IBM i

7.4

Files and file systems

Integrated file system

IBM
 Note
Before using this information and the product it supports, read the information in “Notices” on page
153.

This edition applies to IBM i 7.4 (product number 5770-SS1) and to all subsequent releases and modifications until
otherwise indicated in new editions. This version does not run on all reduced instruction set computer (RISC) models nor
does it run on CISC models.
This document may contain references to Licensed Internal Code. Licensed Internal Code is Machine Code and is
licensed to you under the terms of the IBM License Agreement for Machine Code.
© Copyright International Business Machines Corporation 1999, 2019.
US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract with
IBM Corp.
Contents

Integrated file system............................................................................................1

What's new for IBM i 7.4..............................................................................................................................1
PDF file for Integrated file system...............................................................................................................1
Overview of the integrated file system........................................................................................................2
What the integrated file system is..........................................................................................................2
Why use the integrated file system........................................................................................................3
Integrated file system concepts.................................................................................................................. 4
Directory................................................................................................................................................. 4
Current directory............................................................................................................................... 6
Home directory................................................................................................................................. 6
Provided directories.......................................................................................................................... 6
*TYPE2 directories............................................................................................................................ 9
Link........................................................................................................................................................10
Hard link.......................................................................................................................................... 11
Symbolic link...................................................................................................................................12
Path name.............................................................................................................................................13
Stream file............................................................................................................................................ 14
Name continuity................................................................................................................................... 15
Extended attributes..............................................................................................................................16
Scanning support..................................................................................................................................17
Examples: Scanning for viruses and files being opened............................................................... 17
Related system values.................................................................................................................... 18
Scanning occurrences.....................................................................................................................19
Object change............................................................................................................................19
Signature change.......................................................................................................................20
Different CCSID......................................................................................................................... 21
During save operation............................................................................................................... 21
Object integrity check............................................................................................................... 21
File systems............................................................................................................................................... 21
File system comparison....................................................................................................................... 22
"root" (/) file system............................................................................................................................. 27
Case-sensitivity in the "root" (/) file system.................................................................................. 28
Path names in the "root" (/) file system......................................................................................... 28
Links in the "root" (/) file system.................................................................................................... 28
Use of integrated file system commands in the "root" (/) file system...........................................28
Use of integrated file system APIs in the "root" (/) file system..................................................... 29
Object changes journaling in the "root" (/) file system.................................................................. 29
UDP and TCP devices in the "root" (/) file system..........................................................................29
Open systems file system (QOpenSys)................................................................................................30
Case sensitivity in the QOpenSys file system................................................................................ 30
Path names in the QOpenSys file system...................................................................................... 30
Links in the QOpenSys file system................................................................................................. 31
Use of integrated file system commands and displays in the QOpenSys file system.................. 31
Use of integrated file system APIs in the QOpenSys file system.................................................. 31
Object changes journaling in the QOpenSys file system............................................................... 31
User-defined file systems (UDFSs)...................................................................................................... 32
Temporary user-defined file systems.............................................................................................33
Case sensitivity in an integrated file system user-defined file system......................................... 34
Path names in an integrated file system user-defined file system............................................... 35
Links in an integrated file system user-defined file system.......................................................... 35
Use of integrated file system commands in a user-defined file system....................................... 36

iii
 Use of integrated file system APIs in a user-defined file system..................................................36
Graphical user interface for a user-defined file system................................................................ 37
Creating an integrated file system user-defined file system......................................................... 37
Deleting an integrated file system user-defined file system......................................................... 37
Displaying an integrated file system user-defined file system......................................................38
Mounting an integrated file system user-defined file system....................................................... 38
Unmounting an integrated file system user-defined file system.................................................. 38
Saving and restoring an integrated file system user-defined file system..................................... 38
Object changes journaling in a user-defined file system...............................................................39
User-defined file system and independent auxiliary storage pools..............................................39
Library file system (QSYS.LIB)............................................................................................................. 39
QPWFSERVER authorization list in the QSYS.LIB file system........................................................40
File-handling restrictions in the QSYS.LIB file system.................................................................. 40
Support for user spaces in the QSYS.LIB file system.................................................................... 40
Support for save files in the QSYS.LIB file system.........................................................................40
Case-sensitivity in the QSYS.LIB file system................................................................................. 41
Path names in the QSYS.LIB file system........................................................................................ 41
Links in the QSYS.LIB file system................................................................................................... 41
Use of integrated file system commands and displays in the QSYS.LIB file system.................... 42
Use of integrated file system APIs in the QSYS.LIB file system.................................................... 42
Independent ASP QSYS.LIB.................................................................................................................42
QPWFSERVER authorization list in the independent ASP QSYS.LIB file system.......................... 43
File handling restrictions in the independent ASP QSYS.LIB file system......................................43
Support for user spaces in the independent ASP QSYS.LIB file system....................................... 43
Support for save files in the independent ASP QSYS.LIB file system........................................... 43
Case-sensitivity in the independent ASP QSYS.LIB file system.................................................... 44
Path names in the independent ASP QSYS.LIB file system...........................................................44
Links in the independent ASP QSYS.LIB file system......................................................................44
Use of integrated file system commands and displays in the independent ASP QSYS.LIB file
system........................................................................................................................................45
Use of integrated file system APIs in the independent ASP QSYS.LIB file system.......................45
Document library services file system (QDLS).................................................................................... 46
Integrated file system and HFS in the QDLS file system............................................................... 46
User enrollment in the QDLS file system........................................................................................46
Case sensitivity in the QDLS file system.........................................................................................46
Path names in the QDLS file system...............................................................................................46
Links in the QDLS file system..........................................................................................................47
Use of integrated file system commands and displays in the QDLS file system...........................47
Use of integrated file system APIs in the QDLS file system...........................................................48
Optical file system (QOPT)................................................................................................................... 48
Integrated file system and HFS in the QOPT file system...............................................................48
Case-sensitivity in the QOPT file system....................................................................................... 49
Path names in the QOPT file system.............................................................................................. 49
Links in the QOPT file system......................................................................................................... 49
Use of integrated file system commands and displays in the QOPT file system.......................... 49
Use of integrated file system APIs in the QOPT file system.......................................................... 50
IBM i NetClient file system (QNTC)......................................................................................................50
Authorities and ownership in the QNTC file system...................................................................... 51
Case sensitivity in the QNTC file system........................................................................................ 51
Path names in the QNTC file system.............................................................................................. 51
Links in the QNTC file system......................................................................................................... 52
Use of integrated file system commands and displays in the QNTC file system.......................... 52
Use of integrated file system APIs in the QNTC file system.......................................................... 52
QNTC environment variables..........................................................................................................53
Creating directories in the QNTC file system................................................................................. 54
Use of integrated file system APIs in the Network File System...............................................55
Enabling QNTC file system for Network Authentication Service................................................... 55
IBM i file server file system (QFileSvr.400)......................................................................................... 56

iv
 Case-sensitivity in the QFileSvr.400 file system............................................................................56
Path names in the QFileSvr.400 file system...................................................................................56
Communications in the QFileSvr.400 file system.......................................................................... 57
Security and object authority in the QFileSvr.400 file system.......................................................59
Securing QFileSvr.400 file system connections.............................................................................59
Links in the QFileSvr.400 file system............................................................................................. 60
Use of integrated file system commands and displays in the QFileSvr.400 file system.............. 61
Use of integrated file system APIs in the QFileSvr.400 file system.............................................. 61
Network File System (NFS).................................................................................................................. 62
Characteristics of the Network File System................................................................................... 62
Variations of servers and clients in the Network File System....................................................... 62
Links in the Network File System................................................................................................... 63
Use of integrated file system commands in the Network File System..........................................63
Use of integrated file system APIs in the Network File System.................................................... 64
Comparison of Network File System version 4 to Prior Versions.................................................. 65
Setting up a network for RPCSEC-GSS...........................................................................................65
Identity mapping.............................................................................................................................67
Accessing the integrated file system.........................................................................................................67
Accessing using menus and displays...................................................................................................68
Accessing using CL commands............................................................................................................ 69
Path name rules for CL commands and displays........................................................................... 72
Working with output of the RTVDIRINF and PRTDIRINF commands......................................... 74
Accessing the data of RTVDIRINF........................................................................................... 89
Using the data of RTVDIRINF...................................................................................................89
Collecting and analyzing folder attributes with IBM Navigator for i........................................ 90
Accessing using APIs........................................................................................................................... 91
Accessing using IBM i (SQL) services.................................................................................................. 91
Accessing using a PC............................................................................................................................ 91
Accessing using IBM Navigator for i.................................................................................................... 92
Accessing using IBM i NetServer......................................................................................................... 92
Accessing using File Transfer Protocol................................................................................................ 93
Integrated file system conversion.............................................................................................................94
Converting directories from *TYPE1 to *TYPE2.................................................................................. 94
Overview of *TYPE1 to *TYPE2 conversion................................................................................... 94
Directory conversion considerations..............................................................................................95
Conversion status determination..............................................................................................95
User profiles creation................................................................................................................ 96
Objects renamed....................................................................................................................... 96
User profile considerations....................................................................................................... 96
Auxiliary storage requirements.................................................................................................97
Tips: Symbolic link.................................................................................................................... 97
Tips: Independent ASP............................................................................................................. 98
Tips: Saving and restoring......................................................................................................... 98
Tips: Reclaiming integrated file system objects.......................................................................98
Integrated file system scanning................................................................................................98
Converting names to support additional characters........................................................................... 99
Overview of automatic name conversion....................................................................................... 99
Name conversion considerations.................................................................................................100
Conversion status determination........................................................................................... 100
Objects renamed.....................................................................................................................100
User profile considerations.....................................................................................................100
Tips: Symbolic link.................................................................................................................. 100
Tips: Independent ASP........................................................................................................... 100
Tips: Saving and restoring.......................................................................................................101
Tips: Reclaiming integrated file system objects.....................................................................101
Journaling objects................................................................................................................................... 101
Journaling overview........................................................................................................................... 101
Journal management....................................................................................................................101

v
 Objects you should journal...........................................................................................................102
Journaled integrated file system objects.....................................................................................102
Journaled operations....................................................................................................................103
Special considerations for journal entries................................................................................... 104
Considerations for multiple hard links and journaling.................................................................105
Starting journaling..............................................................................................................................105
Changing journaling............................................................................................................................106
Ending journaling................................................................................................................................106
Reclaim operation of the "root" (/), QOpenSys, and user-defined file systems.................................... 106
Reclaim Object Links (RCLLNK) and Reclaim Storage (RCLSTG) commands
comparison....................................................................................................................................107
Reclaim Object Links (RCLLNK) command........................................................................................108
Re-creation of integrated file system provided objects....................................................................108
Examples: Reclaim Object Links (RCLLNK) command.........................................................109
Example: Correcting problems for an object............................................................................... 109
Example: Correcting problems that exist in a directory subtree.................................................110
Example: Finding all damaged objects in the "root" (/), QOpenSys, and mounted user-
defined file systems.................................................................................................................110
Example: Deleting all damaged objects in the "root" (/), QOpenSys, and mounted user-
defined file systems.................................................................................................................110
Example: Running multiple RCLLNK commands to quickly reclaim all objects in the "root"
(/), QOpenSys, and mounted user-defined file systems........................................................ 110
Programming support..............................................................................................................................111
Copying data between stream files and database files.................................................................... 111
Copying data using CL commands............................................................................................... 111
Copying data using APIs...............................................................................................................112
Copying data using data-transfer functions.................................................................................112
Transferring data from a database file to a stream file.......................................................... 113
Transferring data from a stream file to a database file.......................................................... 113
Transferring data into a newly created database file definition and file............................... 114
Creating a format description file........................................................................................... 114
Copying data between stream files and save files............................................................................ 114
Performing operations using APIs..................................................................................................... 115
ILE C functions..............................................................................................................................122
Large file support.......................................................................................................................... 122
Path name rules for APIs..............................................................................................................123
File descriptor............................................................................................................................... 124
Security......................................................................................................................................... 125
Socket support................................................................................................................................... 125
Naming and international support.....................................................................................................126
Data conversion..................................................................................................................................126
Example: Integrated file system C functions.................................................................................... 127
Working with files and folders using IBM Navigator for i........................................................................131
Creating a folder.................................................................................................................................131
Removing a file or folder.................................................................................................................... 132
Moving files or folders to another file system................................................................................... 133
Setting permissions............................................................................................................................134
Setting up file text conversion........................................................................................................... 135
Sending a file or folder to another system.........................................................................................135
Changing options for sending a file or folder.....................................................................................136
Creating a file share........................................................................................................................... 136
Changing a file share.......................................................................................................................... 137
Removing a file share......................................................................................................................... 137
Creating a new user-defined file system........................................................................................... 138
Mounting a user-defined file system................................................................................................. 138
Unmounting a user-defined file system............................................................................................ 139
Working with dynamically mounted file systems..............................................................................139
Setting whether objects should be scanned or not...........................................................................141

vi
 Checking in objects............................................................................................................................ 142
Checking out objects..........................................................................................................................143
Downloading a file or folder............................................................................................................... 144
Uploading a file...................................................................................................................................145
Navigate directly to an integrated file system folder........................................................................ 145
Save integrated file system folder as a favorite................................................................................ 145
Transport-independent remote procedure call...................................................................................... 146
Network selection APIs..................................................................................................................... 146
Name-to-address translation APIs................................................................................................... 147
eXternal Data Representation (XDR) APIs........................................................................................ 147
Authentication APIs........................................................................................................................... 149
Transport-independent RPC (TI-RPC) APIs...................................................................................... 149
TI-RPC simplified APIs................................................................................................................. 149
TI-RPC top-level APIs.................................................................................................................. 149
TI-RPC intermediate-level APIs...................................................................................................150
TI-RPC expert-level APIs............................................................................................................. 150
Other TI-RPC APIs........................................................................................................................ 150
Related information for integrated file system....................................................................................... 151

Notices..............................................................................................................153
Programming interface information........................................................................................................ 154
Trademarks.............................................................................................................................................. 154
Terms and conditions.............................................................................................................................. 155

vii
viii
Integrated file system
The integrated file system is a part of the IBM i operating system that supports stream input/output and
storage management similar to personal computer and UNIX operating systems, while providing you with
an integrating structure over all information stored in the system.
Note: By using the code examples, you agree to the terms of the “Code license and disclaimer
information” on page 151.

