chapter 16 Reviewing Security

349

User Activity

350

PATROL Knowledge Module for Unix User Guide

Chapter

17
352 352 353 353 354 354 355 356 357 360 361 362 363 364 364 365 366 367 369 370

17

Monitoring the Health of SNMP

The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor the PATROL SNMP system. Simple Network Management Protocol (SNMP) support is built into the PATROL Agent. The application class checks to see if the SNMP Master Agent and subagent are running, lists configuration information, and provides diagnostic tests with which you can troubleshoot problems. The following sections appear in this chapter: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PATROL SNMP System Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Role of Each Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SNMP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Changing SNMP Query Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Listing Configuration Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing PATROL SNMP MIB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Status of SNMP Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting the SNMP Master Agent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Starting the SNMP Subagent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stopping the SNMP Subagent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tests SNMP Functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Testing the SNMP Master Agents Ability to Receive Traps . . . . . . . . . . . . . . . . Generating Manual SNMP Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Querying the PATROL Agent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Querying the SNMP PATROL platform.km . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determining if Any SNMP Managers are Listening . . . . . . . . . . . . . . . . . . . . . . . Testing PATROL Event Managers Automated SNMP Traps . . . . . . . . . . . . . . .

Chapter 17

Monitoring the Health of SNMP

351

Introduction

Introduction
The SNMPHealth application monitors the PATROL SNMP system. This application class also provides the ability to
s s s s s s

change the SNMP query port list PATROL SNMP system configuration information view PATROL SNMP MIB database information start SNMP Master Agent start and stop SNMP subagent test SNMP functionality

For information on how to configure the PATROL SNMP system, see the chapters on SNMP in the PATROL Agent Reference Manual.

Object Hierarchy
The SNMPHealth application class contains parameters with information on the PATROL SNMP system configuration, status of the SNMP subagent, and results from test run against the system. Figure 57 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 57 SNMPHealth Application Object Hierarchy

OS container
UNIX OS

Application class

Parameters

352

PATROL Knowledge Module for Unix User Guide

PATROL SNMP System Architecture

PATROL SNMP System Architecture

The PATROL SNMP architecture comprises an SNMP subagent that is part of the PATROL Agent and an SNMP Master Agent that is a separate, external process.

Architecture
Figure 58 shows the relationships between the PATROL SNMP Master Agent, the PATROL Agent, PATROL SNMP subagents, SNMP subagents, and the SNMP manager. It also lists their default port numbers. Figure 58 PATROL SNMP System Architecture
SNMP Manager 1. generates/sends requests to master or subagent 2. listens for traps

PATROL Agent sends traps to SNMP Manager

PA

PATROL Agent
resides on each remote machine PATROL Event Manager

SNMP Subagent
Makes PATROL KMs available via SNMP and sends traps to SNMP managers

Requester is an SNMP Manager and has nothing to do with PATROL

Port 8161PATROL standard port Port 161industry standard port

Port 199 Third-Party SNMP Subagents

PATROL SNMP Master Agent

looks at requests and dispatches information

Table 59 describes the role that each component plays in the PATROL SNMP architecture.

NOTE
Port 161 is an industry standard, while port 8161 is a PATROL standard. By running on port 8161, you can avoid possible conflicts with operating system vendors who are already using port 161 for an SNMP agent.

Chapter 17

Monitoring the Health of SNMP

353

SNMP Configuration

Role of Each Component

Table 59 lists each component in the PATROL SNMP system and describes its role. Table 59 PATROL SNMP Component Definitions SNMP Role generates traps provides SNMP information to SNMP managers in the form of responses to SNMP queries

Component PATROL Agent PATROL SNMP Master Agent

PATROL SNMP Subagent responds to SNMP queries by going into internal data structures and sending information back SNMP subagents can also generate or send PATROL traps directly to the SNMP Manager. PATROL SNMP Manager Functionality (PSL Functions) receives traps and issues queries to an SNMP agent The manager controls the agent by making SNMP requests of the agent and by setting variables in the agents management information base (MIB).

SNMP Configuration
The SNMPHealth application class enables you to review the PATROL SNMP system settings and change the most crucial setting, the query port number. With this application, you can also examine the information stored in the SNMP Management Information Base (MIB) by the PATROL Agent. SNMP configuration tasks are described in detail on the following pages: Task Changing SNMP Query Port Listing Configuration Information Viewing PATROL SNMP MIB Page 355 356 357

354

PATROL Knowledge Module for Unix User Guide

SNMP Configuration

Changing SNMP Query Port

This task describes how to change SNMP Query Port. External processes send queries to the PATROL Agent through its SNMP Query Port. This port is defined in $PATROL_HOME/log/snmpmagt.cfg. The default value of the SNMP Query Port is 8161, which is registered to BMC Software, Inc. Change this port setting if some other process already uses this port. The PATROL SNMP Master Agent will not start if the port designated as the SNMP Query Port is used by another process.

NOTE
For the change to take effect, you must stop and restart the PATROL SNMP Master Agent.

Before You Begin

Make sure the port that you want to use is available.

To Change the SNMP Query Port 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select SNMP Configuration => Change SNMP Query Port.

PATROL displays the Change SNMP Query Port dialog box (Figure 59). Figure 59 Change SNMP Query Port Dialog Box

Chapter 17

Monitoring the Health of SNMP

355

SNMP Configuration

3 Delete the existing port number and type in a new one. 4 Click Accept.
PATROL closes the dialog box and changes the value in the snmpmagt.cfg file. To force the PATROL SNMP Master Agent to use the new port, you must stop and restart the snmpmagt process. For information on starting and stopping the master agent, see Starting the SNMP Master Agent on page 361.

Listing Configuration Information

This task describes how to display your PATROL SNMP system configuration information. The command lists
s

the PATROL SNMP variables and their values stored in the PATROL Agent configuration database the status of the SNMPStart parameter the status of the snmpmagt Query Port the status of the snmpmagt process the status of the SNMP subagent the contents of the configuration file, snmpmagt.cfg

For additional information on the PATROL SNMP system, see the PATROL Agent Reference Manual.

To View Configuration Information 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select SNMP Configuration => List Configuration Information.

PATROL writes the information to a PATROL task object (List Configuration Information) in the UNIX OS container.

356

PATROL Knowledge Module for Unix User Guide

SNMP Configuration

3 Access the List Configuration Information task object as described in Accessing

KM Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### SNMPHealth Configuration Information collected for host <host> at Tue Aug 28 16:47:34 CCYY SNMPHealth Configuration SNMPHealth Configuration SNMPHealth Configuration SNMPHealth Configuration <continued> Info computer class: SOLARIS Info... /snmp/agent_auto_start: "yes" Info... /snmp/agent_r_community: "public" Info... /snmp/agent_w_community: "private"

To Save Configuration Information to an ASCII File

Perform step 1 through step 3 in To View Configuration Information on page 356, and then perform the following steps:

1 Select File => Save As.

PATROL displays the Save Script File dialog box.

2 Enter the filename and location where you want to save this information. 3 Click OK.
PATROL writes the configuration information to the designated file and location. For additional information on specific configuration settings, see the PATROL Agent Reference Manual.

Viewing PATROL SNMP MIB

This task describes how to check the SNMP information stored in a MIB. The dialog defaults to the PATROL Agent MIB. However, if you know the computer name, listening port, community string, and OID (object identifier) of another process that uses SNMP, you can walk that processs MIBs. This feature is primarily used to test that the PATROL SNMP Master Agent is populating its MIB.

Chapter 17

Monitoring the Health of SNMP

357

SNMP Configuration

Before You Begin

For detailed information on the PATROL SNMP MIB, see the appendix, PATROL SNMP MIB, in the PATROL Agent Reference Manual.

To View the SNMP Management Information Base for PATROL 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Test PATROL SNMP System => Walk PATROL SNMP MIB.
PATROL displays the Walk PATROL SNMP MIB dialog box (Figure 60). Figure 60 Walk PATROL SNMP MIB Dialog Box

3 Define the portion of the PATROL SNMP MIB that you want to view.
Field Description

Host to Query the machine whose SNMP MIB you want to view Listening Port the port on which the master agent is listening Community

the SNMP community string defines the relationship between an SNMP server system and the client systems and acts like a password to control the clients' access to the server the amount of time PATROL waits before exiting; the default is set in the PATROL Agent Configuration variable /SNMP/default_timeout

Timeout

358

PATROL Knowledge Module for Unix User Guide

SNMP Configuration

Field
Retries

Description the number of times PATROL attempts to access the hosts MIB; the default is set in the PATROL Agent Configuration variable /SNMP/default_timeout the object ID

OID

4 Click OK.
PATROL writes the information to a PATROL task object (Walk PATROL SNMP MIB) in the UNIX OS container.

5 Access the Walk PATROL SNMP MIB task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
snmp_get returned: 2.1.2.1.2.1.2121.1.1.1.2.0 1 string / 2.1.2.1.2.1.2121.1.1.1.3.1.1.8.80.82.79.67.67.79.78.84 8 string PROCCONT 2.1.2.1.2.1.2121.1.1.1.3.1.1.6.75.69.82.78.69.76 6 string KERNEL 2.1.2.1.2.1.2121.1.1.1.3.1.1.3.76.79.71 3 string LOG 2.1.2.1.2.1.2121.1.1.1.3.1.1.9.66.108.111.99.107.83.105.122.101 9 string BlockSize 2.1.2.1.2.1.2121.1.1.1.3.1.1.14.83.101.116.117.112.67.108.101.97.114.84.101.120. 116 14 string SetupClearText 2.1.2.1.2.1.2121.1.1.1.3.1.1.10.67.79.77.80.79.83.73.84.69.83 10 string COMPOSITES 2.1.2.1.2.1.2121.1.1.1.3.1.1.9.78.84.95.72.69.65.76.84.72 9 string NT_HEALTH 2.1.2.1.2.1.2121.1.1.1.3.1.1.8.80.82.79.67.80.82.69.83 8 string PROCPRES 2.1.2.1.2.1.2121.1.1.1.3.1.1.10.70.73.76.69.83.89.83.84.69.77 10 string FILESYSTEM 2.1.2.1.2.1.2121.1.1.1.3.1.1.4.68.73.83.75 4 string DISK 2.1.2.1.2.1.2121.1.1.1.3.1.1.6.76.79.71.77.79.78 6 string LOGMON 2.1.2.1.2.1.2121.1.1.1.3.1.1.11.68.70.66.108.111.99.107.83.105.122.101 11 string DFBlockSize 2.1.2.1.2.1.2121.1.1.1.3.1.1.8.83.69.67.85.82.73.84.89 8 string SECURITY 2.1.2.1.2.1.2121.1.1.1.3.1.1.9.83.78.77.80.83.116.97.114.116 9 string SNMPStart 2.1.2.1.2.1.2121.1.1.1.3.1.1.3.83.77.80 3 string SMP <continued>

Chapter 17

Monitoring the Health of SNMP

359

Status of SNMP Agents

Output Format
The output has the following format. Table 60 describes each expression.
name type length value

Table 60 Field name type

Walk PATROL SNMP MIB Command Output Format Description name of an SNMP MIB variable represented by its numeric object ID (OID) the data type of the value Valid data types include: string integer length of the value if its data type is string If the data type is integer, this field is empty. the value of the variable If the variable does not have a value, this field is empty.

length value

Status of SNMP Agents

The SNMPHealth application class enables you to control the status of the PATROL SNMP system. With the application, you can start the master agent and start and stop the subagent.