What's new for IBM i 7.4

Read about new or changed information for the Integrated file system topic collection.

Configuring secure QFileSvr.400 file system connections

For details, see “Securing QFileSvr.400 file system connections” on page 59.

QFileSvr.400 file system transport layer security

For details, see “Communications in the QFileSvr.400 file system” on page 57.

What's new for since the first IBM i 7.4 publication

The following revisions or additions have been made to the Integrated files system documentation since
the first IBM i 7.4 publication:
• April 2019 update :
– The “QNTC environment variables” on page 53 topic was updated.

How to see what is new or changed

To help you see where technical changes have been made, the information center uses:
• The image to mark where new or changed information begins.
• The image to mark where new or changed information ends.
In PDF files, you might see revision bars (|) in the left margin of new and changed information.
To find other information about what is new or changed this release, see the Memo to users.

PDF file for Integrated file system

You can view and print a PDF file of this information.
To view or download the PDF version of this document, select Integrated file system.

Saving PDF files

To save a PDF on your workstation for viewing or printing:
1. Right-click the PDF link in your browser.
2. Click the option that saves the PDF locally.
3. Navigate to the directory in which you want to save the PDF.
4. Click Save.

© Copyright IBM Corp. 1999, 2019 1

 Downloading Adobe Reader
You need Adobe Reader installed on your system to view or print these PDFs. You can download a free
copy from the Adobe Web site (www.adobe.com/products/acrobat/readstep.html) .

Overview of the integrated file system

Here is some basic information about the integrated file system on your IBM i operating system and how it
can be of use on your system.

What the integrated file system is

The integrated file system is a part of the IBM i operating system that supports stream input/output
and storage management similar to personal computer and UNIX operating systems, while providing an
integrating structure over all information stored on your system.
The integrated file system comprises 10 file systems, each with its own set of logical structures and rules
for interacting with information in storage.
The key features of the integrated file system are as follows:
• Support for storing information in stream files that can contain long continuous strings of data. These
strings of data might be, for example, the text of a document or the picture elements in a picture. The
stream file support is designed for efficient use in client/server applications.
• A hierarchical directory structure that allows objects to be organized like fruit on the branches of a tree.
You can access an object by specifying the path through the directory to the object.
• A common interface that enables users and applications to access not only the stream files but also
database files, documents, and other objects that are stored on your system.
• A common view of stream files that are stored locally on your system, on an Integrated xSeries Server
(IXS), or on a remote Windows NT server. Stream files can also be stored remotely on a local area
network (LAN) server, on another remoteIBM i product, or on a Network File System (NFS) server.

Figure 1. A structure over all information stored in the IBM i operating system

Related concepts
File systems

2 IBM i: Integrated file system

 A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.

Why use the integrated file system

The integrated file system enhances the already extensive data management capabilities of IBM i with
additional capabilities to better support emerging and future forms of information processing, such as
client/server, open systems, and multimedia.
You can use the integrated file system to do the following things:
• Provide fast access to IBM i data, especially for applications such as IBM i Access that use the IBM i file
server.
• Allow more efficient handling of types of stream data, such as images, audio, and video.
• Provide a file system base and a directory base for supporting open system standards based on the
UNIX operating system, such as Portable Operating System Interface for Computer Environments
(POSIX) and X/Open Portability Guide (XPG). This file structure and this directory structure also
provides a familiar environment for users of PC operating systems, such as Disk Operating System
(DOS) and Microsoft Windows operating systems.
• Allow file support with unique capabilities (such as record-oriented database files, UNIX operating
system-based stream files, and file serving) to be handled as separate file systems, while allowing them
all to be managed through a common interface.

System i Users

Applications

Integrated File System Interface

File File File

System System System

Figure 2. A common interface to separate file systems

• Allow PC users to take better advantage of their graphical user interface. For example, Windows users
can use the Windows graphical tools to operate on IBM i stream files and other objects in the same way
they operate on files stored on their PCs.
• Provide continuity of object names and associated object information across national languages. For
example, this ensures that individual characters remain the same when switching from the code page of
one language to the code page of another language.
Related concepts
File systems

Integrated file system 3

 A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.

Integrated file system concepts

This topic introduces the basic concepts of integrated file system, such as directory, link, path name,
stream file, name continuity, extended attributes, and scanning support.

Directory
A directory is a special object that is used to locate objects by names that you specify. Each directory
contains a list of objects that are attached to it. That list can include other directories.
The integrated file system provides a hierarchical directory structure for you to access all objects in
your system. You might think of this directory structure as an inverse tree where the root is at the top
and the branches below. The branches represent directories in the directory hierarchy. These directory
branches have subordinate branches that are called subdirectories. Attached to the various directory and
subdirectory branches are objects such as files. Locating an object requires specifying a path through the
directories to the subdirectory to which the object is attached. Objects that are attached to a particular
directory are sometimes described as being in that directory.
A particular directory branch, along with all of its subordinate branches (subdirectories) and all of the
objects that are attached to those branches, is referred to as a subtree. Each file system is a major subtree
in the integrated file system directory structure. In the QSYS.LIB and independent ASP QSYS.LIB file
systems' subtrees, a library is handled the same way as a subdirectory. Objects in a library are handled
like objects in a subdirectory. Because database files contain objects (database file members), they
are handled like subdirectories rather than objects. In the document library services file system (QDLS
subtree), folders are handled like subdirectories and documents in folders are handled like objects in a
subdirectory.
Because of differences in file systems, the operations you can perform in one subtree of the directory
hierarchy may not work in another subtree.
The integrated file system directory support is similar to the directory support that is provided by the DOS
file system. In addition, it provides features typical of UNIX systems, such as the ability to store a file only
once but access it through multiple paths by using links.
File systems and objects are branches on the integrated file system directory tree. See the following figure
for an example of an integrated file system directory tree.

4 IBM i: Integrated file system

Figure 3. Sample integrated file system directory tree

Integrated file system 5

 Current directory
The current directory is similar to the idea of the current library. It is also called the current working
directory, or just working directory.
Your current directory is the first directory in which the operating system looks for your programs and files
and stores your temporary files and output. When you request an operation on an object, such as a file,
the system searches for the object in your current directory unless you specify a different directory path.

Home directory
The home directory is used as the current directory when you sign on to the system. The name of the
home directory is specified in your user profile.
When your job is started, the system looks in your user profile for the name of your home directory. If
a directory by that name does not exist on the system, the home directory is changed to the "root" (/)
directory.
Typically, the system administrator who creates the user profile for a user also creates the user's home
directory. Creating individual home directories for each user under the /home directory is preferable.
The /home directory is a subdirectory under the "root" (/) directory. The system default expects the name
of the home directory of a user to be the same name as the user profile.
For example, the command CRTUSRPRF USRPRF(John) HOMEDIR(*USRPRF) will assign the home
directory for John to /home/JOHN. If the directory /home/JOHN does not exist, the "root" (/) directory
becomes the home directory for John.
You can specify a directory other than the home directory as your current directory at any time after you
sign on by using the Change Current Directory (CHGCURDIR) CL command, the chdir( ) API, or the fchdir()
API.
The home directory chosen during process initiation will remain each thread's home directory by default.
This is true regardless of whether your active user profile for the thread has changed after initiation.
However, there is support provided by the Change Job (QWTCHGJB) API that can be used to change
the home directory being used for a thread to that thread's current user profile's home directory (or the
"root" (/) directory if that home directory does not exist). Secondary threads will always inherit the home
directory of the thread that created it. Note that the process' current directory does not change when
you use QWTCHGJB to change the thread's home directory. The current directory is scoped to the process
level, and the home directory is scoped to the thread level. Changing the current working directory in any
thread changes it for the whole process. Changing the home directory for a thread does not change its
current working directory.
Related information
Change current directory (CHGCURDIR) command
chdir()--Change Current Directory API
fchdir()--Change Current Directory by Descriptor API
Application programming interfaces (APIs)

Provided directories
When the system is restarted, the integrated file system creates the directories listed here if they do not
already exist. These directories should not be moved or renamed after being created by the system.
Note: Do not replace the following system-created directories with symbolic links to other objects. For
example, do not replace /home with a symbolic link to a directory on an independent ASP. Otherwise,
there might be problems on the independent ASP as well as problems when creating new user profiles.
/tmp
The /tmp directory gives applications a place to store temporary objects. This directory is a
subdirectory of the "root" (/) directory, so its path name is /tmp.

6 IBM i: Integrated file system

 Once an application puts an object in the /tmp directory, the object stays there until you or the
application removes it. The system does not automatically remove objects from the /tmp directory or
perform any other special processing for objects in the /tmp directory.
You can use the user displays and commands that support the integrated file system to manage
the /tmp directory and its objects. For example, you can use the Work with Object Links display or
the WRKLNK command to copy, remove, or rename the /tmp directory or objects in the directory. All
users are given *ALL authority to the directory, which means that they can perform most valid actions
on the directory.
An application can use the application programming interfaces (API) that support the integrated file
system to manage the /tmp directory and its objects. For example, the application program can
remove an object in the /tmp directory by using the unlink() API.
If the /tmp directory is removed, it is automatically created again during the next restart of the
system.
The /tmp directory can have the restricted rename and unlink attribute set to Yes for
operating system commonality and security purposes.
Note: The restricted rename and unlink attribute is equivalent to the S_ISVTX mode bit for a
directory.
If the restricted rename and unlink attribute is set to Yes, you cannot rename or unlink
objects within the /tmp directory unless one of the following conditions is true:
• You are the owner of the object.
• You are the owner of the directory.
• You have all object (*ALLOBJ) special authority.
If the attribute is set to Yes, and you do not have the appropriate authorities, you will see
error number 3027 (EPERM) or message MSGCPFA0B1 (Requested operation not allowed.
Access problem) for rename or unlink failures when using the following commands and APIs:
• Remove Link (RMVLNK, DEL, and ERASE) command
• Remove Directory (RMVDIR, RD, and RMDIR) command
• Rename Object (RNM and REN) command
• Move Object (MOV and MOVE) command
• Rename File or Directory (rename()) API
• Rename File or Directory, Keep "new" If It Exists (Qp0lRenameKeep()) API
• Rename File or Directory, Unlink "new" If It Exists (Qp0lRenameUnlink()) API
• Remove Directory (rmdir()) API
• Remove Link to File (unlink()) API
The restricted rename and unlink attribute and S_ISVTX mode bit can be modified using
the Change Attribute (CHGATR) command or the Set Attributes (Qp0lSetAttr()) or
Change File Authorizations (chmod) APIs if you are the owner of the object, or if you have
all object (*ALLOBJ) special authority. But, if the attribute is changed to No, you will lose the operating
system commonality and security benefits that the Yes setting provides.
When the /tmp directory is created during a restart of the system, the attribute is set to Yes. If
the /tmp directory already exists during a restart of the system, the attribute is not changed.
/home
System administrators use the /home directory to store a separate directory for every user. The
system administrator often sets the home directory that is associated with the user profile to be the
user's directory in /home, for example/home/john.
/etc
The /etc directory stores administrative, configuration, and other system files.