NOTE
The PATROL KM for Unix and Linux does not provide the ability to stop the SNMP Master Agent. This action requires root access.

The SNMPHealth start and stop tasks are described in detail on the following pages: Task Starting the SNMP Master Agent Starting the SNMP Subagent Stopping the SNMP Subagent Page 361 362 363

360

PATROL Knowledge Module for Unix User Guide

Status of SNMP Agents

Starting the SNMP Master Agent

This task describes how to start the PATROL SNMP Master Agent.

Before You Begin

s

Check to see that the master agent has stopped running. You can verify the agents status by using one of the following methods: view the SNMPHealth capture diagnostic text parameter run the SNMP List Configuration menu command search the process table for the snmpmagt process

This command will fail if the SMUX port (tcp port 199) or the SNMP query port (defined in PATROL_HOME/lib/snmpmagt.cfg) is used by some other process.

To Start the Master Agent 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Start/Stop SNMP Agent => Start SNMP Master Agent.

After attempting to start the SNMP Master Agent, PATROL refreshes the process cache, sleeps for 60 seconds, and then stops and restarts the PATROL SNMP subagent. PATROL writes its progress to the system output window. Your results should resemble this example:
CCYYMMDDHHMMSS CCYYMMDDHHMMSS CCYYMMDDHHMMSS CCYYMMDDHHMMSS CCYYMMDDHHMMSS Start SNMP Master Agent... snmpmagt Start SNMP Master Agent... Executing Start command: /apps/patrol/ Refresh Process cache and then sleep 60 seconds Start SNMP Master Agent... Patrol SNMP sub-agent stopped Start SNMP Master Agent... Patrol SNMP sub-agent stopped

Chapter 17

Monitoring the Health of SNMP

361

Status of SNMP Agents

Starting the SNMP Subagent

This task describes how to start the SNMP subagent. Generally, this task is performed when the SNMP master agent is running but the SNMP subagent is not. The SNMPHealth application class uses a parameter without an icon, snmp_subagent_start, to ensure that PATROL attempts to start the subagent if it is not started. The parameter checks the status of the subagent every 60 seconds. Perform this task if either the parameter cannot restart the subagent or you do not want to wait for the next restart attempt.

Before You Begin

This command will fail if the SNMP Master Agent is not running.

To Start the subagent 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Start/Stop SNMP Agent => Start SNMP Sub-agent.

PATROL starts the subagent. If the subagent is already running, PATROL stops the subagent and restarts it. PATROL writes its progress to the system output window. Your results should resemble this example:
CCYYMMDDHHMMSS CCYYMMDDHHMMSS CCYYMMDDHHMMSS CCYYMMDDHHMMSS SNMPHealth SNMPHealth SNMPHealth SNMPHealth Start Start Start Start SNMP SNMP SNMP SNMP sub-agent: sub-agent: sub-agent: sub-agent: SNMP sub-agent is already active. Stopping SNMP sub-agent. Starting SNMP sub-agent. SNMP sub-agent is active.

362

PATROL Knowledge Module for Unix User Guide

Status of SNMP Agents

Stopping the SNMP Subagent

This task describes how to stop the SNMP subagent. This task is used primarily for testing the snmp_subagent_monitor and snmp_subagent_start parameters.

Before You Begin

The consequences of stopping the subagent with this menu command include
s

the PATROL SNMP system will not function properly until the SNMP subagent is restarted PATROL changes the snmp_subagent_monitor parameter state to ALARM the snmp_subagent_start parameter attempts to restart the subagent within 60 seconds

NOTE
If you want to stop the SNMP subagent for an extended period of time (more than 60 seconds), you must disable the snmp_subagent_start parameter.

To Stop the Subagent 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Start/Stop SNMP Agent => Stop SNMP Sub-agent.

PATROL stops the subagent. If the subagent is not running, PATROL recognizes that it is inactive and ignores the command. PATROL writes its progress to the system output window.Your results should resemble this example:
CCYYMMDDHHMMSS SNMPHealth Diagnostics: SNMP sub-agent is active. CCYYMMDDHHMMSS SNMPHealth Diagnostics: Stopping SNMP sub-agent. CCYYMMDDHHMMSS SNMPHealth Diagnostics: SNMP sub-agent is not active.

Chapter 17

Monitoring the Health of SNMP

363

Tests SNMP Functionality

Tests SNMP Functionality

The SNMPHealth application class serves as a diagnostic utility. It provides a number of menu commands with which you can test SNMP functionality to identify potential problems and isolate the cause. The SNMPHealth tests are described in detail on the following pages: Task Testing the SNMP Master Agents Ability to Receive Traps Generating Manual SNMP Traps Querying the PATROL Agent Querying the SNMP PATROL platform.km Determining if Any SNMP Managers are Listening Testing PATROL Event Managers Automated SNMP Traps Page 364 365 366 367 369 370

Testing the SNMP Master Agents Ability to Receive Traps

This task describes how to test if the SNMP Master Agent is listening for SNMP traps.

Before You Begin

s

The trap port is stored in the agent configuration variable /snmp/trap_port. The default value is 162. You might need to change this port setting if some other process already uses this port. This task will fail if the port is used by another process. This task should be performed in conjunction with the task, Generating Manual SNMP Traps on page 365. This task provides a mechanism (a PATROL task object) with which to view the manually generated traps.

To Test the Master Agents Ability to Receive Traps 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Test PATROL SNMP System => Listen for SNMP Traps.
PATROL writes the information to a PATROL task object (Listen for SNMP Traps) in the UNIX OS container.

364

PATROL Knowledge Module for Unix User Guide

Tests SNMP Functionality

3 Access the Listen for SNMP Traps task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
SNMP Listen Running...

NOTE
While the task object is executing, it displays any SNMP traps received at the trap port.

Generating Manual SNMP Traps

This task describes how to manually create SNMP traps. The trap port is stored in the agent configuration variable /snmp/trap_port. The default value is 162. You might need to change this port setting if some other process already uses this port. This task will fail if the port is used by another process.

Before You Begin

This task should be performed in conjunction with the task, Testing the SNMP Master Agents Ability to Receive Traps on page 364. This task sends traps to the port on which PATROL is listening for traps.

To Generate Manual Traps 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Test PATROL SNMP System => Generate Manual SNMP Trap.
PATROL displays the Generate a Manual SNMP Trap dialog box.

Chapter 17

Monitoring the Health of SNMP

365

Tests SNMP Functionality

3 Edit the Target host/port/community information to correspond to your PATROL

SNMP system configuration. If you do not provide a community string, PATROL uses the one stored in the agent configuration variable /SNMP/default_r_community.

4 Click OK.
PATROL closes the dialog box and sends the trap to the SNMP listening process started in step 2.

5 Access the Listen for SNMP Traps task object. Your results should resemble this
example:
SNMP Listen Running... From: 27.137.19.172 (2.1.2.1.2.1.2121) Enterprise Specific (1) Uptime: 2 day(s) 4:22:11 (1) 2.1.2.1.2.1.2121.1 6 string test 1 2.1.2.1.2.1.2121.2 6 string test 2 2.1.2.1.2.1.2121.2 6 string test 2 2.1.2.1.2.1.2121.2 6 string test 2

Querying the PATROL Agent

This task describes how to determine if the PATROL Agent can make SNMP queries. It demonstrates that PATROL can make SNMP requests of the PATROL SNMP Master Agent.

To Test the Master Agents Ability to Receive Traps 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Test PATROL SNMP System => SNMP Query PATROL Master Agent.
PATROL displays the SNMP Query PATROL Master Agent dialog box.

366

PATROL Knowledge Module for Unix User Guide

Tests SNMP Functionality

Figure 61

SNMP Query PATROL Master Agent Dialog Box

3 Click OK.
PATROL writes the information to a PATROL task object (SNMP Query PATROL Master Agent) in the UNIX OS container.

4 Access the SNMP Query PATROL Master Agent task object as described in
Accessing KM Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
host: listening_port: community: timeout: retries: oid: <accounting_db_3> <1161> <public> <> <> <.1.3.6.1.2.1.1.1.0>

SNMP Query Master Agent: errno: 0 snmp_get returned: 1.3.6.1.2.1.1.1.0

111 string

Querying the SNMP PATROL platform.km

This task describes how to determine if the PATROL Agent application class can receive SNMP queries. It demonstrates that PATROL can make SNMP requests inside the PATROL Agent and access the platform.km, which is automatically loaded by the agent.

Chapter 17

Monitoring the Health of SNMP

367

Tests SNMP Functionality

To Test the Ability of the Platform.km to Receive Traps 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Test PATROL SNMP System => SNMP Query PATROL platform.km.
PATROL displays the SNMP Query platform.km dialog box. Figure 62 SNMP Query PATROL Platform.km Dialog Box

3 Click OK.
PATROL writes the information to a PATROL task object (SNMP Query PATROL platform.km) in the UNIX OS container.

368

PATROL Knowledge Module for Unix User Guide

Tests SNMP Functionality

4 Access the SNMP Query PATROL platform.km task object as described in

Accessing KM Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
host: listening_port: community: timeout: retries: oid: <accounting_db_4> <1161> <public> <> <> <.2.1.2.1.2.1.2121.1.1.1.6.1.1.0>

SNMP Query PATROL <platform>.km: errno: 0 snmp_get returned: 2.1.2.1.2.1.2121.1.1.1.6.1.1.0 UNIX

4 string

Determining if Any SNMP Managers are Listening

This task describes how to determine if any SNMP Managers are configured in either the agent configuration variable /snmp/piV1m_list or in the SNMP configuration file, snmpmagt.cfg, or both and are listening on the default port. It demonstrates that PATROL can make SNMP queries to the industry standard SNMP query port, 161.

Before You Begin

This task uses the PSL command, Snmp_h_get( ), which automatically uses the port number defined in the agent configuration variable /snmp/default_port.

To determine if any SNMP managers are listening 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Test PATROL SNMP System => SNMP Query Using snmp_h_get( ).
PATROL displays the SNMP Query Using snmp_h_get( ) dialog box.

3 Click OK.
Chapter 17 Monitoring the Health of SNMP 369

Tests SNMP Functionality

PATROL writes the information to a PATROL task object [SNMP Query Using snmp_h_get( )] in the UNIX OS container.

4 Access the SNMP Query Using snmp_h_get( ) task object as described in

Accessing KM Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
host: oid: <inventory_db_dallas> <.1.3.6.1.2.1.1.1.0>

SNMP Query using snmp_h_get()... UDP port 161:

NOTE
If an SNMP manager is not listening on port 161, PATROL times out and returns an error (errorno=95).

Testing PATROL Event Managers Automated SNMP Traps

This task describes how to test whether automatic SNMP traps (that is those based on the PATROL event manager system) are generated by PATROL when a parameter changes state. They demonstrate that certain PATROL Events have the send trap option turned on.

Before You Begin

To generate automatic SNMP traps, at least one SNMP Manager must be registered to receive traps. A list of SNMP Managers are stored in the agent configuration variable /snmp/piV1m_list. The format is host/port/community, where the port and community string default to 162 and public, respectively.

To Test the Event Managers Automated SNMP Traps 1 Access the SNMPHealth application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Test PATROL SNMP System => Test SNMP Traps from PEM.
PATROL creates a task object entitled Test SNMP Traps from PEM. PATROL writes the information to a PATROL task object (Test SNMP Traps from PEM) in the UNIX OS container.