Integrated file system 7

 /usr
The /usr directory includes subdirectories that contain information that is used by the system. Files
in /usr typically do not change often.
/usr/bin
The /usr/bin directory contains the standard utility programs.
/QIBM
The /QIBM directory is the system directory and is provided with the system.
/QIBM/ProdData
The /QIBM/ProdData directory is a system directory used for Licensed program data.
/QIBM/UserData
The /QIBM/UserData directory is a system directory used for Licensed Program user data such as
configuration files.
/QOpenSys/QIBM
The /QOpenSys/QIBM directory is the system directory for the QOpenSys file system.
/QOpenSys/QIBM/ProdData
The /QOpenSys/QIBM/ProdData directory is the system directory for the QOpenSys file system and
is used for Licensed program data.
/QOpenSys/QIBM/UserData
The /QOpenSys/QIBM/UserData directory is the system directory for the QOpenSys file system and
is used for Licensed Program user data such as configuration files.
/asp_name/QIBM
The /asp_name/QIBM directory is the system directory for any independent ASPs that exist on your
system, where asp_name is the name of the independent ASP.
/asp_name/QIBM/UserData
The /asp_name/QIBM/UserData directory is a system directory used for Licensed Program user
data such as configuration files for any independent ASPs that exist on your system, where asp_name
is the name of the independent ASP.
/dev
The /dev directory contains various system files and directories.
/dev/xti
The /dev/xti directory contains the UDP and TCP device drivers.
Related concepts
Home directory
The home directory is used as the current directory when you sign on to the system. The name of the
home directory is specified in your user profile.
Related reference
UDP and TCP devices in the "root" (/) file system
The "root" (/) file system under the directory of /dev/xti will now hold two device drivers named udp and
tcp.
Open systems file system (QOpenSys)
The QOpenSys file system is compatible with open system standards based on UNIX, such as POSIX and
X/Open Portability Guide (XPG). Like the "root" (/) file system, this file system takes advantage of the
stream file and directory support that is provided by the integrated file system.
Related information
Work with Object Links (WRKLNK) command

8 IBM i: Integrated file system

*TYPE2 directories
The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format. The *TYPE2 directory format is an enhancement of the original *TYPE1 directory
format.
Note: The concept of *TYPE1 and *TYPE2 stream files is different from the concept of *TYPE1 and
*TYPE2 directory formats. One does not relate to the other.
*TYPE2 directories have a different internal structure and different implementation than *TYPE1
directories.
The advantages of *TYPE2 directories are:
• Improved performance
• Improved reliability
• Added functionality
• In many cases, less auxiliary storage space
*TYPE2 directories improve file system performance over *TYPE1 directories, especially when creating
and deleting directories.
*TYPE2 directories are more reliable than *TYPE1 directories. After a system abnormally ends, *TYPE2
directories are completely recovered unless there has been an auxiliary storage failure. *TYPE1
directories may require the use of the Reclaim Storage (RCLSTG) command in order to recover
completely.
*TYPE2 directories provide the following added functionality:
• *TYPE2 directories support renaming the case of a name in a monocase file system (for example,
renaming from A to a).
• An object in a *TYPE2 directory can have up to one million links compared to 32 767 links for *TYPE1
directories. This means you can have up to 1 million hard links to a stream file, and a *TYPE2 directory
can contain up to 999 998 subdirectories.
• Using IBM Navigator for i, the list of entries are automatically sorted in binary order when you open a
directory that has the *TYPE2 format.
• Some new functions such as integrated file system scanning support are only available for objects in
*TYPE2 directories.
Typically, *TYPE2 directories that have fewer than 350 objects require less auxiliary storage than *TYPE1
directories with the same number of objects. *TYPE2 directories with more than 350 objects are ten
percent larger (on average) than *TYPE1 directories.
There are several ways to get *TYPE2 directories on your system:
• New IBM i platforms that are preinstalled with OS/400® V5R2 or IBM i V5R3, or later, have *TYPE2
directories. No conversion is needed for "root" (/), QOpenSys, and UDFSs in ASPs 1-32.
• If you install OS/400 V5R2 or IBM i V5R3, or later, on a IBM i platform for the first time, the platform has
*TYPE2 directories. No conversion is needed for "root" (/), QOpenSys, and UDFSs in ASPs 1-32.
• The V5R2 conversion utility is used to convert the file systems. For more information about the
conversion utility, see the Convert to a *TYPE2 directory section in the V5R2 iSeries Information Center.
• If the UDFSs in an independent ASP have not yet been converted to the *TYPE2 format, they will be
converted the first time that the independent ASP is varied on to a system installed with OS/400 V5R2
or IBM i V5R3 or later.
• All other supported file systems except UDFSs on independent ASPs that are still using *TYPE1
directories are converted automatically by the system. This conversion begins after the installation
of IBM i V5R3 or later releases. It should not significantly impact your system activity.
To determine the directory format for the file systems on your system, use the Convert Directory
(CVTDIR) command:

Integrated file system 9

 CVTDIR OPTION(*CHECK)
Note: *TYPE2 directories are supported on OS/400 V5R2 or IBM i V5R3 or later, but there are some
differences from normal *TYPE2 directory support.
Related reference
Converting directories from *TYPE1 to *TYPE2
The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format.
Related information
Reclaim Storage (RCLSTG) command
Convert Directory (CVTDIR) command

Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.
For users of directory-based file systems, it is convenient to think of an object, such as a file, as something
that has a name that identifies it to the system. In fact, it is the directory path to the object that identifies
it. You can sometimes access an object by giving just the object's name. You can do this because the
system is designed to assume the directory part of the path under certain conditions. The idea of a link
takes advantage of the reality that it is the directory path that identifies the object. The name is given to
the link rather than the object.
After you get used to the idea that the link has the name rather than the object, you begin to see
possibilities that were hidden before. There can be multiple links to the same object. For example, two
users can share a file by having a link from each user's home directory to the file (see “Home directory”
on page 6). Certain types of links can cross file systems, and can exist without an object existing.
There are two types of links: Hard link and Symbolic link. When using path names in programs, you have
a choice of using a hard link or a symbolic link. Each type of link has advantages and disadvantages. The
conditions under which one type of link has an advantage over the other type is as follows:

Table 1. Comparison of hard link and symbolic link

Item Hard link Symbolic link
Name resolution Faster. A hard link contains a Slower. A symbolic link contains
direct reference to the object. a path name to the object, which
must be resolved to find the
object.
Object existence Required. An object must exist in Optional. A symbolic link can be
order to create a hard link to it. created when the object it refers
to does not exist.
Object deletion Restricted. All hard links to Unrestricted. An object can
an object must be unlinked be deleted even if there are
(removed) to delete the object. symbolic links referring to it.
Static objects (attributes do not Faster. For a static object, Slower. Name resolution is
change) name resolution is the primary slower when symbolic links are
performance concern. Name used.
resolution is faster when hard
links are used.
Scope Restricted. Hard links cannot Unrestricted. Symbolic links can
cross file systems. cross file systems.

10 IBM i: Integrated file system

Hard link
A hard link, which is sometimes just called a link, cannot exist unless it is linked to an actual object.
When an object is created in a directory (for example, by copying a file into a directory), the first hard
link is established between the directory and the object. Users and application programs can add other
hard links. Each hard link is indicated by a separate directory entry in the directory. Links from the same
directory cannot have the same name, but links from different directories can have the same name.
If supported by the file system, there can be multiple hard links to an object, either from the same
directory or from different directories. The one exception is where the object is another directory. There
can be only one hard link from a directory to another directory.

Figure 4. A directory entry defines each hard link

Hard links can be removed without affecting the existence of an object as long as there is at least one
remaining hard link to the object. When the last hard link is removed, the object is removed from the
system, unless an application has the object open. Each application that has the object open can continue
to use it until that application closes the object. When the object is closed by the last application using it,
the object is removed from the system. An object cannot be opened after the last hard link is removed.
The concept of a hard link can also be applied to the QSYS.LIB or independent ASP QSYS.LIB file systems
and the document library services (QDLS) file system, but with a restriction. A library, in effect, has one
hard link to each object in the library. Similarly, a folder has one hard link to each document in the folder.
Multiple hard links to the same object are not allowed in QSYS.LIB, independent ASP QSYS.LIB, or in
QDLS, however.
A hard link cannot cross file systems. For example, a directory in the QOpenSys file system cannot have a
hard link to an object in the QSYS.LIB or independent ASP QSYS.LIB file systems or to a document in the
QDLS file system.

Integrated file system 11

 Symbolic link
A symbolic link, which is also called a soft link, is a path name contained in a file.
When the system encounters a symbolic link, it follows the path name provided by the symbolic link and
then continues on any remaining path that follows the symbolic link. If the path name begins with a /, the
system returns to the / ("root") directory and begins following the path from that point. If the path name
does not begin with a /, the system returns to the immediately preceding directory and follows the path
name in the symbolic link beginning at that directory.
Consider the following example of how a symbolic link might be used:

Figure 5. An example of using a symbolic link

You select a menu option to show the status of customer accounts. The program displaying the menu
uses the following path name:

/Customer/Status/Summary

The system follows the Customer link, which leads to a directory 1, and then follows the Status link. The
Status link is a symbolic link, which contains a path name 2. Because the path name begins with a /, the
system returns to the / ("root") directory and follows the links Records and Accounts in sequence. This
path leads to another directory 3. Now the system completes the path in the path name provided by the
program. It follows the Summary link, which leads to a file 4 containing the data you will need.
Unlike a hard link, a symbolic link is an object (of object type *SYMLNK); it can exist without pointing to an
object that exists. You can use a symbolic link, for example, to provide a path to a file that will be added or
replaced later.
Also unlike a hard link, a symbolic link can cross file systems. For example, if you are working in one
file system, you can use a symbolic link to access a file in another file system. Although the QSYS.LIB,
independent ASP QSYS.LIB, and QDLS file systems do not support creating and storing symbolic links, you
can create a symbolic link in the "root" (/) or QOpenSys file system that allows you to:

12 IBM i: Integrated file system

 • Access a database file member in the QSYS.LIB or independent ASP QSYS.LIB file systems.
• Access a document in the QDLS file system.

Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.
The path name is expressed as a sequence of directory names followed by the name of the object.
Individual directories and the object name are separated by a slash (/) character; for example:

directory1/directory2/file

For your convenience, the backslash (\) can be used instead of the slash in integrated file system
commands.
There are two ways of indicating a path name:
• An absolute path name begins at the highest level, or "root" directory (which is identified by the /
character). For example, consider the following path from the / directory to the file named Smith.

Figure 6. The components of a path name

The absolute path name to the Smith file is as follows:

/Dept2/Photo/Smith

The absolute path name is also known as the full path name.
• If the path name does not begin with the / character, the system assumes that the path begins at your
current directory. This type of path name is called a relative path name. For example, if your current
directory is Dept2 and it has a subdirectory named Photo containing the file Smith, the relative path
name to the file is:

Photo/Smith

Integrated file system 13

 Notice that the path name does not include the name of the current directory. The first item in the name
is the directory or object at the next level below the current directory.
Related reference
Path name rules for APIs
When using an API to operate on an integrated file system object, you identify the object by supplying its
path name.
Path name rules for CL commands and displays
When using a CL command or display to operate on an object, you identify the object by supplying its path
name.

Stream file
A stream file is a randomly accessible sequence of bytes, with no further structure imposed by the system.
The integrated file system provides support for storing and operating on information in the form of stream
files. Documents that are stored in your system's folders are stream files. Other examples of stream files
are PC files and the files in UNIX systems. An integrated file system stream file is a system object that has
an object type of *STMF.
To better understand stream files, it is useful to compare them with IBM i database files. A database
file is record-oriented; it has predefined subdivisions that consist of one or more fields that have specific
characteristics, such as length and data type.

Figure 7. Comparison of a stream file and a record-oriented file

Stream files and record-oriented files are structured differently, and this difference in structure affects
how the files are used. The structure affects how an application is written to interact with the files and
where each type of file is best used in an application. A record-oriented file, for example, is well suited
for storing customer statistics such as name, address, and account balance. A record-oriented file allows
these predefined fields to be individually accessed and manipulated, using the extensive programming
facilities of your system. But a stream file is better suited for storing information, such as a customer's
picture, which is composed of a continuous string of bits representing variations in color. Stream files are
particularly well suited for storing strings of data, such as the text of a document, images, audio, and
video.

14 IBM i: Integrated file system

 A file has one of two format options: *TYPE1 stream file or *TYPE2 stream file. The file format depends on
the release the file was created on, or if a file was created in a user-defined file system, the value that was
specified for that file system.
Note: The concept of *TYPE1 and *TYPE2 stream files is different from the concept of *TYPE1 and
*TYPE2 directory formats. One does not relate to the other.

*TYPE1 stream files

A *TYPE1 stream file has the same format as stream files created on releases before OS/400 V4R4.
A *TYPE1 stream file has a minimum size of 4096 bytes and a maximum object size of approximately 128
GB (1 GB equals approximately 1 073 741 824 bytes).

*TYPE2 stream files

A *TYPE2 stream file has high-performance file access.
*TYPE2 stream files have a maximum object size of approximately 1 TB (1 TB equals approximately 1 099
511 627 776 bytes) in the "root" (/), QOpenSys and user-defined file systems. Otherwise, the maximum is
approximately 256 GB. It is also capable of memory mapping, as well as the ability to specify an attribute
to optimize main storage allocation. All files created with OS/400 V4R4 and newer systems are *TYPE2
stream files, unless they were created in a user-defined file system that specified a file format of *TYPE1.
Note: Any files larger than 256 GB cannot be saved or restored to systems before IBM i V5R3.

Name continuity
When you use the "root" (/), QOpenSys, and user-defined file systems, you can take advantage of system
support that ensures characters in object names remain the same.
This also applies when you use these file systems across systems and connected devices that have
different character encoding schemes (code pages). Your system stores the characters in the names in a
16-bit form that is known as UCS2 Level 1 (also called Unicode) for *TYPE1 directories and UTF-16 for
*TYPE2 directories. UCS2 Level 1 and UTF-16 are subsets of the ISO 10646 standard. When the name is
used, the system converts the stored form of the characters into the appropriate character representation
in the code page being used. The names of extended attributes associated with each object are also
handled the same way.

Figure 8. Keep characters the same across encoding schemes

This support makes it easier to interact with a system from devices using different code pages. For
example, PC users can access an IBM i file using the same file name, even though their PCs do not
have the same code page as your system. The conversion from one code page to another is handled
automatically by your system. Of course, the device must be using a code page that contains the
characters used in the name.

Integrated file system 15

 Related concepts
*TYPE2 directories
The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format. The *TYPE2 directory format is an enhancement of the original *TYPE1 directory
format.
Overview of automatic name conversion
Some file systems that are not case sensitive, such as “root” (/) and the UDFSs that are created with
CASE(*MONO), support names stored in Unicode Standard 4.0. The system runs an automatic name
conversion to support the additional characters in names.

Extended attributes
An extended attribute is information associated with an object that provides additional details about the
object. The extended attribute consists of a name, which is used to refer to it, and a value. The value can
be text, binary data, or another type of data.
The extended attributes for an object exist only as long as the object exists.
Extended attributes come in many varieties and can be used to contain a variety of information. You may
need to be aware of the following three extended attributes, in particular:
.SUBJECT
A brief description of the content or purpose of the object.
.TYPE
The type of data in the object. The type of data might be text, binary, source for a program, a compiled
program, or other information.
.CODEPAGE
The code page to be used for the object. The code page used for the object is also used for the
extended attribute associated with the object.
A period (.) as the first character of the name means that the extended attribute is a standard system
extended attribute (SEA), which is reserved for system use.
Various objects in the various file systems may or may not have extended attributes. The QSYS.LIB and
independent ASP QSYS.LIB file systems support three predefined extended attributes: .SUBJECT, .TYPE,
and .CODEPAGE. In the document library services (QDLS) file system, folders and documents can have
any kind of extended attribute. Some folders and documents may have extended attributes, and some
may not. In the "root" (/), QOpenSys, and user-defined file systems, all directories, stream files, and
symbolic links can have extended attributes of any kind. Some, however, may not have any extended
attributes at all.
The Work with Object Links (WRKLNK) command and Display Object Links (DSPLNK)
command can be used to display the .SUBJECT extended attribute for an object. There is no other
integrated file system support through which applications or users can access and change extended
attributes. The only exceptions to this rule are the Display a UDFS (DSPUDFS) and the Display
Mounted File System Information (DSPMFSINF) CL commands, which present extended
attributes to users.
Extended attributes associated with some objects in the QDLS can, however, be changed through
interfaces provided by the hierarchical file system (HFS).
If a client PC is connected to a IBM i platform through the Windows operating systems, the programming
interfaces of the respective operating system (such as DosQueryFileInfo and DosSetFileInfo) can be used
to query and set the extended attributes of any file object.
If you define extended attributes, use the following naming guidelines:
• The name of an extended attribute can be up to 255 characters long.
• Do not use a period (.) as the first character of the name. An extended attribute whose name begins with
a period is interpreted as a standard system extended attribute.

16 IBM i: Integrated file system

 • To minimize the possibility of name conflicts, use a consistent naming structure for extended attributes.
The following form is recommended:

CompanyNameProductName.Attribute_Name

Scanning support
With the IBM i operating system, you can scan integrated file system objects.
This support creates flexibility for users by allowing scans for various items; users decide when the scan
should occur and what actions to take based on the results of their scans.
The two exit-points related to this support are:
• QIBM_QP0L_SCAN_OPEN - Integrated File System Scan on Open Exit Program
For this exit point, the integrated file system scan on open exit program is called to do scan processing
when an integrated file system object is opened under certain conditions.
• QIBM_QP0L_SCAN_CLOSE - Integrated File System Scan on Close Exit Program
For this exit point, the integrated file system scan on close exit program is called to do scan processing
when an integrated file system object is closed under certain conditions.
Note: Only objects in file systems that have been fully converted to *TYPE2 directories will be scanned.
Related tasks
Setting whether objects should be scanned or not
You can specify whether objects should be scanned or not in the "root" (/), QOpenSys and user-defined
file systems. Follow these steps to set the scanning options.
Related information
Integrated File System Scan on Open Exit Program
Integrated File System Scan on Close Exit Program

Examples: Scanning for viruses and files being opened

These examples show what the exit program can scan for.
• Viruses
The exit programs can scan for viruses. If a virus is located in a file, the antivirus program can act
accordingly by repairing the problem or by attempting to quarantine the virus. Because the IBM
i platform itself cannot be infected by the virus, what this accomplishes is a reduction in virus
transmissions between systems.
• Calls to know when a file was opened
You may also scan to find out when a file was opened. By enacting this scan, you are able to track the
date and time of when certain files were accessed. This is useful when you want to track the behavior of
certain users.
The scan can occur at two different times depending on how the system values are set and how the scan
environment is established. The following list describes different kinds of scanning depending on the time
they occur.
1. Runtime scanning
A runtime scan is a scan on a file or files during normal day-to-day activities. This ensures the integrity
of your files every time they are accessed. Scanning during your normal activities allows you to ensure
that your file or files are current for whatever standards you are scanning.
Example of a runtime scan for viruses
You choose to access a file on the integrated file system from your PC. When the file is opened from
the PC, it is scanned. Because an open exit program is registered and the QSCANFS system value is
set to scan files in the "root" (/), QOpenSys and UDFS file systems. The scan shows one virus is found

Integrated file system 17

 and the anti-virus exit program proceeds to repair the problem. After the exit program repairs the file,
the file is no longer infected. Thus, the access from the PC is not infected and it cannot spread the
infection.
Now say that instead of scanning for viruses on that access, you choose not to do a runtime scan.
Then, after accessing the infected file from your PC, the virus might be transferred to your PC. By
employing a runtime scan, the virus can be detected before it spreads to your PC.
The main shortcoming of this method is that resource time is required to do the scans. Users
attempting to access a file must wait until the scan is completed, before being able to use the file.
The system ensures that scanning is performed only when required, not on every access.
2. Mass or manually activated scanning
You can use this option if you want to scan multiple items at the same time. In this instance, you can
set the scan to occur when your system is offline, such as during the weekend. This has very little
impact on accessing files during your normal day-to-day activities. The scan is done offline. Therefore,
it can reduce runtime scan overhead for files that do not change after the mass scan is completed
because re-scans are not required when such files are accessed again.
Related concepts
Related system values
You can use the QSCANFS and QSCANFSCTL system values to establish the scanning environment you
want for your system.
Related information
Integrated File System Scan on Open Exit Program
Integrated File System Scan on Close Exit Program

Related system values

You can use the QSCANFS and QSCANFSCTL system values to establish the scanning environment you
want for your system.
The following list includes the names of the scan-related system values and the descriptions for each one.
These system values and their control options are described for the IBM Navigator for i. The comparable
character-based interface values are listed in parenthesis following the IBM Navigator for i names. For
example, for the system value QSCANFSCTL, when the Scan accesses through file servers only control
option in IBM Navigator for i is selected, you would essentially be creating the same results by specifying
the character-based control option *FSVRONLY.
The name and descriptions of these system values are as follows:
1. Use registered exit program to scan the "root" (/), QOpenSys, and user-defined file systems
(QSCANFS)
This system value can be used to specify whether file systems should be scanned. Only objects in the
"root" (/), QOpenSys, and user-defined file systems will be scanned if the file system has been fully
converted. This value specifies whether objects should be scanned by exit programs registered with
any of the integrated file system scan-related exit points.
The default value is that objects will be scanned if any exit programs are registered.
2. Scan control (QSCANFSCTL)
For this system value, you can use the default control options or you can use specified control options.
Brief descriptions of the different specified control options based on the IBM Navigator for i system
values follow:
• Scan accesses through file servers only (*FSVRONLY specified)
A scan will only take place if you access the IBM i platform from a file server. If this option is not
selected, all accesses are scanned.
• Fail request if exit program fails (*ERRFAIL specified)

18 IBM i: Integrated file system

 If there are errors when the exit program is called, the request or operation that triggered the call to
the exit program will fail. If this option is not selected, the system will skip the failing exit program
and the object will be treated as if it was not scanned.
• Perform write access upgrades (*NOWRTUPG not specified)
The access upgrade will occur for the scan descriptor passed to the exit program to include write
access. If *NOWRTUPG option is not selected, the system will not attempt to do the write access
upgrade.
If *NOWRTUPG is specified, the system will not attempt to upgrade the access for the scan
descriptor passed to the exit program to include write access. If *NOWRTUPG is not specified, the
system will attempt to do the write access upgrade.
• Use 'only when objects have changed' attribute to control scan (*USEOCOATR specified)
The 'object change only' attribute (only scan the object if it has been modified) will be used. If
this option is not selected, this attribute will not be used and the object will be scanned after it is
modified and when scan software indicates an update.
• Fail close request if scan fails during close (*NOFAILCLO not specified)
If an object failed a scan during close processing, the close request will be failed. If this option
is not selected, the close request will not be failed. When not selected, this value overrides the
specification of the 'fail request if exit program fails' value.
If *NOFAILCLO is specified, the system will not fail the close request with an indication of scan
failure, even if the object failed a scan which was done as part of the close processing.
• Scan on next access after object has been restored (*NOPOSTRST not specified)
Objects will be scanned after they are restored. If 'the object will not be scanned' attribute is
specified, the object will be scanned once after being restored. If the attribute 'object change only' is
specified, the object will be scanned after being restored.
If *NOPOSTRST is specified while the objects are restored, they will not be scanned just because
they were restored. If the object attribute is 'the object will not be scanned', the object will not be
scanned at any time. If the object attribute is 'object change only', the object will only be scanned if it
is modified after being restored.
Related information
Security system values: Use registered exit programs to scan the root (/), QOpenSys, and user-defined file
systems
Security system values: Scan control

Scanning occurrences
Scanning can occur for a variety of reasons. Here is some information about when and why a scan might
occur.
To see the current scan status and attribute of an object, you can use the Work with Object Links
(WRKLNK) command, the Display Object Links (DSPLNK) command, the Get Attributes
(Qp0lGetAttr()) API, or the Properties page in IBM Navigator for i.
Related information
Work with Object Links (WRKLNK) command
Display Object Links (DSPLNK) command
Qp0lGetAttr()--Get Attributes API

Object change
A scan would occur if the object is accessed after an object has changed or has been modified.
Normally, the modification occurs in the object's data. Examples of modifications to an object are writing
to the object directly, or through memory mapping, truncating the object, or clearing the object. If the
object's CCSID attribute changes, this will also trigger a scan on the next access.

Integrated file system 19

 Signature change
A scan occurs when the object is accessed if the global signature is different from the object's signature.
The global or independent ASP group signatures represent the level of software associated with the
scan-related exit programs. The object signature reflects the global or independent ASP signature when
the object was last scanned. When an object is not in an independent ASP group, the object signature
is compared to the global scan signature. If the object is in an independent ASP, the object signature is
compared to the associated independent ASP group scan signature.
Note: In the following example, the phrases scan key and scan key signature are used. The scan key is a
method to identify one set of scanning software. An example of this is for a specific company. The scan
key signature allows the set of scanning software to indicate the level of support it provides. One example
of this is a set of virus definitions.
Here is an example of when an object is not in an independent ASP group and a scan occurs:
1. An exit program is registered to the QIBM_QP0L_SCAN_OPEN exit point. A scan key and a scan key
signature were specified as follows:

Scan key: XXXXXX

Scan key signature: 0000000000

The global scan signature is 0000 and is not updated.

2. An exit program is then registered to the QIBM_QP0L_SCAN_CLOSE exit point. A scan key and a scan
key signature were specified as follows:

Scan key: XXXXXX

Scan key signature: 1111111111

The global scan signature is then updated to 0001.

3. Next, a file is opened that currently has an object signature of 0000. The existence of the exit
programs, coupled with the difference in global scan signatures (0000 to 0001), initiates a scan. When
the scan completes successfully, the file signature is updated to 0001.
4. If the file is opened by another user, it will not be re-scanned since the object and global signatures
match.
The example below displays that the exit program wants to cause a re-scan to occur:
1. Support has been added to the system to scan for new types of viruses. The Change Scan
Signature (QP0LCHSG) API is called to update the scan keys' scan key signature. A scan key and a
scan key signature are specified as follows:

Scan key: XXXXXX

Scan key signature: 2222222222

The global scan key signature is then updated to 0002.

2. If the previously scanned file is now opened, the difference in signatures will cause a re-scan.
The example continues on to show when an object is in an independent ASP group:
1. An independent ASP is varied on for the first time and a file in the independent ASP is opened. When
the first file is opened, the independent ASP scan key list is compared to the system scan key list. The
two are different because of the fact that there is no independent ASP scan key list. In this case, the
independent ASP scan key list obtains the global scan key list. The independent ASP scan key list then
has a scan key of XXXXXX and a scan key signature of 2222222222. As a result, the independent ASP
scan signature is changed to 0001. When the file in the independent ASP is opened that currently has
an object signature of 0000, it is then compared to the independent ASP scan signature of 0001, and
because of the difference the file is scanned. When scanned successfully, the file signature is updated
to 0001.
Note: A signature change will trigger a scan unless the object has the 'object change only' attribute and
the *USEOCOATR system value specified.

20 IBM i: Integrated file system

 Related information
Integrated File System Scan on Open Exit Program
Integrated File System Scan on Close Exit Program
Change Scan Signature (QP0LCHSG) API

Different CCSID
If an object is accessed with a different coded character set identifier (CCSID) than was previously
scanned for that object, a scan would be triggered.
An example of this scan is when a file with data stored in CCSID 819 is opened in CCSID 1200, and
scanned successfully. As long as the file's data is not changed, then every time that file is opened in
CCSID 1200, a scan is not triggered. However, if that file is opened in a different CCSID, for example, 37,
a scan is triggered for that CCSID 37. If that scan is also successful, then any subsequent accesses with
CCSID 1200 and 37 will not trigger an additional scan.
Only two CCSIDs and one binary indication are kept in an effort to minimize data stored on the system. If
you typically access the same object with many different CCSIDs, then this can cause a lot of additional
scanning to occur.