370

PATROL Knowledge Module for Unix User Guide

Tests SNMP Functionality

3 Access the Listen for SNMP Traps task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
SNMP Listen Running...

4 Select one of the following tests: Test PATROL SNMP System =>
s s s

Set testparm Parameter OKtest PATROL standard event class 9 Set testparm Parameter Alarmtest PATROL standard event class 11 Set testparm Parameter Warntest PATROL standard event class 11

PATROL changes the state of testparm in the SNMPHealth application. PATROL also sends a message to the message area. Figure 63 State Change Messages Based on Automated Traps

NOTE
If you attempt to run a test and the testparm is already in that state (for example, select Set testparm Parameter OK when testparm is already in the OK state), PATROL recognizes this fact and does not send a message to the message area.

5 Access the Listen for SNMP Traps task object. Your results should resemble this
example, a trap for Set testparm parameter to Warn command:
From: 27.137.19.172 (2.1.2.1.2.1.2121.1.1.2) Enterprise Specific (4) Uptime: 0 day(s) 20:17:29 (7304900) 2.1.2.1.2.1.2121.1.1.2.1.0 93 string Alarm #1 of global parameter 'testparm' triggered on 'SNMPHealth.SNMPHealth'. 1 <= 1.00 <= 1 2.1.2.1.2.1.2121.1.1.2.2.0 31 string /SNMPHealth/SNMPHealth/testparm 2.1.2.1.2.1.2121.1.1.2.3.0 0 string

Chapter 17

Monitoring the Health of SNMP

371

Tests SNMP Functionality

372

PATROL Knowledge Module for Unix User Guide

Chapter

18
374 374 375 375 376 377 378 379 380

18

Monitoring Users
The PATROL Knowledge Module for Unix and Linux gives you the ability to monitor resource consumption by individual users. With it, you can also track the number of users and sessions running on the system. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Object Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Resources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Print Jobs that Belong to a User. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing Processes that Belong to a User . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Amount of Disk Space Consumed by a User . . . . . . . . . . . . . . . . . . Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Number of Users on the System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing the Number of Sessions Running on the System . . . . . . . . . . . . . . . . . .

Chapter 18

Monitoring Users

373

Introduction

Introduction
The PATROL Knowledge Module for Unix and Linux monitors the resource consumption of individual users as well as the number of users and sessions that run on the system. For individual users, it monitors
s s s

the number of print jobs issued by a user the amount of disk space consumed by a user the number of processes owned by a user

For the systems, it monitors

s s

the number of users logged on the number of sessions running

Object Hierarchy
The USERS application class contains several parameters with information about the number of users logged onto the system and the number of sessions running. Figure 64 illustrates the basic hierarchy but does not necessarily display all instances and parameters. Figure 64 USERS Application Object Hierarchy

OS container
UNIX OS

Application class

Parameters

374

PATROL Knowledge Module for Unix User Guide

Resources

Resources
The USERS application class enables you to monitor an individual users resource consumption. With this feature, you can discover how many resources (CPU, disk space, printer) an individual user is consuming. The USERS tasks are described in detail on the following pages: Task Viewing Print Jobs that Belong to a User Viewing Processes that Belong to a User Viewing the Amount of Disk Space Consumed by a User Page 375 376 377

Viewing Print Jobs that Belong to a User

This task describes how to view all print jobs that were requested by a specific user.

Before You Begin

PATROL can list print jobs by user and by printer. This task lists all print jobs owned by a single user. For information on how to view print jobs on a specific printer, see Viewing All Print Jobs Assigned to a Printer on page 231.

To List a Users Print Jobs 1 Access the USERS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select List Print Jobs.

PATROL displays the List Print Jobs dialog box. Figure 65 List Print Jobs Dialog Box

Chapter 18

Monitoring Users

375

Resources

3 Type the user name and click Apply.

PATROL writes the results to a PATROL task object (List Print Jobs Users) in the UNIX OS container.

4 Access the List Print Jobs Users task object as described in Accessing KM
Commands and InfoBoxes on page 402 and view the results.

Viewing Processes that Belong to a User

This task describes how to view all processes that were started by a specific user.

To List a Users Processes 1 Access the USERS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Process List.

PATROL displays the Process List dialog box.

3 Type the user name and click Apply.

PATROL writes the results to a PATROL task object (Process List Users) in the UNIX OS container.

4 Access the Process List Users task object as described in Accessing KM

Commands and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> List processes for user: 3Suprv 2335 2334 3Suprv S 1768 0 ksh -ksh 2450 1 3Suprv S 12168 2 Patrol Patrol dev 2463 4747 3Suprv S 1760 0 ksh bin/ksh 2334 142 3Suprv S 3728 0 xterm usr/openwin/bin/xterm -ls -sb -cr -white -display 172.19.137.27:0 -name ---------------- >>>>> DONE <<<<< ----------------

weights

376

PATROL Knowledge Module for Unix User Guide

Resources

Output Format
The output has the following format. Table 61 defines each field.
pid ppid user s size cput cmd name cmd line

Table 61 Field pid ppid user s

Process List Command Output Format Description the process ID of the process the process ID of the parent process the name of the user that owns the process the status of the process Iintermediate 0nonexistent Rrunning Ssleeping Tstopped Wwaiting Xgrowing Zzombie, terminated the total size of the process in virtual memory, in blocks the first word of the command line that started the process the complete command line (command and arguments up to 80 characters) that started the process; note that the command line can be modified during the process execution

size cmd name cmd line

Viewing the Amount of Disk Space Consumed by a User

This task describes how to view how much disk space a specific user is using.

To List the Amount of Disk Space Consumed by a User 1 Access the USERS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Disk Space.

PATROL displays the Disk Space dialog box.

3 Type the user name and click Apply.

PATROL writes the results to a PATROL task object (Disk Space Users) in the UNIX OS container.

Chapter 18

Monitoring Users

377

Numbers

4 Access the Disk Space Users task object as described in Accessing KM Commands
and InfoBoxes on page 402 and view the results. Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> show user: 3Suprv disk space information Processing, this might take a while, please wait ... NIS is detected, get information from NIS database Disk usage for user: 3Suprv 20 /home/3Suprv/pcache/tmp 68 /home/3Suprv/pcache/knowledge 16 /home/3Suprv/pcache/psl 16 /home/3Suprv/nsmail ---------------- >>>>> DONE <<<<< ----------------

Output Format
The output has the following format. Table 62 defines each field.
size /path/

Table 62 Field size

Disk Space Command Output Format Description the amount of space allocated to a file File size is written in 512-byte units, rounded up to the next 512-byte unit. For example, a 513-byte file is reported as using 2 units, 1024 bytes. the directory path in which files owned by the user are stored

/path/

Numbers
The USERS application class enables you to monitor the amount of usage in terms of users and sessions. With this feature, you can discover how many users are on the system and how many sessions are running. You can view the information for the present moment in time or in the recent past. The USERS tasks are described in detail on the following pages: Task Viewing the Number of Users on the System Viewing the Number of Sessions Running on the System Page 379 380

378

PATROL Knowledge Module for Unix User Guide

Numbers

Viewing the Number of Users on the System

This task describes how to view the total number of users on the system at the present time or in the last couple of hours.

To View the Total Number of Users Now on the System

Access the USERS application InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the USERS InfoBox, which provides the number of users. Figure 66 USERS InfoBox

To View the Total Number of Users Recently on the System 1 Access the USERS application so that you can view its parameters. 2 Open the USRNoUser parameter.
PATROL displays a graph that shows the number of users on the system over time.

Chapter 18

Monitoring Users

379

Numbers

Viewing the Number of Sessions Running on the System

This task describes how to view the total number of sessions running on the system at the present time or in the last couple of hours.

To View the Total Number of Session Now Running on the System

Access the USERS application InfoBox as described in Accessing KM Commands and InfoBoxes on page 402. PATROL displays the USERS InfoBox, which provides the number of sessions.

To View the Total Number of Sessions Recently Running on the System 1 Access the USERS application so that you can view its parameters. 2 Open the USRNoSessions parameter.
PATROL displays a graph that shows the number of sessions running on the system over time.

380

PATROL Knowledge Module for Unix User Guide

Appendix

A
382 382 383 385 386 388 388 389 390 391 392 392 393 394 395 396 396 397 398 398 399 400 400

Debugging and Running Diagnostics

The PATROL Knowledge Module for Unix and Linux comes equipped with diagnostic and debugging utilities. The following sections appear in this chapter. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Unix Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Capturing DUMP Information for Selected Components. . . . . . . . . . . . . . . . . . . Viewing Installation Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Capturing Log File Diagnostics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Collector Debug Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging FILESYSTEM Data Collection (DFColl). . . . . . . . . . . . . . . . . . . . . . . . Debugging NETWORK Data Collection (NETColl). . . . . . . . . . . . . . . . . . . . . . . . Debugging NFS Data Collection (NFSColl). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging PRINTER Data Collection (PRINTERColl) . . . . . . . . . . . . . . . . . . . . . Debugging PROCESS Data Collection (PSColl) . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging USER and PROCESS Data Collection (USRPROCColl) . . . . . . . . . . Debugging CPU, MEMORY, and KERNAL Data Collection (VMColl) . . . . . . . Debugging DCM Data Collection (DCMColl) . . . . . . . . . . . . . . . . . . . . . . . . . . . . Application Discovery Debug Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging DISK Discovery Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging FILESYSTEM Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging NETWORK Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging PRINTER Discovery Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging SMP Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging SWAP Discovery Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Application Class Debug Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Debugging the SCHEDULER Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Appendix A

Debugging and Running Diagnostics

381

Introduction

Introduction
The PATROL Knowledge Module for Unix and Linux provides diagnostic and debugging utilities to help pinpoint errors if and when they occur. The capabilities include
s s

Viewing installation statistics Capturing DUMP information for selected components Log file diagnostics Debugging Collectors for the following application classes: CPU FILESYSTEM KERNEL MEMORY NETWORK NFS PRINTER PROCESS USERS Debugging Discovery for the following application classes: DISK FILESYSTEM NETWORK PRINTER SMP SWAP

Unix Diagnostics
This PATROL Knowledge Module for Unix and Linux gathers installation and configuration information as well as monitor important operating system log files. The UNIX OS tasks related to the diagnostic procedures are described in detail on the following pages: Task Capturing DUMP Information for Selected Components Page 383

382

PATROL Knowledge Module for Unix User Guide

Unix Diagnostics

Task Viewing Installation Diagnostics Capturing Log File Diagnostics

Page 385 386

Capturing DUMP Information for Selected Components

This task describes how to gather basic system information. With it, you can gather system information as well as to review the current state of various aspects of PATROL such as which KMs are loaded, how many PSL errors have occurred since the agent started, and which tasks are currently being performed.

To Capture Unix Diagnostic Information 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Capture Unix Diagnostics.

PATROL displays the Select Categories of Diagnostics dialog box (Figure 67). Figure 67 Select Categories of Diagnostics Dialog Box

3 Select the %DUMP Options check box. 4 Place the cursor in the Output File for Diagnostics field and type the path and
filename to which you want the diagnostic information written.