During save operation

This provides yet another example of when a scan might occur. A scan can be requested when an object is
saved.
The Save Object (SAV) command now includes a SCAN parameter that allows for specification as to
whether the files will be scanned when being saved. You may also request for the object not to be saved
if it has either previously failed a scan or if it fails a scan during the save. This prevents files who fail the
scan from being put on media and possibly moved to other systems.
Note: This does not mean that when your object is restored it will be marked as having been scanned.
Whenever objects are restored, the entire scan status history is cleared.
Related information
Save Object (SAV) command

Object integrity check

Lastly, a scan can be requested if the SCANFS parameter on the Check Object Integrity
(CHKOBJITG) command is specified with a value of *YES.
This option is ideal if you want to determine whether a file is good without opening it. If SCANFS
(*STATUS) is specified, then all objects which have failed previous scans will log a scan failure violation.
Related information
Change Object Integrity (CHGOBJITG) command

File systems
A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.
Each file system has a set of logical structures and rules for interacting with information in storage. These
structures and rules may be different from one file system to another. In fact, from the perspective of
structures and rules, the IBM i support for accessing database files and various other object types through
libraries can be thought of as a file system. Similarly, the IBM i support for accessing documents (which
are really stream files) through the folders structure might be thought of as a separate file system.
The integrated file system treats the library support and folders support as separate file systems. Other
types of file management support that have differing capabilities are also treated as separate file systems.
You can interact with any of the file systems through a common interface. This interface is optimized
for the input and output of stream data, in contrast to the record input and output that is provided

Integrated file system 21

 through the data management interfaces. The provided commands, menus and displays, and application
programming interfaces (APIs) allow interaction with the file systems through this common interface.

Figure 9. File systems, file servers, and the integrated file system interface

Using Network File Systems through the integrated file system interface
The Network File System (NFS) is accessible through the integrated file system interface. Be aware of
these considerations and limitations.
Related information
Optical storage
Plan integrated file system security
Planning integrated file system security

File system comparison

These tables summarize the features and limitations of each file system.

Table 2. File system summary (Part 1 of 2)

Capability "root" (/) QOpenSys QSYS.LIB16 QDLS QNTC
Standard part of IBM i Yes Yes Yes Yes Yes
Type of file Stream Stream Record 12 Stream Stream
File size limit T2=1 TB; T2=1 TB; Database 4 GB Varies 17
T1=128 GB T1=128 GB file sizes
Access through IBM i file server Yes Yes Yes Yes Yes
Direct access through file server I/O No No No No Yes
processor 1
Comparative speed for open/close Medium 2 Medium 2 Low 2 Low 2 Medium 2

22 IBM i: Integrated file system

Table 2. File system summary (Part 1 of 2) (continued)
Capability "root" (/) QOpenSys QSYS.LIB16 QDLS QNTC
Case-sensitive name search No Yes No 4 No 5 No
Maximum length of each component in 255 255 10/6 8/3 255
path name characters 3 characters 3 characters 6 characters characters 3
7

Maximum length of path name 8 16MB 16MB 55 – 66 82 255

characters4 characters characters
Maximum length of extended attributes 2GB 2GB Varies 9 32KB 0 18
for an object
Maximum levels of directory hierarchy No limit 10 No limit 10 3 32 127
within file system
Maximum links per object 11 Varies 15 Varies 15 1 1 1
Supports symbolic links Yes Yes No No No
Object or file can have owner Yes Yes Yes Yes No
Supports integrated file system Yes Yes Yes Yes Yes
commands
Supports integrated file system APIs Yes Yes Yes Yes Yes
Supports hierarchical file system (HFS) No No No Yes No
APIs
Threadsafe 13 Yes Yes Yes No Yes
Supports object journaling Yes Yes Yes14 No No

Integrated file system 23

Table 2. File system summary (Part 1 of 2) (continued)
Capability "root" (/) QOpenSys QSYS.LIB16 QDLS QNTC

Notes:
1. The file server I/O processor is hardware used by the LAN Server.
2. The speed applies when the file system is accessed through the IBM i file server.
3. For certain CCSID values, the maximum length can be less than 255 characters.
4. The QSYS.LIB file system has a maximum path name length of 55 characters. The independent ASP
QSYS.LIB file system has a maximum path length of 66 characters.
5. See “Document library services file system (QDLS)” on page 46 for details.
6. This value can be up to 10 characters for the object name and up to 6 characters for the object type.
7. This value can be up to 8 characters for the name and 1 to 3 characters for the file type extension (if any).
8. The values are based on the assumption that an absolute path name begins with / followed by the file
system name (such as /QDLS...).
9. The QSYS.LIB and independent ASP QSYS.LIB file systems support three predefined extended
attributes: .SUBJECT, .CODEPAGE, and .TYPE. The maximum length is determined by the combined length
of these three extended attributes.
10. In practice, directory levels are limited by program and system space limits.
11. An exception to this is a directory that can have only one link to another directory.
12. The user spaces in QSYS.LIB and independent ASP QSYS.LIB file systems support stream file input and
output.
13. Integrated file system APIs are threadsafe when the operation is directed to an object that resides in a
threadsafe file system. When these APIs are operating on objects in file systems that are not threadsafe
when multiple threads are running in the job, the API will fail.
14. QSYS.LIB and independent ASP QSYS.LIB file systems support journaling different object types than the
"root" (/), UDFS, and QOpenSys file systems.
15. *TYPE2 directories have a limit of one million links per object and a limit of 999 998 subdirectories.
*TYPE1 directories have a limit of 32 767 links per object.
16. Data in this column refers to both the QSYS.LIB file system and the independent ASP QSYS.LIB file system.
17. This limit depends on the system being accessed.
18. QNTC does not support extended attributes.
Abbreviations
• T1 = *TYPE1 *STMF
• T2 = *TYPE2 *STMF
• B = bytes KB = kilobytes MB = megabytes GB = gigabytes TB = terabytes

Table 3. File system summary (Part 2 of 2)

Capability QOPT QFileSvr.400 UDFS NFS
Standard part of IBM i Yes Yes Yes Yes
Type of file Stream Stream Stream Stream
File size limit Varies16 Varies 3 T2 = 1 TB; T1=128 Varies 14
GB
Access through IBM i file Yes Yes Yes Yes
server

24 IBM i: Integrated file system

Table 3. File system summary (Part 2 of 2) (continued)
Capability QOPT QFileSvr.400 UDFS NFS
Direct access through the file No No No No
server I/O processor 1
Comparative speed for open/ Low Low 2 Medium 2 Medium 2
close
Case-sensitive name search No No 2 Yes 11 Varies 2
Maximum length of each Varies 4 Varies 2 255 characters 15 Varies 2
component in path name
Maximum length of path name 294 characters No limit 2 16 MB No limit 2
5

Maximum length of extended 8 MB 06 2 GB 10 06

attributes for an object
Maximum levels of directory No limit 7 No limit 2 No limit 7 No limit 2
hierarchy within file system
Maximum links per object 8 1 1 Varies 13 Varies 2
Supports symbolic links No No Yes Yes 2
Object or file can have owner No No 9 Yes Yes 2
Supports integrated file Yes Yes Yes Yes
system commands
Supports integrated file Yes Yes Yes Yes
system APIs
Supports hierarchical file Yes No No No 2
system (HFS) APIs
Threadsafe 12 Yes Yes Yes Yes
Supports object journaling No No Yes No

Integrated file system 25

Table 3. File system summary (Part 2 of 2) (continued)
Capability QOPT QFileSvr.400 UDFS NFS

Notes:
1. The file server I/O processor is hardware used by the LAN Server.
2. This value depends on which remote file system is being accessed.
3. When connected to a system earlier than V6R1, the file size limit is 2 GB-1. Otherwise, the file size limit
depends on the file system being accessed.
4. See “Optical file system (QOPT)” on page 48 for details.
5. The values are based on the assumption that an absolute path name that begins with / followed by the file
system name.
6. The QFileSvr.400 file system does not return extended attributes even if the file system being accessed
supports extended attributes.
7. In practice, directory levels are limited by program and system space limits.
8. An exception to this is a directory that can have only one link to another directory.
9. The file system being accessed might support object owners.
10. The maximum length of extended attributes for the UDFS itself cannot exceed 40 bytes.
11. Case sensitivity can be specified when a UDFS is created. If the *MIXED parameter is used when creating a
UDFS, it will allow a case-sensitive search.
12. Integrated file system APIs are threadsafe when they are accessed in a multithread capable process. The
file system does not allow accesses to the file systems that are not threadsafe.
13. *TYPE2 directories have a limit of one million links per object. *TYPE1 directories have a limit of 32 767
links per object.
14. This limit depends on the system being accessed.
15. For certain CCSID values, the maximum length can be less than 255 characters.
16. 9,999,999,827,968 bytes when accessing through integrated file system. 4,294,705,152 bytes when
accessing through hierarchical file system (HFS).
Abbreviations
• T1 = *TYPE1 *STMF
• T2 = *TYPE2 *STMF
• B = bytes KB = kilobytes MB = megabytes GB = gigabytes TB = terabytes

Related reference
"root" (/) file system
The "root" (/) file system takes full advantage of the stream file support and hierarchical directory
structure of the integrated file system. It also has some characteristics of the DOS file system.
Open systems file system (QOpenSys)
The QOpenSys file system is compatible with open system standards based on UNIX, such as POSIX and
X/Open Portability Guide (XPG). Like the "root" (/) file system, this file system takes advantage of the
stream file and directory support that is provided by the integrated file system.
User-defined file systems (UDFSs)
The user-defined file systems (UDFSs) reside on the auxiliary storage pool (ASP) or independent auxiliary
storage pool (ASP) of your choice. You can create and manage these file systems.
Library file system (QSYS.LIB)
The QSYS.LIB file system supports the IBM i library structure.
Independent ASP QSYS.LIB

26 IBM i: Integrated file system

 The independent ASP QSYS.LIB file system supports the IBM i library structure in independent auxiliary
storage pools (ASPs) you create and define. This file system provides access to database files and all of
the other IBM i object types that the library support manages in the independent ASPs.
Document library services file system (QDLS)
The QDLS file system supports the folders structure. It provides access to documents and folders.
Optical file system (QOPT)
The QOPT file system provides access to stream data that is stored on optical media.
IBM i NetClient file system (QNTC)
The QNTC file system provides access to data and objects that are stored on an Integrated xSeries Server
(IXS) running Windows NT 4.0 or later, or Linux operating system. The QNTC file system also provides
access to data and objects that are stored on remote servers running Windows NT 4.0 or later, Linux
Samba 3.0 or later, or supported versions of IBM i NetServer.
IBM i file server file system (QFileSvr.400)
The QFileSvr.400 file system provides transparent access to other file systems that reside on remote IBM
i platforms. It is accessed through a hierarchical directory structure.
Network File System (NFS)
The Network File System (NFS) provides the user with access to data and objects that are stored on a
remote NFS server.
Related information
Journal management

"root" (/) file system

The "root" (/) file system takes full advantage of the stream file support and hierarchical directory
structure of the integrated file system. It also has some characteristics of the DOS file system.
In addition, it:
• Is optimized for stream file input and output.
• Supports multiple hard links and symbolic links.
• Supports local sockets.
• Supports threadsafe APIs.
• Supports *FIFO objects.
• Supports the /dev/null and /dev/zero *CHRSF objects as well as other *CHRSF objects.
• Supports the journaling of object changes.
• Supports the scanning of objects using the integrated file system scan-related exit points.
The "root" (/) file system has support for the character special files (*CHRSF) called /dev/null
and /dev/zero. Character special files are associated with a device or resource of a computer system.
They have path names that appear in directories and have the same access protection as regular files.
The /dev/null or /dev/zero character special files are always empty, and any data written to /dev/
null or /dev/zero is discarded. The files /dev/null and /dev/zero have an object type of *CHRSF
and can be used like regular files, except that no data is ever read in the/dev/null file, and the /dev/
zero file always returns successfully with the data cleared to zeros.

Accessing the "root" (/) file system

The "root" (/) file system can be accessed through the integrated file system interface using either the
IBM i file server or the integrated file system commands, user displays, and APIs.

Integrated file system 27

 Case-sensitivity in the "root" (/) file system
The file system preserves the same uppercase and lowercase form in which object names are entered,
but no distinction is made between uppercase and lowercase when the system searches for names.

Path names in the "root" (/) file system

Path names have specific form in the "root" (/) file system.

/Directory/Directory . . . /Object

• Each component of the path name can be up to 255 characters long, much longer than in the QSYS.LIB
or QDLS file systems. The full path name can be extremely long, up to 16 megabytes.
• There is no limit to the depth of the directory hierarchy other than program and system space limits.
• The characters in names are converted to UCS2 Level 1 form (for *TYPE1 directories) and UTF-16 (for
*TYPE2 directories) when the names are stored.
Related concepts
Name continuity
When you use the "root" (/), QOpenSys, and user-defined file systems, you can take advantage of system
support that ensures characters in object names remain the same.
*TYPE2 directories
The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format. The *TYPE2 directory format is an enhancement of the original *TYPE1 directory
format.
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Links in the "root" (/) file system

Multiple hard links to the same object are allowed in the "root" (/) file system. Symbolic links are fully
supported.
A symbolic link can be used to link from the "root" (/) file system to an object in another file system, such
as QSYS.LIB, Independent ASP QSYS.LIB, or QDLS.
Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands in the "root" (/) file system
All of the commands listed in the Accessing using CL commands topic and the displays described in the
Accessing using menus and displays topic can operate on the "root" (/) file system. However, it might not
be safe to use these commands in a multithread-capable process.
Related tasks
Accessing using menus and displays
You can perform operations on files and other objects in the integrated file system by using a set of menus
and displays provided by your system.
Related reference
Accessing using CL commands

28 IBM i: Integrated file system

All of the operations that you can do through the integrated file system menus and displays can be done
by entering control language (CL) commands. These commands can operate on files and other objects in
any file system that are accessible through the integrated file system interface.

Use of integrated file system APIs in the "root" (/) file system
All of the APIs listed in the Performing operations using APIs topic can operate on the "root" (/) file
system.
Related reference
Performing operations using APIs
Many of the application programming interfaces (APIs) that perform operations on integrated file system
objects are in the form of C language functions.
Related information
Application programming interfaces (APIs)

Object changes journaling in the "root" (/) file system

Some object types in the "root" (/) file system can be journaled. This function enables you to recover the
changes to an object that have occurred since the object was last saved.
Related concepts
Journaling objects
The primary purpose of journaling is to enable you to recover the changes to an object that have occurred
since the object was last saved. Additionally, a key use of journaling is to assist in the replication of object
changes to another system either for high availability or workload balancing.

UDP and TCP devices in the "root" (/) file system

The "root" (/) file system under the directory of /dev/xti will now hold two device drivers named udp and
tcp.
Both of the drivers are character special files (*CHRSFs) and are created during the first initial program
load (IPL). The User Datagram Protocol (UDP) and Transmission Control Protocol (TCP) device drivers are
used to open a connection to the UDP and TCP transport providers. Both of these drivers are user devices
and receive a new device major number. They also have cloned open operations, which means that each
open operation obtains a unique instance of the device. The use of these devices is only supported in the
IBM i Portable Application Solutions Environment (PASE). The following table contains the objects that
will be created and their properties.

Table 4. Device driver objects and properties

Path name Type Major Minor Owner Owner Group Group Public
data data data
authoriti authoriti authoriti
es es es
/dev/xti *DIR N/A N/A QSYS *RWX None *RX *RX
/dev/xti/tcp *CHRSF Clone TCP QSYS *RW None *RW *RW
/dev/xti/udp *CHRSF Clone UDP QSYS *RW None *RW *RW

Related information
i5/OS PASE

Integrated file system 29

Open systems file system (QOpenSys)
The QOpenSys file system is compatible with open system standards based on UNIX, such as POSIX and
X/Open Portability Guide (XPG). Like the "root" (/) file system, this file system takes advantage of the
stream file and directory support that is provided by the integrated file system.
In addition, it:
• Is accessed through a hierarchical directory structure similar to UNIX systems.
• Is optimized for stream file input and output.
• Supports multiple hard links and symbolic links.
• Supports case-sensitive names.
• Supports local sockets.
• Supports threadsafe APIs.
• Supports *FIFO objects.
• Supports the journaling of object changes.
• Supports the scanning of objects using the integrated file system scan-related exit points.
The QOpenSys file system has the same characteristics as the "root" (/) file system, except it is case
sensitive to enable support for UNIX-based open systems standards.

Accessing QOpenSys
QOpenSys can be accessed through the integrated file system interface using either the IBM i file server
or the integrated file system commands, user displays, and APIs.

Case sensitivity in the QOpenSys file system

Unlike the "root" (/) file system, the QOpenSys file system distinguishes between uppercase or lowercase
characters when searching for object names.
For example, a character string supplied in all uppercase characters will not match the same character
string in which any of the characters is lowercase.
This case sensitivity allows you to use duplicate names, provided there is some difference in uppercase
and lowercase of the characters making up the name. For example, you can have an object named
Payroll, an object named PayRoll, and an object named PAYROLL in the same directory in QOpenSys.

Path names in the QOpenSys file system

Path names have specific form in the QOpenSys file system.

/QOpenSys/Directory/Directory/ . . . /Object

• Each component of the path name can be up to 255 characters long. The full path name can be up to 16
MB long.
• There is no limit to the depth of the directory hierarchy other than program and system space limits.
• The characters in names are converted to UCS2 Level 1 form (for *TYPE1 directories) and UTF-16 (for
*TYPE2 directories) when the names are stored.
Related concepts
Name continuity
When you use the "root" (/), QOpenSys, and user-defined file systems, you can take advantage of system
support that ensures characters in object names remain the same.
*TYPE2 directories

30 IBM i: Integrated file system

The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format. The *TYPE2 directory format is an enhancement of the original *TYPE1 directory
format.
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Links in the QOpenSys file system

Multiple hard links to the same object are allowed in the QOpenSys file system. Symbolic links are fully
supported.
A symbolic link can be used to link from the QOpenSys file system to an object in another file system.
Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands and displays in the QOpenSys file
system
All of the commands that are listed in the Accessing using CL commands topic and the displays that are
described in the Accessing using menus and displays topic can operate on the QOpenSys file system.
However, it may not be safe to use these commands in a multithread capable process.
Related tasks
Accessing using menus and displays
You can perform operations on files and other objects in the integrated file system by using a set of menus
and displays provided by your system.
Related reference
Accessing using CL commands
All of the operations that you can do through the integrated file system menus and displays can be done
by entering control language (CL) commands. These commands can operate on files and other objects in
any file system that are accessible through the integrated file system interface.

Use of integrated file system APIs in the QOpenSys file system

All the APIs listed in the Performing operations using APIs topic can operate on the QOpenSys file system.
Related reference
Performing operations using APIs
Many of the application programming interfaces (APIs) that perform operations on integrated file system
objects are in the form of C language functions.
Related information
Application programming interfaces (APIs)

Object changes journaling in the QOpenSys file system

Some object types in the QOpenSys file system can be journaled. This function enables you to recover the
changes to an object that have occurred since the object was last saved.
Related concepts
Journaling objects

Integrated file system 31

 The primary purpose of journaling is to enable you to recover the changes to an object that have occurred
since the object was last saved. Additionally, a key use of journaling is to assist in the replication of object
changes to another system either for high availability or workload balancing.

User-defined file systems (UDFSs)

The user-defined file systems (UDFSs) reside on the auxiliary storage pool (ASP) or independent auxiliary
storage pool (ASP) of your choice. You can create and manage these file systems.
In addition, they:
• Provide a hierarchical directory structure similar to PC operating systems such as DOS
• Are optimized for stream file input and output
• Support multiple hard links and symbolic links
• Support local sockets
• Support threadsafe APIs
• Support *FIFO objects
• Support the journaling of object changes
• Support the scanning of objects using the integrated file system scan-related exit points
You can create multiple UDFSs by giving each a unique name. You can specify other attributes for a UDFS
during its creation, including:
• An ASP number or independent ASP name where the objects that are located in the UDFS are stored.
• The case-sensitivity characteristics of the object names that are located within a UDFS.
The case-sensitivity of a UDFS determines whether uppercase and lowercase characters will match
when searching for object names within the UDFS.
• The create object scanning attribute which defines what the scan attribute should be for objects created
in a UDFS.
• The value for the restricted, rename and unlink attribute
• The auditing value for an object created in the UDFS.
• The different format, *TYPE1 and *TYPE2, for stream files created in the UDFS.
• The disk storage option for stream files created in the UDFS.
• The main storage option for stream files created in the UDFS.

User-defined file system concepts

In a user-defined file system (UDFS), as in the "root" (/) and QOpenSys file systems, you can create
directories, stream files, symbolic links, local sockets, and *FIFO objects.
A single block special file object (*BLKSF) represents a UDFS. As you create UDFSs, you also
automatically create block special files. The block special file is only accessible to the user through the
integrated file system generic commands, APIs, and the QFileSvr.400 interface.
Note: Changing the attributes or authority of a block special file results in equivalent changes to the root
directory of the UDFS and vice versa. The changes do not affect other objects in the UDFS.
A UDFS exists only in two states: mounted and unmounted. When you mount a UDFS, the objects within
it are accessible. When you unmount a UDFS, the objects within it become inaccessible.
In order to access the objects within a UDFS, you must mount the UDFS on a directory (for example, /
home/JON). When you mount a UDFS on a directory, the original contents of that directory, including
objects and subdirectories, become inaccessible. When you mount a UDFS, the contents of the UDFS
become accessible through the directory path that you mount the UDFS over. For example, the /
home/JON directory contains a file /home/JON/payroll. A UDFS contains three directories mail,
action, and outgoing. After mounting the UDFS on /home/JON, the /home/JON/payroll file is
inaccessible, and the three UDFS directories become accessible as /home/JON/mail, /home/JON/

32 IBM i: Integrated file system

action, and /home/JON/outgoing. After unmounting the UDFS, the /home/JON/payroll file is
accessible again, and the three directories in the UDFS become inaccessible.
An initial program load (IPL) of the system or a Reclaim Storage (RCLSTG) operation of the directories
unmounts all UDFSs. Therefore, the UDFSs need to be remounted after performing an IPL or running a
RCLSTG command on the directories.
Note: A UDFS on an independent ASP cannot be mounted over.

Accessing a user-defined file system through the integrated file system interface
A user-defined file system (UDFS) can be accessed through the integrated file system interface using
either the IBM i file server or the integrated file system commands, user displays, and APIs.
In using the integrated file system interface, you should be aware of the following considerations and
limitations.
Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.
Stream file
A stream file is a randomly accessible sequence of bytes, with no further structure imposed by the system.
Related information
Create User-Defined FS (CRTUDFS) command

Temporary user-defined file systems

The temporary user-defined file system can increase performance by reducing auxiliary storage
operations.
The integrated file system performs auxiliary storage operations to ensure that files and directories
persist across IPLs or crashes of the system. However, many applications use temporary working files
and directories that do not need to persist across IPLs of the system. These applications are needlessly
slowed down by objects being forced to permanent storage. Users can create and mount a special type of
UDFS that contains only temporary objects. Temporary objects do not require any extra auxiliary storage
operations because they are automatically deleted by the system when the system is restarted, or when
the file system is unmounted.
A temporary file system is created with the "Create User-Defined FS (CRTUDFS)" command using a new
naming convention. The names of permanent UDFSs must end with ".udfs". The names of temporary
UDFS's must end with ".tmpudfs", for example:

CRTUDFS UDFS('/dev/QASP01/new.tmpudfs')

The creator of a temporary UDFS must have *ALLOBJ special authority. Temporary UDFSs can only be
created in the system auxiliary storage pool (ASP). That is, they can only be created in '/dev/QASP01'.
After it is created, the temporary UDFS can then be mounted and used much like a permanent UDFS, with
the following restrictions:
• Temporary objects cannot be secured by authorization lists.
• User journaling of temporary objects is not allowed.
• Objects cannot be saved from, nor restored into, a temporary file system.
• Extended attributes are not supported for temporary objects.
• Object signing of temporary objects is not allowed.
• You cannot mount a temporary UDFS as a read-only file system.
Users of temporary file systems need to be aware of the following considerations:

Integrated file system 33

 • The storage for temporary objects is not accounted to the owning user profile nor to any process.
Therefore, it is possible for a user profile to create temporary objects that exceed the maximum storage
allowed for that user profile.
• When any file system is unmounted, it can cause delays in other processes that are attempting to
access mounted file systems, including system-supplied file systems such as QDLS or QSYS.LIB.
Normally these delays are brief and are not noticeable. When a temporary UDFS is unmounted, all
the contents of the UDFS are deleted. If there are large numbers of objects in the temporary UDFS,
other processes could experience significant delays accessing other file systems. If you are affected by
this consideration, it is recommended that you use the Remove Directory (RMVDIR) and Remove Link
(RMVLNK) commands as needed to delete the contents of the temporary UDFS before unmounting the
file system. The SUBTREE(*ALL) value for the RMVDIR command can be especially helpful in this regard.
• Because unmounting a temporary UDFS deletes the entire contents of the UDFS, the unmount
operation is subject to many of the same restrictions as the Delete User-Defined FS (DLTUDFS)
command. The unmount can fail if any of the following are true:
– The user does not have *OBJEXIST authority to all the objects in the UDFS.
– The user does not have *WX data authority to all the non-empty directories in the UDFS.
– Any object in the UDFS is checked out.
– Any object in the UDFS is read-only.
– The "restricted rename and unlink" attribute of a directory is "Yes", and the user is not the owner
of the directory, does not own all the objects in the directory, and does not have *ALLOBJ special
authority.
When the unmount fails for any of these reasons, diagnostic messages can appear in the job log stating
the reason for the failure, along with a message stating how many objects have been removed.
• The block special file (BLKSF) object associated with the temporary UDFS is itself a permanent object.
It is only the contents of the UDFS that are temporary. However, this BLKSF cannot be saved nor can its
attributes be changed to allow it to be saved. Also, this BLKSF cannot be secured by an authorization
list.