Appendix A

Debugging and Running Diagnostics

383

Unix Diagnostics

5 Click Accept.
PATROL displays the DUMP Options dialog box (Figure 68 on page 384). Figure 68 %DUMP Options Dialog Box

6 Select the check boxes of the components for which you want PATROL to provide
information. Table 63 lists the options in alphabetic order. Table 63 Options ALL APP_INSTS APPS CONSOLES ERRORS GLOBALS KM_LIST LIBRARIES PARAMS RTLIST Capture Unix Diagnostics %DUMP Options Description information about PATROL Agent data structure information about each application instances list of applications list of connected consoles list of PSL errors that have occurred since the last reboot list of global channels list of loaded KMs list of loaded libraries list of PATROL Agent parameters information about processes in the PATROL Agent run-time queue

CHANNELS list of open PSL global file channels and open process channels

384

PATROL Knowledge Module for Unix User Guide

Unix Diagnostics

Table 63 Options RUNQ TASKS %PSLPS

Capture Unix Diagnostics %DUMP Options Description list of items scheduled in the run queue provides a list of current tasks provides a list of current PSL processes

7 Click Accept.
PATROL writes the information to the file that you specified in step 4 and to a task object (Capture Unix Diagnostics) in the UNIX OS container.

Viewing Installation Diagnostics

This task describes how to gather PATROL installation information. It writes some general information to a task object and writes the remainder of the information to a log file that you specify.

To Capture Installation Diagnostic Information 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Capture Unix Diagnostics.

PATROL displays the Select Categories of Diagnostics dialog box. To view an example, see Figure 67 on page 383.

3 Select the Installation Diagnostics check box. 4 Place the cursor in the Output File for Diagnostics field and type the path and
filename to which you want the diagnostic information written.

5 Click Accept.
PATROL writes the information to a task object (Capture Unix Diagnostics) in the UNIX OS container. Your results should resemble this example:
# PSL Serial No.: ######### <NOT INTERACTIVE> The remaining diagnostics will be written to the diagnostic log file but will not be displayed in the task window.

Appendix A

Debugging and Running Diagnostics

385

Unix Diagnostics

Capturing Agent Configuration Variables. Capturing Default Agent Configuration Variables. Capturing Non-Default/Overridden Agent Configuration Variables. Capturing diagnostics related to the PATROL installation and environment. Capturing list of symbolic links under PATROL installation. Capturing setuid and setgid files that exist under PATROL installation. Capturing information about apidll, dcm, and bgscollect. BEST1_HOME directory = /local/Patrol35/Solaris-2-8-sparc-64/best1

Capturing setuid and setgid files that exist under BEST1_HOME installation. Capturing license details for PATROL installation. Capturing df -k. Capturing ls -ld on all directores under /local/Patrol35. Capturing ls -lRu /local/Patrol35. Capturing ls -lR /local/Patrol35. Closing OutputDiagFile: /tmp/PDiag.log

Capturing Log File Diagnostics

This task describes how to gather basic system information. With it, you can gather system information as well as to generate debug information for different components of the PATROL Knowledge Module for Unix and Linux.

To Capture Unix Diagnostic Information 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Capture Unix Diagnostics.

PATROL displays the Select Categories of Diagnostics dialog box. To view an example, see Figure 67 on page 383.

3 Select the Log Files check box. 4 Place the cursor in the Output File for Diagnostics field and type the path and
filename to which you want the diagnostic information written.

5 Click Accept.
PATROL displays the Select Log Files to Collect dialog box (Figure 69 on page 387).
386 PATROL Knowledge Module for Unix User Guide

Unix Diagnostics

Figure 69

Select Log Files to Collect Dialog Box

6 Select the check boxes of the log files that you want PATROL to capture. Table 64
lists the options in the order that they appear in the dialog box. Table 64 Options Select Log Files to Collect Options Description

All Logs Contained in gathers log information from all the log files stored in the $PATROL_HOME/log PATROL log directory Agent Error Log DCM Debug Log apidll Log gathers log information the PATROL Agent gathers log information from the Data Collection Manager executable used in the DCM collection method gathers log information from the C library loaded by the PATROL Agent during runtime so that it can read and write to FIFO gathers log information from the bgscollect executable, which accesses the kernel, retrieves the requested data, and writes it to the shared memory segment gathers log information from the MrConsumer library, which is used by DCM to read the data from the shared memory segment written by bgscollect executable

bgscollect Log

MrConsumer Log

7 Click Accept.
PATROL writes some of the information to a task object (Capture Unix Diagnostics) in the UNIX OS container. PATROL writes the entire contents of all the log files that you selected to the .tar file that you specified.

Appendix A

Debugging and Running Diagnostics

387

Collector Debug Information

Collector Debug Information

This PATROL Knowledge Module for Unix and Linux gathers debug information for application collectors. This debugging information is available only for collectors that employ the PSL collection method. The UNIX OS tasks related to the debugging collectors for various application classes are described in detail on the following pages: Task Debugging FILESYSTEM Data Collection (DFColl) Debugging NETWORK Data Collection (NETColl) Debugging NFS Data Collection (NFSColl) Debugging PRINTER Data Collection (PRINTERColl) Debugging PROCESS Data Collection (PSColl) Debugging USER and PROCESS Data Collection (USRPROCColl) Debugging CPU, MEMORY, and KERNAL Data Collection (VMColl) Debugging DCM Data Collection (DCMColl) Page 388 389 390 391 392 392 393 394

Debugging FILESYSTEM Data Collection (DFColl)

This task describes how to turn the debugging feature on and off for the DFColl parameter, which gathers information for the FILESYSTEM application class.

388

PATROL Knowledge Module for Unix User Guide

Collector Debug Information

Before You Begin

This debug command provides the following information for each monitored file system instance
s s s

rule state of all instances the debug command used to collect information the information returned by the command for each instance file system blocks file system block size file system blocks available total blocks available blocks used blocks file system block capacity represented as an integer and a decimal number of files in the file system number of used i-nodes file system i-node capacity represented as an integer and a decimal

To Turn Debugging On/Off for DFColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle DFColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 DFColl debug set to TRUE CCYY0621151414 DFColl debug set to FALSE

When the debug feature is turned on, PATROL writes the debug information to the system output window when the collector runs.

Debugging NETWORK Data Collection (NETColl)

This task describes how to turn the debugging feature on and off for the NETColl parameter, which gathers information for the NETWORK application class.

Appendix A

Debugging and Running Diagnostics

389

Collector Debug Information

To Turn Debugging On/Off for NETColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle NETColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 NETColl debug set to TRUE CCYY0621151414 NETColl debug set to FALSE

When the debug feature is turned on, PATROL writes the debug information to the system output window when the collector runs.

Debugging NFS Data Collection (NFSColl)

This task describes how to turn the debugging feature on and off for the NFSColl parameter, which gathers information for the NFS (network file system) application class.

Before You Begin

This debug command provides the following information for the network file system.
s

s s s s

call information for server rpc: connectionless and connection-oriented server nfs: versions 2 and 3 server nfs_acl: versions 2 and 3 client rpc: connectionless and connection-oriented server nfs_acl: version 2 and 3 values of the last nfs_stat command cross-reference list between file systems delta data values of the current nfs_stat command

390

PATROL Knowledge Module for Unix User Guide

Collector Debug Information

To Turn Debugging On/Off for NFSColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle NFSColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 NFSColl debug set to TRUE CCYY0621151414 NFSColl debug set to FALSE

When the debug feature is turned on, PATROL writes the debug information to the system output window when the collector runs.

Debugging PRINTER Data Collection (PRINTERColl)

This task describes how to turn the debugging feature on and off for the PRINTERColl parameter, which gathers information for the PRINTER application class.

To Turn Debugging On/Off for PRINTERColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle PRINTERColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 PRINTERColl debug set to TRUE CCYY0621151414 PRINTERColl debug set to FALSE

When the debug feature is turned on, PATROL writes the debug information to the system output window when the collector runs.

Appendix A

Debugging and Running Diagnostics

391

Collector Debug Information

Debugging PROCESS Data Collection (PSColl)

This task describes how to turn the debugging feature on and off for the PSColl parameter, which gathers information for the PROCESS application class.

Before You Begin

This debug command provides the following information for the processes running on the system.
s s s s s s s s

machine type and operating system level the Unix list active processes (ps) command output results from previous Unix active process command list of process by process ID size of the processes runtime of the processes number of total processes number of shared processes

To Turn Debugging On/Off for PSColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle PSColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 PSColl debug set to TRUE CCYY0621151414 PSColl debug set to FALSE

When the debug feature is turned on, PATROL writes the debug information to the system output window when the collector runs.

Debugging USER and PROCESS Data Collection (USRPROCColl)

This task describes how to turn the debugging feature on and off for the USRPROCColl parameter, which gathers information for the PROCES and USER application classes.

392

PATROL Knowledge Module for Unix User Guide

Collector Debug Information

Before You Begin

This debug command provides the following information for users and the processes that they are running on the system.
s s s

nonroot logins nonroot users with processes user process statistics number of user processes average number of processes per user total number of processes number of zombie processes

To Turn Debugging On/Off for USRPROCColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle USRPROCColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 USRPROCColl debug set to TRUE CCYY0621151414 USRPROCColl debug set to FALSE

When the debug feature is turned on, PATROL writes the debug information to the system output window when the collector runs.

Debugging CPU, MEMORY, and KERNAL Data Collection (VMColl)

This task describes how to turn the debugging feature on and off for the VMColl parameter, which gathers information for the CPU, MEMORY, and KERNEL application classes.

Appendix A

Debugging and Running Diagnostics

393

Collector Debug Information

Before You Begin

This debug command provides the following information for the systems cpu, memory, and kernel using the vmstat command.
s s s s s s

the number of running, blocked, and waiting processes swap and free memory sizes paging statistics disk statistics faults statistics cpu statistics

To Turn Debugging On/Off for VMColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle VMColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
20010621151414 VMColl debug set to TRUE 20010621151414 VMColl debug set to FALSE

When the debug feature is turned on, PATROL writes the debug information to the system output window when the collector runs.

Debugging DCM Data Collection (DCMColl)

This task describes how to turn the debugging feature on and off for the DCMColl parameter.

NOTE
This task does not provide debugging information for the DCM collection method. Rather, it provides debugging information for the PSL collector that monitors the DCM collection method and that populates the DCM parameters.

394

PATROL Knowledge Module for Unix User Guide

Application Discovery Debug Information

Before You Begin

The command is a toggle, which means that when you select it, PATROL sets the collectors debug state to the opposite of its current state. If debug is on, selecting this command turns it off. If debug is off, selecting this command turns it on.

To Turn Debugging On/Off for DCMColl 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Debug and Diagnostics => Toggle DCMColl Debug.

Depending upon the collectors current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 DCMColl debug set to TRUE CCYY0621151414 DCMColl debug set to FALSE

Application Discovery Debug Information

This PATROL Knowledge Module for Unix and Linux gathers debug information for application discovery. This debugging information is available only for a subset of the application classes that employ the PSL collection method. The UNIX OS tasks related to the debugging discovery for various applications are described in detail on the following pages: Task Debugging DISK Discovery Process Debugging FILESYSTEM Discovery Process Debugging NETWORK Discovery Process Debugging PRINTER Discovery Process Debugging SMP Discovery Process Debugging SWAP Discovery Process Page 396 396 397 398 398 399

Appendix A

Debugging and Running Diagnostics

395

Application Discovery Debug Information

Debugging DISK Discovery Process

This task describes how to turn the debugging feature on and off for the DISK application discovery process.

Before You Begin

This debug command provides the following information about the DISK discovery process.
s

access rates reads per second writes per second amount of data processed kilobytes read per second kilobytes written per second transactions service average number of transactions waiting for service average number of transactions actively being serviced average service time in milliseconds

To Turn Debugging On/Off for DISK Discovery 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Knowledge Module Admin => DISK Application => Toggle DISK Disc. Debug.
Depending upon the current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 DISK debug set to TRUE CCYY0621151414 DISK debug set to FALSE

Debugging FILESYSTEM Discovery Process

This task describes how to turn the debugging feature on and off for the FILESYSTEM application discovery process.