Case sensitivity in an integrated file system user-defined file system

You can specify whether object names in the user-defined file system (UDFS) are case sensitive or not
case sensitive when you create it.
When you select case sensitivity, uppercase and lowercase characters are distinguished when searching
for object names. For example, a name that is supplied in all uppercase characters will not match the
same name in which any of the characters are lowercase. Therefore, /home/MURPH/ and /home/murph/
are recognized as different directories. To create a case-sensitive UDFS, you can specify *MIXED for the
CASE parameter when using the Create User-Defined File System (CRTUDFS) command.
When you select not case sensitive, the system does not distinguish between uppercase and lowercase
characters during searches for names. Therefore, the system recognizes /home/CAYCE and /HOME/
cayce as the same directory, not as two separate directories. To create a case insensitive UDFS, you can
specify *MONO for the CASE parameter when using the CRTUDFS command.
In either case, the file system saves the same uppercase and lowercase forms in which the user enters
object names. The case sensitivity option only applies to how the user searches for names through the
system.
Related information
Create User-Defined FS (CRTUDFS) command

34 IBM i: Integrated file system

Path names in an integrated file system user-defined file system
A block special file (*BLKSF) represents a user-defined file system (UDFS) when the entire UDFS and all of
the objects within it need to be manipulated.
If your UDFS resides on the system or on a basic user ASP, and is not a temporary UDFS, block special file
names must be of the form

/dev/QASPXX/udfs_name.udfs

where XX is the ASP number where you store the UDFS, and udfs_name is the unique name of the UDFS
within that ASP. Note that the UDFS name must end with the .udfs extension.
If your UDFS is a temporary UDFS, block special file names must be of the form

/dev/QASP01/udfs_name.tmpudfs

where udfs_name is the unique name of the UDFS. Because temporary UDFSs can only exist on the
system ASP, the block special file can only be created in /dev/QASP01. Note that the temporary UDFS
name must end with the .tmpudfs extension.
If your UDFS resides on an independent ASP, block special file names must be of the form

/dev/asp_name/udfs_name.udfs

where asp_name is the name of independent ASP where you store the UDFS and udfs_name is the
unique name of the UDFS within that independent ASP. Note that the UDFS name must end with
the .udfs extension.
Path names for objects within a UDFS are relative to the directory over which you mount a UDFS. For
example, if you mount the UDFS /dev/qasp01/wysocki.udfs over /home/dennis, then the path
names for all objects within the UDFS will begin with /home/dennis.
Additional path name rules:
• Each component of the path name can be up to 255 characters long. The full path name can be up to 16
MB long.
• There is no limit to the depth of the directory hierarchy other than program and server space limits.
• The characters in names are converted to UCS2 Level 1 form (for *TYPE1 directories) and UTF-16 (for
*TYPE2 directories) when the names are stored.
Related concepts
Name continuity
When you use the "root" (/), QOpenSys, and user-defined file systems, you can take advantage of system
support that ensures characters in object names remain the same.
*TYPE2 directories
The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format. The *TYPE2 directory format is an enhancement of the original *TYPE1 directory
format.
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Links in an integrated file system user-defined file system

A user-defined file system (UDFS) allows multiple hard links to the same object and fully supports
symbolic links.
A symbolic link can create a link from a UDFS to an object in another file system.
Related concepts
Link

Integrated file system 35

 A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands in a user-defined file system

All of the commands that are listed in the Accessing using CL commands topic and the displays that are
described in the Accessing using menus and displays topic can operate on a user-defined file system.
There are some CL commands that are specific to the UDFS and other mounted file systems in general.
The following table describes them.

Table 5. User-defined file system CL commands

Command Description
ADDMFS Add Mounted File System. Places exported, remote server file
systems over local client directories.
CRTUDFS Create UDFS. Creates a user-defined file system.
DLTUDFS Delete UDFS. Deletes a user-defined file system.
DSPMFSINF Display Mounted File System Information. Displays information
about a mounted file system.
DSPUDFS Display UDFS. Displays information about a user-defined file system.
MOUNT Mount a File System. Places exported, remote server file systems
over local client directories. This command is an alias for the ADDMFS
command.
RMVMFS Remove Mounted File System. Removes exported, remote server file
systems from the local client namespace.
UNMOUNT Unmount a File System. Removes exported, remote server file
systems from the local client namespace. This command is an alias
for the RMVMFS command.

Note: You must mount a UDFS before any integrated file system commands can operate on the objects
that are stored in that UDFS.
Related tasks
Accessing using menus and displays
You can perform operations on files and other objects in the integrated file system by using a set of menus
and displays provided by your system.
Related reference
Accessing using CL commands
All of the operations that you can do through the integrated file system menus and displays can be done
by entering control language (CL) commands. These commands can operate on files and other objects in
any file system that are accessible through the integrated file system interface.

Use of integrated file system APIs in a user-defined file system

All of the APIs that are listed in Perform operations using APIs topic can operate on a user-defined file
system.
Note: You must mount a UDFS before any integrated file system APIs can operate on the objects that are
stored in that UDFS.
Related reference
Performing operations using APIs

36 IBM i: Integrated file system

Many of the application programming interfaces (APIs) that perform operations on integrated file system
objects are in the form of C language functions.
Related information
Application programming interfaces (APIs)

Graphical user interface for a user-defined file system

IBM Navigator for i, a web interface on an IBM i system, provides easy and convenient access to user-
defined file systems (UDFSs).
This interface enables you to create, delete, display, mount, and unmount a UDFS from a Windows client.
You can perform operations on a UDFS through IBM Navigator for i. Basic tasks include:
• “Creating a new user-defined file system” on page 138
• “Mounting a user-defined file system” on page 138
• “Unmounting a user-defined file system” on page 139

Creating an integrated file system user-defined file system

The Create User-Defined File System (CRTUDFS) command creates a file system that can be
made visible through the integrated file system namespace, APIs, and CL commands.
The ADDMFS or MOUNT commands place the user-defined file system (UDFS) on top of the already existing
local directory. You can create a UDFS in an ASP or independent ASP of your choice.
You can also specify the following items for a UDFS:
• An ASP number or independent ASP name where the objects that are located in the UDFS are stored.
• The case-sensitivity characteristics of the object names that are located within a UDFS.
The case-sensitivity of a UDFS determines whether uppercase and lowercase characters will match
when searching for object names within the UDFS.
• The create object scanning attribute which defines what the scan attribute should be for objects created
in a UDFS.
• The value for the restricted, rename and unlink attribute
• The auditing value for an object created in the UDFS.
• The different format, *TYPE1 and *TYPE2, for stream files created in the UDFS.
• The disk storage option for stream files created in the UDFS.
• The main storage option for stream files created in the UDFS.
Related information
Create User-Defined FS (CRTUDFS) command
Add Mounted FS (ADDMFS) command

Deleting an integrated file system user-defined file system

The Delete User-Defined File System (DLTUDFS) command deletes an existing, unmounted
user-defined file system (UDFS), and all the objects within it.
The command will fail if you have mounted the UDFS. Deletion of a UDFS will cause the deletion of all
objects in the UDFS. If you do not have appropriate authority to delete all of the objects within a UDFS,
then none of the objects will be deleted.
Related information
Delete User-Defined FS (DLTUDFS) command

Integrated file system 37

 Displaying an integrated file system user-defined file system
The Display User-Defined File System (DSPUDFS) command presents the attributes of an
existing user-defined file system (UDFS), whether mounted or unmounted.
The Display Mounted File System Information (DSPMFSINF) command will also present
information about a mounted UDFS as well as any mounted file system.
Related information
Display User-Defined FS (DSPUDFS) command
Display Mounted FS Information (DSPMFSINF) command

Mounting an integrated file system user-defined file system

The Add Mounted File System (ADDMFS) and MOUNT commands make the objects in a file system
accessible to the integrated file system namespace.
To mount a user-define file system (UDFS), you need to specify *UDFS for the TYPE parameter on the
ADDMFS command.
An initial program load (IPL) of the system or a Reclaim Storage (RCLSTG) operation of the directories
unmounts all UDFSs. Therefore, the UDFSs need to be remounted after performing an IPL or running a
RCLSTG command on the directories.
Note: A UDFS on an independent ASP cannot be mounted over.
Related information
Add Mounted FS (ADDMFS) command

Unmounting an integrated file system user-defined file system

The unmount command makes the contents of a user-defined file system (UDFS) inaccessible to the
integrated file system interfaces.
Note: An initial program load (IPL) of the system or a Reclaim Storage (RCLSTG) operation of the
directories unmounts all UDFSs as well.
The objects in a UDFS will not be individually accessible once the UDFS is unmounted. The Remove
Mounted File System (RMVMFS) or UNMOUNT commands will make a mounted file system
inaccessible to the integrated file system namespace. If any of the objects in the file system is in use
(for example, a file is opened) at the time of using the command, you will receive an error message. The
UDFS will remain mounted. If you have mounted over any part of the UDFS, then this UDFS cannot be
unmounted until it is uncovered.
For example, you mount a UDFS /dev/qasp02/jenn.udfs over /home/judy in the integrated file
system namespace. If you then mount another file system /pubs over /home/judy, then the contents of
jenn.udfs will become inaccessible. Furthermore, you cannot unmount jenn.udfs until you unmount
the second file system from /home/judy.
Note: A UDFS on an independent ASP cannot be mounted over.
Related information
Remove Mounted FS (RMVMFS) command

Saving and restoring an integrated file system user-defined file system

You can save and restore user-defined file system (UDFS) objects, as well as their associated authorities,
unless the UDFS is a temporary UDFS.
The Save Object (SAV) command allows you to save objects in a UDFS, whereas the Restore
Object (RST) command allows you to restore UDFS objects. Both commands function whether the
UDFS is mounted or unmounted. However, to correctly save the UDFS attributes, and not just the objects
within the UDFS, the UDFS should be unmounted.

38 IBM i: Integrated file system

 Related information
Save Object (SAV) command
Restore Object (RST) command

Object changes journaling in a user-defined file system

Some object types in user-defined file systems (UDFSs) can be journaled. This function enables you to
recover the changes to an object that have occurred since the object was last saved.
Related concepts
Journaling objects
The primary purpose of journaling is to enable you to recover the changes to an object that have occurred
since the object was last saved. Additionally, a key use of journaling is to assist in the replication of object
changes to another system either for high availability or workload balancing.

User-defined file system and independent auxiliary storage pools

When you vary on an independent auxiliary storage pool (ASP), several changes occur within the "root" (/)
file system.
These changes are:
• A directory is created inside the /dev directory for the independent ASP. The name of this directory
matches the name of the device description associated with the ASP. If this directory exists before the
vary on request, and this directory is not empty, then the vary on will proceed but you will not be able
to work with any UDFS's on the ASP. If this occurs, vary off the independent ASP, either rename the
directory or remove its contents and then try the vary on request again.
• Within the /dev/asp_name directory you will find the block special file objects associated with all
UDFS's that reside on the independent ASP. There will always be a system-provided default UDFS. The
path to the default UDFS's block special file is: /dev/asp_name/QDEFAULT.UDFS
• The default UDFS is mounted over the directory /asp_name. The /asp_name directory does not need
to exist before the vary on request. However, if it does exist, it must be empty. If it is not empty, the
ASP will still be varied on, but the default UDFS will not be mounted. If this occurs, either rename the
directory or remove its contents and then either vary off and try the vary on again or use the MOUNT
command to mount the default UDFS.
• If the independent ASP is either a primary or secondary ASP, and the default UDFS was successfully
mounted, then an additional file system will be mounted. The independent ASP QSYS.LIB file system
will be mounted over /asp_name/QSYS.LIB.
Note: This file system cannot be mounted or unmounted independently of the default UDFS. It will
always be mounted or unmounted automatically.
Related reference
Independent ASP QSYS.LIB
The independent ASP QSYS.LIB file system supports the IBM i library structure in independent auxiliary
storage pools (ASPs) you create and define. This file system provides access to database files and all of
the other IBM i object types that the library support manages in the independent ASPs.

Library file system (QSYS.LIB)

The QSYS.LIB file system supports the IBM i library structure.
This file system provides you with access to database files and all of the other IBM i object types that the
library support manages on the system and in the basic user auxiliary storage pools (ASPs).
In addition, it:
• Supports all user interfaces and programming interfaces that operate on IBM i libraries and objects in
those libraries
• Supports all programming languages and facilities that operate on database files

Integrated file system 39

 • Provides extensive administrative support for managing IBM i objects
• Supports stream I/O operations on physical file members, user spaces, and save files
Before the integrated file system was introduced in Version 3 of OS/400, the QSYS.LIB file system was the
only file system. Programmers who used languages, such as RPG or COBOL, and facilities, such as DDS,
to develop applications were using the QSYS.LIB file system. System operators who used commands,
menus, and displays to manipulate output queues were using the QSYS.LIB file system, as were system
administrators who were creating and changing user profiles.
All of these facilities and the applications based on these facilities work as they did before the
introduction of the integrated file system. These facilities cannot, however, access QSYS.LIB through
the integrated file system interface.

Accessing QSYS.LIB through the integrated file system interface

The QSYS.LIB file system can be accessed through the integrated file system interface using either the
i5/OS file server or the integrated file system commands, user displays, and APIs.

QPWFSERVER authorization list in the QSYS.LIB file system

The QPWFSERVER is an authorization list (object type *AUTL) that provides additional access
requirements for all objects in the QSYS.LIB file system being accessed through remote clients.
The authorities specified in this authorization list apply to all objects within the QSYS.LIB file system.
The default authority to this object is PUBLIC *USE authority. The administrator can use the EDTAUTL
(Edit Authorization List) or WRKAUTL (Work With Authorization List) commands to
change the value of this authority. The administrator can assign PUBLIC *EXCLUDE authority to the
authorization list so that the general public cannot access QSYS.LIB objects from remote clients.

File-handling restrictions in the QSYS.LIB file system

Here are some restrictions to be aware of when handling files in the QSYS.LIB file system.
• Logical files are not supported.
• Physical files supported for text mode access are program-described physical files containing a single
field and source physical files containing a single text field. Physical files supported for binary mode
access include externally-described physical files in addition to those files supported for text mode
access.
• Byte-range locking is not supported. For more information about byte-range locking, see the fcntl()--
Perform File Control Command topic.
• If any job has a database file member open, only one job is given write access to that file member at any
time. Other requests are allowed only read access.