396

PATROL Knowledge Module for Unix User Guide

Application Discovery Debug Information

Before You Begin

This debug command provides the following information about the FILESYSTEM discovery process.
s s

mounted systems file system instances name PATROL object labels PATROL object rule state file system type machine type options

To Turn Debugging On/Off for FILESYSTEM Discovery 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Knowledge Module Admin => FILESYSTEM Application => Toggle

FILESYSTEM Disc. Debug.

Depending upon the current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 FILESYSTEM debug set to TRUE CCYY0621151414 FILESYSTEM debug set to FALSE

Debugging NETWORK Discovery Process

This task describes how to turn the debugging feature on and off for the NETWORK application discovery process.

To Turn Debugging On/Off for NETWORK Discovery 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Knowledge Module Admin => NETWORK Application => Toggle NETWORK
Disc. Debug.

Appendix A

Debugging and Running Diagnostics

397

Application Discovery Debug Information

Depending upon the current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 NETWORK debug set to TRUE CCYY0621151414 NETWORK debug set to FALSE

Debugging PRINTER Discovery Process

This task describes how to turn the debugging feature on and off for the PRINTER application discovery process.

To Turn Debugging On/Off for PRINTER Discovery 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Knowledge Module Admin => PRINTER Application => Toggle PRINTER Disc.
Debug.

Depending upon the current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 PRINTER debug set to TRUE CCYY0621151414 PRINTER debug set to FALSE

Debugging SMP Discovery Process

This task describes how to turn the debugging feature on and off for the SMP application discovery process.

To Turn Debugging On/Off for SMP Discovery 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Knowledge Module Admin => SMP Application => Toggle SMP Disc. Debug.

398

PATROL Knowledge Module for Unix User Guide

Application Discovery Debug Information

Depending upon the current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 SMP debug set to TRUE CCYY0621151414 SMP debug set to FALSE

Debugging SWAP Discovery Process

This task describes how to turn the debugging feature on and off for the SWAP application discovery process.

To Turn Debugging On/Off for SWAP Discovery 1 Access the UNIX OS application menu as described in Accessing KM Commands
and InfoBoxes on page 402.

2 Select Knowledge Module Admin => SWAP Application => Toggle SWAP Disc.
Debug.

Depending upon the current debug state, PATROL writes one of the following messages to the system output window. TRUE means that debug is turned on. FALSE means that debug is turned off.
CCYY0621151414 SWAP debug set to TRUE CCYY0621151414 SWAP debug set to FALSE

Appendix A

Debugging and Running Diagnostics

399

Application Class Debug Information

Application Class Debug Information

This PATROL Knowledge Module for Unix and Linux gathers debug information for some application classes. This debugging information is available only for a subset of the application classes. The tasks related to debugging various applications are described in detail on the following pages: Task Debugging the SCHEDULER Application Page 400

Debugging the SCHEDULER Application

This task provides information about how to collect PSL debugger data for the SCHEDULER application. For more information about PSL debugger information, see the PATROL Script Language Reference Manual Volume 1PSL Essentials, or the PATROL Scheduler PSL Debugger dialog box online Help topic.

To Collect PSL Debugger Data for the SCHEDULER Application 1 Access the SCHEDULER application menu as described in Accessing KM
Commands and InfoBoxes on page 402.

2 Select Scheduling Debug from the pop-up menu.

The PATROL Scheduler PSL Debugger dialog box is displayed.

3 Select the check boxes for the PSL debug options that you want to turn on, and
click Accept.

NOTE
The left side of the dialog box controls debug options for the scheduler daemon. The right side of the dialog box controls debug options for the graphical user interface (GUI). Debug information is only sent to the PATROL Agent Console window when the area being debugged is active. The PATROL KM for Unix does not generate debug data when the scheduler daemon is sleeping, or the GUI is not being used. The left side PATROL begins collecting debug data for the options that you selected. Debug information is displayed in the System Output Window.

400

PATROL Knowledge Module for Unix User Guide

Appendix

Loading, Unloading, and Navigating KMs in the PATROL Consoles

B

BMC Software, Inc. offers several PATROL consoles from which you can view a PATROL Knowledge Module (KM). Due to the environments in which these consoles run, each one uses different methods to display and access information in the KM. This appendix provides instructions on how to access the KM menu commands and InfoBoxes, and how to access the online Help for the KM using each of the PATROL consoles. Accessing KM Commands and InfoBoxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Accessing Online Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Using the Mouse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Using the Mouse to Perform Tasks in PATROL . . . . . . . . . . . . . . . . . . . . . . . . . . . 402 402 404 405

Appendix B

Loading, Unloading, and Navigating KMs in the PATROL Consoles

401

Accessing KM Commands and InfoBoxes

Accessing KM Commands and InfoBoxes

Table 65 provides information about how to access KM commands and InfoBoxes in the various PATROL consoles. Table 65 Console Accessing KM Commands and InfoBoxes To access menu commands To access InfoBoxes Right-click the application icon to display its pop-up menu and then choose InfoBox. In either Tree View or Work Area, right-click a PATROL object and choose InfoBox from the pop-up menu.

PATROL Console for Microsoft Right-click the application icon Windows and choose KM Commands. PATROL Console for Unix PATROL Central - Microsoft Windows Edition

Right-click the application icon. Click MB2 on the instance icon. From an applications page, click KM Commands.

PATROL Central - Web Edition Right-click the application icon From the Monitored Systems and choose Knowledge Module page, click the system that you Commands. want, and click the Infobox tab to view the InfoBox for that system. To view the InfoBox for an application, drill down into a Monitored System until you can click the application instance that you want, and click the Infobox tab.

Accessing Online Help

Table 66 on page 403 provides information about how to access Help on each console.

NOTE
If you are trying to access Help from a Unix console, you may want to refer to the PATROL Installation Guide for specific instructions on how to install and setup a browser in the Unix environment.

402

PATROL Knowledge Module for Unix User Guide

Accessing Online Help

Table 66 Console

Accessing Online Help (Part 1 of 2) To access product help Right-click the PATROL for Unix and Linux application icon and choose KM Commands => Product Help. From the console menu bar, choose Help => Help Topics => PATROL Knowledge Modules. To access application class help Double-click the application class in the KM tab of the console. Then click the Help tab and Show Help in the Application Properties dialog box. To access parameter help Right-click the parameter icon, and choose Help On from the pop-up menu. Double-click the parameter icon; then click the ? icon or Help button in the parameter display window. Double-click the parameter in the KM tab of the console; then click the Help tab and Show Help in the properties dialog box.

PATROL Console for Microsoft Windows

PATROL Console for Unix

Right-click the PATROL for Unix and Linux application icon and choose KM Commands => Product Help. From the console menu bar, choose Help On => Knowledge Modules.

Right-click the Choose Attributes => Application Classes and parameter icon, and click Help On. double-click the application name. Then click Show Help in the Application Definition dialog box.

Appendix B

Loading, Unloading, and Navigating KMs in the PATROL Consoles

403

Using the Mouse

Table 66 Console

Accessing Online Help (Part 2 of 2) To access product help To access application class help From the Operator tab of the console, expand or collapse the list to find the application class. Click the Whats this? Help button and then To access parameter help Double-click the parameter icon to open its display window. Click the Help button.

PATROL Central - From the console menu Microsoft Windows bar, choose Help => Edition Help Topics. Then double-click the name of your KM.

click the application class name. PATROL Central Web Edition Click the Help link, and Click the link for the Click the link for the then click the link for application class, and parameter, and then your KM. then click the Page Help click the Page Help link. link.

Using the Mouse

PATROL assumes that you are using a two-button mouse. On some workstations that have a three-button mouse, clicking the middle mouse button is equivalent to clicking both buttons of a two-button mouse simultaneously. The instructions in this book are written from a right-hand mouse perspective. If you have a left-hand mouse, or have reconfigured your mouse, you will need to make the appropriate substitutions when you are following the instructions in this book. Figure 70 illustrates the typical right- and left-handed mouse configurations. Figure 70 Mouse Button Configuration

404

PATROL Knowledge Module for Unix User Guide

Using the Mouse

Using the Mouse to Perform Tasks in PATROL

You can use the mouse to perform various tasks within the PATROL KM for Unix. The following list provides information about which mouse button(s) should be used to perform several common tasks.
s

Push-buttons are activated by clicking the left mouse button (MB1). InfoBoxes are displayed by clicking the left and right mouse buttons simultaneously, or by clicking the middle mouse button (MB2) on a three-button mouse. Menus are activated by clicking the right mouse button (MB3).

NOTE
PATROL creates menus dynamically, so there may be a delay when you access menus. Click and hold down MB3 until the pop-up menu is displayed.

Appendix B

Loading, Unloading, and Navigating KMs in the PATROL Consoles

405

Using the Mouse

406

PATROL Knowledge Module for Unix User Guide

Appendix

C
408 409 410 411 411 410

Regular Expressions
This appendix describes how to use regular expressions in the context of the PATROL Knowledge Module for Unix and Linux. Not all components of the PATROL KM for Unix and Linux support regular expressions. This appendix lists the components that support regular expressions, defines the regular expression character set for PATROL KM for Unix and Linux, and provides some examples. The following sections appear in this appendix. Components that Use Regular Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Examples. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conventions for Using Regular Expressions with PATROL Objects . . . . . . . . . . . . . Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Examples. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Appendix C

Regular Expressions

407

Components that Use Regular Expressions

Components that Use Regular Expressions

The following components of the PATROL KM for Unix and Linux use regular expressions.
s

PRINTER exclusion list SECURITY List SUID and SGID Permission s File Name field s File Owner field s File Group field

408

PATROL Knowledge Module for Unix User Guide

Characters

Characters
You can use the following special characters when creating a regular expression.

NOTE
In PATROL KM for Unix and Linux, the parenthesis ( ) and pipe | characters are paired with a backslash \. This pairing varies from the standard regular expression character set.

Table 67 Character . (period)

Regular Expression Characters Description matches any single character matches zero or more instances of the previous pattern item matches one or more instances of the previous pattern item matches zero or one instances of the previous pattern item

* (asterisk) + (plus sign) ? (question mark)

\(\ ) (backslash and groups a sub-pattern parenthesis) The repetition and alternation operators apply to the whole proceeding subpattern. \| (backslash and vertical bar or pipe) [ ] (brackets) provides an alternative This character functions similarly to a logical OR. delimits a set of characters Ranges are specified as [x-y]. If the first character in the set is ^, then there is a match if the remaining characters in the set are not present. ^ (caret) anchor the pattern to the beginning of the string This character applies only when first. $ (dollar sign) anchor the pattern to the end of the string This character applies only when last.

Appendix C

Regular Expressions

409

Characters

Examples
The following examples illustrate how to use regular expressions.

any two-character string

..matches any two characters

upper and lower case

[Ww]arningmatches Warning or warning.

exclusion
[^a-zA-Z]matches anything except uppercase and lowercase letters.

a character followed by zero or more occurrences of a character

st*matches a string that contains an s followed by zero or more ts.

repeated sequences
\(st\)+matches a string that has one or more sequences of st.

anything
.*matches anything.

word regardless of case

warning\|Warningmatches warning or Warning.

space or tabs that precede a string

^[ \t]+matches any string that begins with spaces or tabs.

410

PATROL Knowledge Module for Unix User Guide

Conventions for Using Regular Expressions with PATROL Objects

Conventions for Using Regular Expressions with PATROL Objects

In regular expressions, when specifying PATROL objects (/computer:/application/instance/parameter) or pathnames (/directory/subdirectory/subdirectory/file), remove the initial slash and substitute dashes for the remaining slashes.

Format
For a file, the format for a regular expression is
directory-subdirectory-subdirectory-file

For a PATROL object, such as a parameter, the format for a regular expression is
computer:-application-instance-parameter

EXAMPLE
The file system /app/oracle/usr would be represented as app-oracle-usr in a regular expression.

Examples
The following examples illustrate how to use regular expressions with PATROL objects and pathnames.

a directory or object
tmpmatches tmp directory

directories or objects with a similar name

usr[1-9].*matches usr1 through usr9

Appendix C

Regular Expressions

411

Conventions for Using Regular Expressions with PATROL Objects

subdirectories
etc-testmatches directory or file system /etc/test

objects with similar names in object hierarchies

nerd:-DISK-sb[0-9].*matches any disk with a prefix of sb0 through sb9 on the computer nerd PRINTER-.*[Hh]ouston.*matches any printer with the word Houston, capitalized or in all lowercase, in its name

412

PATROL Knowledge Module for Unix User Guide

Appendix

D
414 414 415 415 415 416

Inclusion and Exclusion Lists

The PATROL Knowledge Module for Unix and Linux gives you the ability to determine which instance PATROL monitors for the DISK, FILESYSTEM, and PRINTER applications. The following sections appear in this chapter. Using Exclusion and Inclusion Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods for Creating Exclusion and Inclusion Lists . . . . . . . . . . . . . . . . . . . . . . . Storage Location of Exclusion and Inclusion Lists . . . . . . . . . . . . . . . . . . . . . . . . . Benefits and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benefits of Creating Exclusion Lists Using Menu Commands . . . . . . . . . . . . . . . Inclusion List Limitations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Appendix D

Inclusion and Exclusion Lists

413

Using Exclusion and Inclusion Lists

Using Exclusion and Inclusion Lists

PATROL is designed to monitor all the instances that it discovers. However, some application classes, such as DISKS, FILESYSTEMS, and PRINTERS, allow the user to select which instances PATROL monitors. PATROL provides this feature through exclusion and inclusion lists. The exclusion list is a mechanism for ignoring a small number of instances. If you want PATROL to ignore a majority of instances, you can create an inclusion list, which forces PATROL to ignore all but the instances that you specify.

Methods for Creating Exclusion and Inclusion Lists

PATROL provides three different ways to create exclusion and inclusion lists and subsequently create and populate the associated configuration variables. You can create the list by using
s

the Stop Monitoring menu command from DISK, FILESYSTEM, or PRINTER instances (exclusion list only) the Edit <application> Exclusion List menu command from DISK, FILESYSTEM, or PRINTER instances (exclusion list only) pconfig tool

NOTE
When using xpconfig or wpconfig to create an exclusion list, add both entries (pathname-instance and /pathname/instance) for each instance that you intend to exclude. For more information, see Benefits of Creating Exclusion Lists Using Menu Commands on page 415.

414

PATROL Knowledge Module for Unix User Guide

Benefits and Limitations

Storage Location of Exclusion and Inclusion Lists

PATROL uses variables in the PATROL configuration database to store information about exclusion and inclusion lists. The two variables are
s

/AgentSetup/application.filterListcontains the list of instances to include or exclude depending upon the value of application.filterType

The format is a comma-separated list of instances or regular expressions. The variable has a limit of 1024 characters. Regular Expressions can be used to alleviate this limitation. The error: Exclusion list too long. Please use regular expression exclusion. is generated when a filter exceeds the maximum size allowed for an Agent configuration variable.
s

/AgentSetup/application.filterTypedetermines whether the list is an exclusion or

inclusion If application.filterType does not exist, the list defaults to exclude. Valid values are exclude include

Benefits and Limitations

The exclusion list feature is designed for ignoring a small number of instances. If you want PATROL to ignore a majority of instances, you can create an inclusion list, which forces PATROL to ignore all but the instances that you specify. However, monitoring by inclusion has a few limitations.

Benefits of Creating Exclusion Lists Using Menu Commands

BMC Software recommends that you use the menu options Stop Monitoring and Edit <application> Exclusion List to exclude application instances. These menu options are accessed from the pop-up menu of the instances.

Appendix D

Inclusion and Exclusion Lists

415

Benefits and Limitations

The menu commands add two entries into the /AgentSetup/application.filterList for each instance excluded.
s

The first entry is in the form pathname-instance, where slashes are replaced by dashes and the initial slash is omitted.

EXAMPLE
mnt-hpserv-users indicates the file system /mnt/hpserv/users.

This entry is used by the collector in setting the consumer variables.

s

The second entry is in the form /pathname/instance.

EXAMPLE
/mnt/hpserv/users indicates the file system /mnt/hpser/users.

This entry is used to display the instance in the <application> Exclusions dialog box. If you omit this entry, the instance will not appear in the dialog box.

Inclusion List Limitations

Monitoring application instances by using an inclusion list disables some of the instances features and requires the user to perform additional tasks.
s

PATROL will not automatically monitor any new instances.

NOTE
The user assumes full responsibility for maintaining this list.

The Stop Monitoring menu command on an application will not work. Initially, the instance will disappear from the list of monitored instances. However, the instance reappears during the next discovery cycle.

416

PATROL Knowledge Module for Unix User Guide

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Index
Symbols
/snmp/default_port variable 369 /SNMP/default_r_community variable 366 /snmp/piV1m_list variable 369, 370 /snmp/trap_port variable default value 364

A
accessing InfoBoxes 405 menu commands 405 push-buttons 405 accounts Network Information System (NIS), listing 218 activating application classes 100 parameters 100 PRINTER application class 225 Address Resolution Protocol (ARP) cache, adding hosts 205 cache, deleting hosts 203 cache, viewing 202 addresses host, listing 187 host, output format 188 internal state 184 IP, listing 187 IP, output format 188 Network IP, determining 191 network, output format 185 network, viewing 184 remote, viewing 184 ALL_COMPUTERS application class parameters 50 application classes activating 100 COMPOSITE, hierarchy 118 description 22 HEALTH AT A GLANCE 158 icons 22 NFS 214 PRINTER, activating 225 PRINTER, object hierarchy 224 PROCESS, object hierarchy 244 PROCESS_PRESENCE, object hierarchy 244 SCHEDULER, object hierarchy 312 SECURITY, object hierarchy 340 SNMPHealth, object hierarchy 352 application menus

COLLECTOR 33 COMPOSITES 33 COMPOSITES_COLLECTOR 33 Computer Instance 30 CPU 34 DCM 34 DISK 34 FILESYTEM 35 HEALTH AT A GLANCE 36 KERNEL 36 MEMORY 36 NETWORK 37 NFS 38 PATROL 38 PATROLAGENT 39 PRINTER 41 PROCCONT 42 PROCESS_PRESENCE (PROCCONT) 42 PROCPRES 43 SMP 44 SNMPHealth 45 SWAP 46 UNIX OS 31 USERS 47

B
bang path, Unix to Unix Communication Protocol (UUCP) 206 blackout cancelling a scheduled script 337 creating a script 327 deleting a script 334 modifying a script 331 specifying a script default directory 326 viewing a script 330 BMC Software, contacting 2 build date viewing, kernel 112

C
cache Address Resolution Protocol (ARP), adding hosts 205 Address Resolution Protocol (ARP), deleting hosts 203 Address Resolution Protocol (ARP), viewing 202 CD-ROM systems exclusion from file system monitoring 140 collecting PSL debug data 322325 COLLECTORS application class data collection methods 88 icons 22 menus 33

Index

417

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
object hierarchy 88 parameters 51 commands File Transfer Protocol (ftp), supported by PATROL 198 msu attempts, listing failed 345 Ping, using in PATROL 199 rlogin 200 set user (su) attempts, listing failed 345 Stop Monitoring 144 COMPOSITE application class object hierarchy 118 composite parameters creating 118 creating expressions 119 creating expressions manually 124 creating expressions with the wizard 121 expressions, deleting 126 expressions, editing 125 expressions, examples 120 expressions, syntax rules 120 icon 119 instances 118 object hierarchy 118 relational operators 118 setting the status 121 COMPOSITES application class menus 33 object hierarchy 118 parameters 52 COMPOSITES_COLLECTOR application class icons 22 menus 33 parameter icons 22 parameters 52 Computer Instance application class icons 22 menus 30 computer systems trusted 198 configuration mouse 404 PATROL SNMP, listing 356 PATROL SNMP, saving in ASCII file 357 configuring activating application classes 100 activating parameters 100 context switches, CPU statistics 109 counters, resetting for NFS servers 217 CPU detection 108 hogs, list output format 297 hogs, viewing 295 multi-processor, viewing statistics 110 processes using excessive amounts, viewing 295 viewing, top ten processes using 297 CPU application class icons 22 menus 34 object hierarchy 106 parameters 52 CPU usage monitoring, overall system usage 159 CPU utilization idle time, viewing 108 interrupts, viewing data 109 monitoring 106 monitoring, detection 108 multi-processor, statistics 110 percentage used 108 run queue, viewing 109 switches, viewing 109 system mode, viewing 108 user time, viewing 108 viewing 108 waits, viewing 109 cross-platform monitoring 216 customer support 3

D
Data Collection Manager (DCM) data collection methods 91 data collection methods application classes 88 changing 98 COLLECTORS application class 88 Data Collection Manager (DCM) 91 DCM 90, 92 DCM application class 89 DCM, parameters 91 determining which is used 98 kernel-level data 91 managing 98 object hierarchies 88 PATROL Perform Collection 91 PATROL Perform collector 91 PATROL PSL 95 PSL parameters, activated 95 Unix utilities 92, 95 dblk statistics, viewing 181 DCM collection method 90 DCM application class data collection methods 89 icons 22 menus 34 object hierarchy 89 parameters 53 DCM collection method gathering data 92 DCM data collection deactivated parameters 93

418

PATROL Knowledge Module for Unix User Guide

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
parameters 91 selecting 98 deactivated parameters DCM data collection 93 PSL data collection 95 debug data 322 deleting files, by characteristics 153 files, by date updated 155 files, by group 155 files, by name 155 files, by owner 155 files, on single physical disk 155 files, output format 156 semaphores 168 shared-memory segments 166 destination networks, listing 192 device names, file system mounting 148 devices network, obtaining information 202 directories Unix to Unix Communication Protocol (UUCP) 206 discovery mounted disks 128 network interface, name 179 DISK application class icons 22 menus 34 object hierarchy 128 parameters 56 disk monitoring excluding 130 exclusion list 414 inclusion list 414 resuming 132 resuming monitoring 131 resuming on all disks 133 selecting disks 131 disks discovery 128 files, removing on single disk 155 synchronizing file system cache 149 composite parameters, syntax 120 composite parameters, syntax rules 120 creating manually, composite parameters 124 deleting, composite parameters 126 regular, disk monitoring 131 regular, file searches 151 regular, file system monitoring 139

F
file system capacity managing 141 file system cleanup automating 141 recovery actions, attended mode 143 file system monitoring CD-ROM exclusion 140 exclusion by type 140 exclusion list 414 exclusion lists 138 inclusion 139 inclusion list 414 NFS systems, excluding 140 process control systems, exclusion 140 regular expressions 139 file system searches output format 152 File Systems exclude from being monitored 144 monitoring 144 file systems cache, synchronizing to disk 149 device names, mounting 148 monitoring, overall usage 159 mounting 146 mounting, automatic 147 mounting, manual 147 mounting, paths 148 NFS servers, listing 216 NFS, resetting counters 217 searching 150 searching, on single physical disk 151 unmounting 148 unmounting, automatically 148 unmounting, manually 149 File Transfer Protocol (ftp) commands, supported by PATROL 198 using with PATROL 198 files copying, File Transfer Protocol (ftp) 198 copying, Unix to Unix Communication Protocol (UUCP) 208 deleting based on characteristics 153 deleting, based on name 155 deleting, by date updated 155 deleting, by group 155

E
entries viewing statistics 114 exclusion list 414 file systems 138 exclusion lists disks, monitoring 130 expressions composite parameter wizard 121 composite parameters, creating 119 composite parameters, editing 125 composite parameters, examples 120

Index

419

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
deleting, by owner 155 deleting, on single physical disk 155 global-write privilege, viewing 344 locating 150 locating, by date updated 151 locating, by file name 151 locating, by group 151 locating, by owner 151 locating, on single physical disk 151 locating, output format 152 locating, SGID-protected 341 locating, SUID-protected 341 removing, output format 156 SGID-protected, listed output format 343 SGID-protected, listing 341 SUID-protected, listed output format 343 SUID-protected, listing 341 transferring, File Transfer Protocol (ftp) 198 Unix to Unix Communication Protocol (UUCP) 206 Unix to Unix Communication Protocol (UUCP) log, viewing 210 Unix to Unix Communication Protocol (UUCP), deleting 207 FILESYSTEM application class icons 23 menus 35 object hierarchy 137 parameters 58 filesystem table, recording mounts 148 hosts adding to Address Resolution Protocol (ARP) cache 205 deleting from Address Resolution Protocol (ARP) cache 203 trusted, listing 198 Unix to Unix Communication Protocol (UUCP), listing 210

I
I/O operations CPU statistics, waiting 109 icons application classes 22 composite parameter instance 119 inclusion list 414 InfoBoxes accessing with the mouse 405 i-nodes entries, viewing files located 114 slots, viewing used 114 slots,viewing statistics 114 viewing conversions 114 interrupts, CPU statistics 109 IP address obtaining 202 IP addresses determining 191 listing 187 output format 188 IPV4 protocol, viewing statistics 191

G
global write privilege files, listing 344 group process ID listing nice value of processes 301 groups Network Information System (NIS), list output 220 Network Information System (NIS), listing 219

J
job modifying a scheduled script 321 removing a scheduled script 321 scheduling a command interpreter script 315 scheduling a PSL script 318 scheduling an operating system command script 313 jobs print, viewing for one printer 231 printer, deleting all from a print queue 234 printer, moving all jobs 236 printer, moving one job 235 printers, listing for system 232

H
HEALTH AT A GLANCE application class 158 icons 23 menus 36 object hierarchy 158 parameters 58 help accessing in a PATROL Console for Windows 26 accessing in PATROL Central - Web Edition 26 accessing in PATROL Central - Windows Edition 26 accessing in PATROL Console for Unix 26 online 25 host addresses listing 187 output format 188

K
kernel build date, viewing 112 dynamically allocated memory 113 memory allocation, viewing 113 memory requests 113

420

PATROL Knowledge Module for Unix User Guide

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
monitoring resources 106 monitoring, detection 112 viewing system usage 115 KERNEL application class icons 23 menus 36 object hierarchy 111 parameters 59 kernel data active semaphores 115 directory block reads 115 message operations per second 115 percentage of file slots used 115 percentage of used lock slots 115 percentage of used process slots 115 system calls 115 kernel memory allocation (KMA), viewing 113 kernel-level data collection methods 91 km plug-in 304 knowledge module plug-in 304 memory See also virtual memory semaphores, viewing statistics 167 MEMORY application class icons 23 menus 36 object hierarchy 163 parameters 62 memory buffer allocation failures, viewing 181 memory buffers statistics, output format 183 statistics, viewing 181 memory requests categories 113 large, viewing allocated amount 113 large, viewing available amount 113 large, viewing failed amounts 113 oversized, viewing allocated amounts 113 small, viewing allocated amounts 113 small, viewing available amounts 113 small, viewing failed amounts 113 viewing failed amounts 113 menu commands accessing with the mouse 405 menus COLLECTOR application 33 COMPOSITES application 33 COMPOSITES_COLLECTOR application 33 Computer Instance application 30 CPU application 34 DCM application 34 DISK application 34 FILESYTEM application 35 HEALTH AT A GLANCE application 36 KERNEL application 36 location of KM commands 29 MEMORY application 36 NETWORK application 37 NFS application 38 PATROL application 38 PATROLAGENT application 39 PRINTER application 41 PROCCONT application 42 PROCESS_PRESENCE (PROCCONT) application 42 PROCPRES application 43 SECURITY application 44 SMP application 44 SNMPHealth application 45 SWAP application 46 UNIX OS application 31 USERS application 47 modifying a scheduled task defined by a script 321 monitoring blackout cancelling a scheduled script 337

L
large memory requests viewing allocated amounts 113 viewing available amounts 113 viewing failed amounts 113 linkblk statistics, viewing 181 locating files, by date updated 151 files, by group 151 files, by owner 151 files, on single physical disk 151 files, search output format 152 locating, files 150 log files Unix to Unix Communication Protocol (UUCP), viewing 210 logins remote, using with PATROL 200

M
Machine Access Control (MAC) obtaining 202 Management Information Base (MIB) checking SNMP information 357 managing data collection 98 file system capacity 141 semaphores 167 mblk statistics, viewing 181

Index

421

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
creating a script 327 deleting a script 334 modifying a script 331 specifying a default directory for scripts 326 viewing a script 330 CPU usage, overall for system 159 CPU utilization 106 CPU utilization, viewing statistics 108 cross-platform 216 disks, excluding 130 disks, resuming for all 133 disks, selecting 131 file system usage, overall for system 159 file systems, excluding 138 file systems, selecting 139 kernel resources 106 kernel, detection 112 network data, activating 176 overall system health 158 poll time, parameters 92 printers, excluding 227 printers, including 227 printers, jobs in queue 228 processes waiting on resources 298 processes, configuring a process to alarm when its PPID is not 1 262 processes, configuring a process to alarm when its PPID is owned by unauthorized user 265 processes, configuring a process to alarm when number exceeds set limit 250 processes, configuring a process to alarm when number falls below set limit 253 processes, configuring a process to stop 259 processes, configuring a process to suspend alert for n cycles if not running 268 processes, configuring communication for a process host group 284 processes, configuring multiple instances of 247 processes, configuring to restart a process when it dies 256 processes, creating a host group 283 processes, deleting a process from the monitoring queue 278 processes, modifying host attributes in a process host group 285 processes, modifying the configuration of a process 270 processes, non-root 298 processes, querying a host within a process host group 286 processes, removing a host from a process host group 286 processes, reporting on monitored processes 280 processes, resuming monitoring for a suspended process 280 processes, specifying a flag file directory 279 processes, suspending monitoring for a process 280 processes, synchronizing a local hosts list to a selected hosts list for a process host group 288 processes, synchronizing a local processs list with a selected processs list 289 processes, synchronizing a selected hosts list to a local hosts list for a process host group 288 processes, synchronizing a selected processs list with the local processs list 290 processes, using most CPU 295 processes, viewing the process list for a selected host 289 processes, viewing the process presence list for a selected process 291 resuming for a single disk 132 status, parameters 92 suspend cancelling a scheduled blackout script 337 creating a blackout script 327 deleting a blackout script 334 modifying a blackout script 331 specifying a default directory for blackout scripts 326 viewing a blackout script 330 swap space usage, overall for system 159 total processes running 295 zombie processes, viewing total on system 292 mounting file systems 146 automatically 147 device names 148 filesystem table 148 manually 147 mouse usage 404 accessing InfoBoxes 405 accessing menu commands 405 accessing push-buttons 405 msu command failed execution, list of users 345 multiple sessions running, listing users 348 multi-processor CPUs, viewing statistics 110

N
name server, querying 201 NETWORK application class activating 176 icons 23 menus 37 object hierarchy 175 parameters 66 Network File System (NFS) application class 214 counters, resetting 217 object hierarchy 215 servers, listing 216 Network Information System (NIS)

422

PATROL Knowledge Module for Unix User Guide

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
accounts, listing 218 accounts, output format 219 groups, list output 220 groups, listing 219 servers, list output format 221 servers, listing 220 network instance destroying when unavailable 181 network instances alarming when unavailable 181 PATROL behavior when unavailable 180 network interfaces name discovery 179 output format 179 viewing 178 Network IP addresses, determining 191 networks active uses per route 192 addresses, internal state 184 addresses, output format 185 addresses, viewing 184 dblk statistics, viewing 181 destination, listing 192 devices, obtaining information 202 gateways, listing 192 interface used 192 IP address, obtaining 202 linkblk statistics, viewing 181 listing trusted hosts 198 mblk statistics, viewing 181 node, physical address 202 packets sent per route 192 packets, errors 194 packets, statistics 194 protocol version, viewing 184 protocol, viewing 191 protocols, listing 189 protocols, output format 190 qband statistics, viewing 181 querying name server 201 queue statistics, viewing 181 receive queue size, viewing 184 receive window size, viewing 184 remote address, viewing 184 route designation, viewing 192 router status, viewing 192 send queue size, viewing 184 send window size, viewing 184 sockets, viewing data 184, 185 statistics, viewing 192 stream statistics, viewing 181 strevent statistics, viewing 181 subnet mask, obtaining 202 synq statistics, viewing 181 NFS application class icons 23 menus 38 parameters 66 NFS systems exclusion from file monitoring 140 nice values changing for a process 300 node physical address, obtaining 202 non-VME device interrupts, viewing 109

O
oversized memory requests viewing allocated amounts 113 viewing failed amounts 113

P
packets errors, monitoring 194 network, viewing sent 192 statistics 194 parameters activating 100 COLLECTORS application class 51 composite, creating 118 COMPOSITES application class 52 COMPOSITES_COLLECTOR application class 52 COMPUTERS application class 50 CPU application class 52 DCM application class 53 DCM collection method 91 deactivated, DCM collection method 93 deactivated, DCM data collection 93 deactivated, PSL data collection 95 default settings 8086 DISK application class 56 FILESYSTEM application class 58 HEALTH AT A GLANCE application class 58 KERNEL application class 59 MEMORY application class 62 NETWORK application class 66 NFS application class 66 PATROLAGENT application class 71 PRINTER application class 71 PROCCONT application class 71 PROCESS application class 72 PROCPRES application class 73 PSL data collection 95 SCHEDULER application class 74 SMP application class 74 SNMPHealth application class 77 SWAP application class 78 USERS application class 79 partitions, swap area size 172 passwords

Index

423

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
listing users without a password 346 paths file system mounting 148 PATROL File Transfer Protocol (ftp), supported commands 198 mouse usage 404 ping, supported commands 199 rlogin, supported commands 200 SNMP architecture 353 PATROL Agent port number, default 353 querying 366 receiving SNMP queries, testing 367 SNMP queries, testing 366 system role 354 PATROL application class icons 23 menus 38 PATROL Console for Microsoft Windows menu command location 29 PATROL Console for Unix menu command location 29 PATROL Event Manager automated SNMP traps, testing 370 PATROL Perform Agent -Ctrl parameters 92 data collection methods 91 PATROL PSL collectors -Coll parameters 92 PATROL SNMP configuration information, listing 356 configuration, saving in ASCII file 357 PATROL SNMP Manager system role 354 PATROL SNMP Master Agent port number, default 353 starting 361 system role 354 PATROL SNMP Sub-agent system roles 354 PATROLAGENT application class icons 23 menus 39 parameters 71 permissions semaphores, viewing 167 ping using with PATROL 199 plug-ins knowledge module 304 poll time monitoring parameters 92 port numbers PATROL Agent, default 353 PATROL SNMP Master Agent, default 353 ports SNMP Query, changing 355 print jobs deleting jobs

print, deleting 233 deleting all jobs in queue 234 listing for entire system 232 listing, output 232 moving all jobs 236 moving single job 235 output format 232 print queues spooling, disabling 238 starting 239 stopping 238 viewing number of jobs 228 PRINTER application class activating 225 icons 24 menus 41 object hierarchy 224 parameters 71 printer monitoring exclusion list 414 inclusion list 414 printers deleting all jobs 234 deleting jobs 233 excluding one from monitoring 227 jobs, moving all jobs 236 listing all jobs in system 232 monitoring, including 227 moving one job 235 queues, starting 239 queues, stopping 238 spooling, disabling 238 starting 229 stopping 228 viewing assigned jobs 231 PROCCONT application class menus 42 parameters 71 PROCESS application class icon 24 object hierarchy 244 parameters 72 process control subsystems excluding from monitoring 140 process groups renicing processes within 301 Process KM 304 PROCESS_PRESENCE application class icons 24 menus 42 object hierarchy 244 processes See also zombie processes active, listing 300

424

PATROL Knowledge Module for Unix User Guide

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
active, listing with group process ID 301 listing, by user ID 303 monitoring, configuring a process to alarm when its PPID is not 1 262 monitoring, configuring a process to alarm when its PPID is owned by unauthorized user 265 monitoring, configuring a process to alarm when number exceeds set limit 250 monitoring, configuring a process to alarm when number falls below set limit 253 monitoring, configuring a process to stop 259 monitoring, configuring a process to suspend alert for n cycles if not running 268 monitoring, configuring communication for a process host group 284 monitoring, configuring multiple instances of a process 247 monitoring, configuring to restart a process when it dies 256 monitoring, creating a host group 283 monitoring, deleting a process from the monitoring queue 278 monitoring, modifying host attributes in a process host group 285 monitoring, modifying the configuration of a process 270 monitoring, querying a host within a process host group 286 monitoring, removing a host from a process host group 286 monitoring, reporting on monitored processes 280 monitoring, resuming monitoring for a suspended process 280 monitoring, specifying a flag file directory 279 monitoring, suspending monitoring for a process 280 monitoring, synchronizing a local hosts list to a selected hosts list for a process host group 288 monitoring, synchronizing a local processs list with a selected processs list 289 monitoring, synchronizing a selected hosts list to a local hosts list for a process host group 288 monitoring, synchronizing a selected processs list with the local processs list 290 monitoring, viewing the process list for a selected host 289 monitoring, viewing the process presence list for a selected process 291 non-root, viewing average 299 non-root, viewing average number 299 renicing 300 reprioritizing 300 reprioritizing all within a process group 301 reprioritizing by owner 302 using excessive CPU, output format 297 using excessive CPU, viewing 295 viewing, processes waiting on resources 298 viewing, top ten consumers of CPU 297 viewing, total non-root processes running 298 viewing, total running on system 295 viewing, total user processes running 298 PROCPRES application class icons 24 parameters 73 PROCTopProcs parameter, output 298 product support 3 protocol version, viewing 184 protocols network, listing 189 network, output format 190 statistics, viewing 191 TCP/IP services, listing 196 PSL data collection -Coll parameters 96 exceptions 96 selecting 98 PSL data collection parameters, deactivated by platform 97 PSL data collection parameters, deactivated by platform 96 push-buttons, selecting 405

Q
qband statistics, viewing 181 queues deleting all jobs for single printer 234 printer, deleting a print job 233 printer, starting 239 printer, stopping 238 printer, viewing all jobs on a system 232 printer, viewing jobs 228 printer, viewing jobs for single printer 231 receive 184 send, size 184 statistics, viewing 181

R
RAWIP protocol, viewing statistics 191 receive queue size, viewing 184 receive window size, viewing 184 recovery actions attended mode 143 file system cleanup 141 regular expressions disk monitoring 131 file system monitoring 139 file systems, searching 151 relational clauses, composite parameters 118 relational operators, composite parameters 118 remote addresses viewing 184

Index

425

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
removing a scheduled task defined by a script 321 files by characteristics 153 files, by date updated 155 files, by group 155 files, by name 155 files, by owner 155 files, on single physical disk 155 files, output format 156 semaphores 168 shared memory segments 166 renicing processes 300 processes for a process group 301 processes, by owner 302 single process 300 requirements composite parameters, expression syntax 120 file systems, unmounting 148 mounting file systems 146 rlogin using with PATROL 200 route designation viewing 192 router status, viewing 192 routing table output format 193 viewing 192 run queue, viewing 109 displaying active 115 group, viewing 167 managing 167 memory types, viewing 167 numeric ID, viewing 167 owner, viewing 167 permissions, viewing 167 statistics, output format 168 statistics, viewing 167 send queue size viewing 184 send window size, viewing 184 server name, querying 201 servers Network File System (NFS), listing 216 Network Information System (NIS), list output format 221 Network Information System (NIS), listing 220 services TCP/IP, listing 196 Unix to Unix Communication Protocol (UUCP), listing 210 sessions multiple, listing users 348 set user (su) command failed execution, list of users 345 SGID-protected files, listing 341 shared memory segments statistics, output format 164 viewing statistics 164 shared memory segments, deleting 166 small memory requests viewing allocated amounts 113 viewing available amounts 113 viewing failed amounts 113 SMP application class icons 24 menus 44 parameters 74 SNMP architecture, PATROL 353 configuration tasks 354 listening for traps, testing 364 viewing information in a Management Information Base (MIB) 357 SNMP Managers determining whether listening 369 SNMP Master Agent PATROL, starting 361 traps, ability to receive 364 SNMP queries received by PATROL Agent, testing 367 SNMP Query Port changing 355 SNMP sub-agent starting 362

S
SCHEDULER application class icons 24 menus 44 object hierarchy 312 parameters 74 scheduling task defined by a command interpreter script 315 task defined by a PSL script 318 task defined by an operating system command script 313 security files with global write privilege, viewing 344 files, listing SUID-protected files 341 listing failed msu command attempts 345 listing failed set user (su) command attempts 345 listing SGID-protected files 341 listing users with multiple running sessions 348 listing users without a password 346 SECURITY application class icons 24 menus 44 object hierarchy 340 semaphores deleting 168

426

PATROL Knowledge Module for Unix User Guide

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
stopping 363 SNMP traps manual, generating 365 PATROL Event Manager, testing 370 SNMPHealth application class icons 24 menus 45 object hierarchy 352 parameters 77 sockets network, viewing data 184, 185 Solaris systems Unix to Unix Communication Protocol (UUCP) 207 spooling, disabling for a printer 238 statistics active semaphores 115 CPU interrupts, viewing 109 CPU run queue, viewing 109 CPU switches, viewing 109 CPU utilization 106 CPU utilization, viewing 108 CPU waits, viewing 109 CPU, viewing for multi-processors 110 dblk, viewing 181 kernel resources 106 linkblk, viewing 181 mblk, viewing 181 memory buffer 181 network packets 194 network protocol 191 network, viewing 192 overall CPU usage 159 overall file system usage 159 overall swap usage 159 packets with errors 194 qband, viewing 181 semaphores 167 semaphores, viewing 167 shared memory segments 164 strevent, viewing 181 swap partition size, viewing 172 swap partition space 172 swap space 171 swap space used, viewing 172 swap space, in a swap area 171 swap space, total 171 synq, viewing 181 system calls 115 viewing entries 114 viewing kernel system usage 115 viewing tables 114 viewing, i-node slots 114 virtual memory, viewing 171 status monitoring, parameters 92 setting for composite parameters 121 streams, statistics 181 strevent statistics, viewing 181 subnet mask, obtaining 202 SUID-protected files, listing 341 support, customer 3 suspend monitoring cancelling a scheduled blackout script 337 creating a blackout script 327 deleting a blackout script 334 modifying a blackout script 331 specifying a default directory for blackout scripts 326 viewing a blackout script 330 SWAP application class icons 25 menus 46 object hierarchy 170 parameters 78 swap partition, size 172 swap space overall usage, monitoring 159 percentage used, swap area 172 percentage used, total 172 percentage used, viewing 172 viewing 171 viewing, for a swap area 171 viewing, total swap space 171 swap space size swap area, viewing 172 switches, CPU statistics 109 synchronizing, file system cache and disk 149 synq statistics, viewing 181 syntax rules composite parameter expressions 120 system calls kernel data 115 system health monitoring 158 overall, viewing 160 system roles PATROL Agent 354 PATROL SNMP Manager 354 PATROL SNMP Master Agent 354 PATROL SNMP Sub-agent 354 system settings ports, SNMP Query 355 system usage viewing kernel data 115 systems copying files, File Transfer Protocol (ftp) 198 transferring files, File Transfer Protocol (ftp) 198

T
tables routing, viewing 192

Index

427

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
viewing statistics 114 task collecting debug data 322 modifying a scheduled script 321 removing a scheduled script 321 scheduling a command interpreter script 315 scheduling a PSL script 318 scheduling an operating system command script 313 TCP protocol, viewing statistics 191 TCP/IP name, listing 196 port, listing 196 protocol, listing 196 services, listing 196 services, output format 197 technical support 3 testing SNMP Master Agent, receiving traps 364 traps listening for SNMP, testing 364 manual, generating 365 PATROL Event Manager SNMP, testing 370

V
variables /PUK/FILESYSTEM/moniList 144 /snmp/default_port 369 /SNMP/default_r_community 366 /snmp/piV1m_list 369, 370 /snmp/trap_port 364 viewing 108 files, with global write privilege 344 non-root processes, average 299 PATROL SNMP configuration 356 processes waiting on resources 298 processes, non-root 298 processes, using excessive CPU 295 PROCTopProcs parameter, output format 298 SGID-protected files 341 SUID-protected files 341 system health, overall 160 top ten CPU hogs 297 top ten CPU hogs, output format 298 users without passwords 346 zombie processes on system 293 virtual memory monitoring 171 percentage used, swap area 172 percentage used, viewing 172 swap area, viewing 171, 172 swap partition size, viewing 172 swap space, total 172 total swap space, viewing 171 viewing 171

U
UDP protocol, viewing statistics 191 Unix to Unix Communication Protocol (UUCP) bang path emails 206 copying files 208 copying files, specifying source 209 deleting files, Solaris systems 207 directories 206 files, copying options 208 files, specifying destination 209 hosts, listing 210 log file, viewing 210 services, listing 210 verifying support files 206 Unix utilities data collection 92 data collection methods 95 UNIX_OS application class icons 25 menus 31 unmounting file systems, automatically 148 file systems, manually 149 users listing duplicate user ID entries 348 without passwords, listing 346 USERS application class icons 25 menus 47 parameters 79

W
waits, CPU statistics 109

Z
zombie processes listing 293 viewing total on system 292 viewing, output format 294

428

PATROL Knowledge Module for Unix User Guide

Notes

*03145* *03145* *03145* *03145*

*54130*