Support for user spaces in the QSYS.LIB file system

QSYS.LIB supports stream input and output operations to user space objects.
For example, a program can write stream data to a user space and read data from a user space. The
maximum size of a user space is 16 776 704 bytes.
Be aware that user spaces are not tagged with a CCSID (coded character set identifier). Therefore, the
CCSID returned is the default CCSID of the job.

Support for save files in the QSYS.LIB file system

The QSYS.LIB file system supports stream I/O operations to save file objects.
For example, an existing save file has data that may be read out or copied to another file until it is
necessary to place the data into a different, existing, and empty save file object. When a save file is open
for writing, no other open instances of the file are allowed. A save file does allow multiple open instances

40 IBM i: Integrated file system

for reading, provided no job has more than one open instance of the file for reading. A save file may not
be opened for read/write access. Stream I/O operations to save file data are not allowed when multiple
threads are running in a job.
Stream I/O operations on a save file are not supported when the save file or its directory are being
exported through the Network File System. They can, however, be accessed from PC clients and through
the QFileSvr.400 file system.

Case-sensitivity in the QSYS.LIB file system

In general, the QSYS.LIB file system does not distinguish between uppercase and lowercase characters in
the names of objects.
A search for object names achieves the same result regardless of whether characters in the names are
uppercase or lowercase.
However, if a name is enclosed in quotation marks, the case of each character in the name is preserved. A
search involving quoted names, therefore, is sensitive to the case of the characters in the quoted name.

Path names in the QSYS.LIB file system

Each component of the path name must contain the object name followed by the object type of the
object.
• For example:

/QSYS.LIB/QGPL.LIB/PRT1.OUTQ

/QSYS.LIB/EMP.LIB/PAY.FILE/TAX.MBR

The object name and object type are separated by a period (.). Objects in a library can have the same
name if they are different object types, so the object type must be specified to uniquely identify the
object.
• The object name in each component can be up to 10 characters long, and the object type can be up to 6
characters long.
• The directory hierarchy within QSYS.LIB can be either two or three levels deep (two or three
components in the path name), depending on the type of object being accessed. If the object is a
database file, the hierarchy can contain three levels (library, file, member); otherwise, there can be only
two levels (library, object). The combination of the length of each component name and the number of
directory levels determines the maximum length of the path name.
If "root" (/) and QSYS.LIB are included as the first two levels, the directory hierarchy for QSYS.LIB can
be up to five levels deep.
• The characters in names are converted to CCSID 37 when the names are stored. Quoted names,
however, are stored using the CCSID of the job.
For more information about CCSIDs, see the i5/OS globalization topic.
Related concepts
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Links in the QSYS.LIB file system

Symbolic links cannot be created or stored in the QSYS.LIB file system.
The relationship between a library and objects in a library is the equivalent of one hard link between the
library and each object in the library. The integrated file system handles the library-object relationship
as a link. Thus, it is possible to link from a file system that supports symbolic links to an object in the
QSYS.LIB file system.

Integrated file system 41

 Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands and displays in the QSYS.LIB file
system
Many integrated file system commands and displays are valid in the QSYS.LIB file system.
The commands listed in “Accessing using CL commands” on page 69 can operate on the QSYS.LIB file
system, except for the following restrictions:
• The Add Link (ADDLNK) command can be used only to create a symbolic link to an object in
QSYS.LIB.
• File operations can be done only on program-described physical files and source physical files.
• The Start Journal (STRJRN) and End Journal (ENDJRN) commands cannot be used on
database physical files or libraries.
• These commands are not supported:
– Check In Object (CHKIN)
– Check Out Object (CHKOUT)
– Reclaim Object Links (RCLLNK)
The same restrictions apply to the user displays described in “Accessing using menus and displays” on
page 68.

Use of integrated file system APIs in the QSYS.LIB file system

Many integrated file system APIs are valid in the QSYS.LIB file system.
The APIs listed in “Performing operations using APIs” on page 115 can operate on the QSYS.LIB file
system, except for the following restrictions:
• File operations can be done only on program-described physical files and source physical files.
• The symlink() function can be used only to link to an object in QSYS.LIB from another file system that
supports symbolic links.
• The QjoStartJournal() and QjoEndJournal() APIs cannot be used on database physical files or
libraries.
Related information
Application programming interfaces (APIs)

Independent ASP QSYS.LIB

The independent ASP QSYS.LIB file system supports the IBM i library structure in independent auxiliary
storage pools (ASPs) you create and define. This file system provides access to database files and all of
the other IBM i object types that the library support manages in the independent ASPs.
In addition, it:
• Supports all user interfaces and programming interfaces that operate on IBM i libraries and objects in
those libraries in independent ASPs
• Supports all programming languages and facilities that operate on database files
• Provides extensive administrative support for managing IBM i objects
• Supports stream I/O operations on physical file members, user spaces, and save files

42 IBM i: Integrated file system

Accessing independent ASP QSYS.LIB through the integrated file system interface
The independent ASP QSYS.LIB file system can be accessed through the integrated file system interface
using either the i5/OS file server or the integrated file system commands, user displays, and APIs.
In using the integrated file system interfaces, you should be aware of some considerations and
limitations.

QPWFSERVER authorization list in the independent ASP QSYS.LIB file system

The QPWFSERVER is an authorization list (object type *AUTL) that provides additional access
requirements for all objects in the independent ASP QSYS.LIB file system being accessed through remote
clients.
The authorities specified in this authorization list apply to all objects within the independent ASP
QSYS.LIB file system.
The default authority to this object is PUBLIC *USE authority. The administrator can use the EDTAUTL
(Edit Authorization List) or WRKAUTL (Work With Authorization List) commands to
change the value of this authority. The administrator can assign PUBLIC *EXCLUDE authority to the
authorization list so that the general public cannot access independent ASP QSYS.LIB objects from
remote clients.

File handling restrictions in the independent ASP QSYS.LIB file system

Here are the restrictions to be aware of when handling files in the independent ASP QSYS.LIB file system
• Logical files are not supported.
• Physical files supported for text mode access are program-described physical files containing a single
field and source physical files containing a single text field. Physical files supported for binary mode
access include externally-described physical files in addition to those files supported for text mode
access.
• Byte-range locking is not supported. For more information about byte-range locking, see the fcntl()--
Perform File Control Command topic.
• If any job has a database file member open, only one job is given write access to that file member at any
time. Other requests are allowed only read access.

Support for user spaces in the independent ASP QSYS.LIB file system
Independent ASP QSYS.LIB supports stream input and output operations to user space objects.
For example, a program can write stream data to a user space and read data from a user space. The
maximum size of a user space is 16 776 704 bytes.
Be aware that user spaces are not tagged with a CCSID (coded character set identifier). Therefore, the
CCSID returned is the default CCSID of the job.

Support for save files in the independent ASP QSYS.LIB file system
The independent ASP QSYS.LIB supports stream I/O operations to save file objects.
For example, an existing save file has data that may be read out or copied to another file until it is
necessary to place the data into a different, existing, and empty save file object. When a save file is open
for writing, no other open instances of the file are allowed. A save file does allow multiple open instances
for reading, provided no job has more than one open instance of the file for reading. A save file may not
be opened for read/write access. Stream I/O operations to save file data are not allowed when multiple
threads are running in a job.
Stream I/O operations on a save file are not supported when the save file or its directory are being
exported through the Network File System. They may, however, be accessed from PC clients and through
the QFileSvr.400 file system.

Integrated file system 43

 Case-sensitivity in the independent ASP QSYS.LIB file system
In general, the independent ASP QSYS.LIB file system does not distinguish between uppercase and
lowercase characters in the names of objects.
A search for object names achieves the same result regardless of whether characters in the names are
uppercase or lowercase.
However, if a name is enclosed in quotation marks, the case of each character in the name is preserved. A
search involving quoted names, therefore, is sensitive to the case of the characters in the quoted name.

Path names in the independent ASP QSYS.LIB file system

Each component of the path name must contain the object name followed by the object type of the
object.
• For example:

/asp_name/QSYS.LIB/QGPL.LIB/PRT1.OUTQ

/asp_name/QSYS.LIB/EMP.LIB/PAY.FILE/TAX.MBR

where asp_name is the name of the independent ASP. The object name and object type are separated
by a period (.). Objects in a library can have the same name if they are different object types, so the
object type must be specified to uniquely identify the object.
• The object name in each component can be up to 10 characters long, and the object type can be up to 6
characters long.
• The directory hierarchy within independent ASP QSYS.LIB can be either two or three levels deep (two or
three components in the path name), depending on the type of object being accessed. If the object is a
database file, the hierarchy can contain three levels (library, file, member); otherwise, there can be only
two levels (library, object). The combination of the length of each component name and the number of
directory levels determines the maximum length of the path name.
If /, asp_name, and QSYS.LIB are included as the first three levels, the directory hierarchy for the
Independent ASP QSYS.LIB file system can be up to six levels deep.
• The characters in names are converted to coded character set identifier (CCSID) 37 when the names are
stored. Quoted names, however, are stored using the CCSID of the job.
For more information about CCSID, see the i5/OS globalization topic in the IBM i Information Center.
Related concepts
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Links in the independent ASP QSYS.LIB file system

Symbolic links cannot be created or stored in the independent ASP QSYS.LIB file system.
The relationship between a library and objects in a library is the equivalent of one hard link between the
library and each object in the library. The integrated file system handles the library-object relationship
as a link. Thus, it is possible to link from a file system that supports symbolic links to an object in the
independent ASP QSYS.LIB file system.
Related concepts
Link

44 IBM i: Integrated file system

A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands and displays in the independent ASP
QSYS.LIB file system
Many integrated file system commands and displays are valid in the independent ASP QSYS.LIB file
system.
Nearly all the commands listed in “Accessing using CL commands” on page 69 can operate on the
independent ASP QSYS.LIB file system. But there are a few exceptions:
• The Add Link (ADDLNK) command can be used only to create a symbolic link to an object in
independent ASP QSYS.LIB.
• File operations can be done only on program-described physical files and source physical files.
• The Start Journal (STRJRN) and End Journal (ENDJRN) commands cannot be used on
database physical files or libraries.
• You cannot move libraries in the independent ASP QSYS.LIB file system to basic auxiliary storage pools
(ASPs) using the Move Object (MOV) command. However, you can move libraries in independent ASP
QSYS.LIB to the system ASP or other independent ASPs.
• If you use Save Object (SAV) or Restore Object (RST) to save or restore library objects on an
independent ASP, then that independent ASP must be associated with the job doing the SAV or RST
operation, or the independent ASP must be specified on the ASPDEV parameter. The path name naming
convention of /asp_name/QSYS.LIB/object.type is not supported on SAV and RST.
• These commands are not supported:
– Check In Object (CHKIN)
– Check Out Object (CHKOUT)
– Reclaim Object Links (RCLLNK)
The same restrictions apply to the user displays described in “Accessing using menus and displays” on
page 68.

Use of integrated file system APIs in the independent ASP QSYS.LIB file
system
Many integrated file system APIs are valid in the independent ASP QSYS.LIB file system
The APIs listed in “Performing operations using APIs” on page 115 can operate on the independent ASP
QSYS.LIB file system, except for the following situations:
• File operations can be done only on program-described physical files and source physical files.
• The symlink() function can be used only to link to an object in independent ASP QSYS.LIB from
another file system that supports symbolic links.
• The QjoStartJournal() and QjoEndJournal() APIs cannot be used on database physical files or
libraries.
• If you use QsrSave() or QsrRestore() APIs to save or restore library objects on an independent
ASP, this independent ASP must be associated with the job doing the save or restore operation, or
the independent ASP must be specified on the ASPDEV key. The naming convention of path name
(/asp_name/QSYS.LIB/object.type) is not supported on QsrSave() and QsrRestore() APIs.
Related information
Application programming interfaces (APIs)

Integrated file system 45

Document library services file system (QDLS)
The QDLS file system supports the folders structure. It provides access to documents and folders.
In addition, it:
• Supports IBM i folders and document library objects (DLOs).
• Supports data stored in stream files.

Accessing QDLS through the integrated file system interface

The QDLS file system can be accessed through the integrated file system interface using either the IBM i
file server or the integrated file system commands, user displays, and APIs.
In using the integrated file system interfaces, you should be aware of the following considerations and
limitations.

Integrated file system and HFS in the QDLS file system

Operations can be performed on objects in the QDLS file system not only through the Document Library
Objects (DLO) CL commands but also through either the integrated file system interface or APIs provided
by a hierarchical file system (HFS).
Whereas the integrated file system is based on the Integrated Language Environment® (ILE) program
model, HFS is based on the original IBM i program model.
The HFS APIs allow you to perform a few additional operations that the integrated file system does not
support. In particular, you can use HFS APIs to access and change directory extended attributes (also
called directory entry attributes). Be aware that the naming rules for using HFS APIs are different from the
naming rules for APIs using the integrated file system interface.
Related information
Hierarchical file system APIs

User enrollment in the QDLS file system

You must be enrolled in the system distribution directory when working with objects in the QDLS file
system.

Case sensitivity in the QDLS file system

The QDLS file system converts the lowercase English alphabetic characters a to z to uppercase when
used in object names. Therefore, a search for object names using only those characters is not case
sensitive.
All other characters are case sensitive in QDLS.
Related information
Folder and document name

Path names in the QDLS file system

Each component of the path name can consist of just a name.
• For example:

/QDLS/FLR1/DOC1

or a name plus an extension (similar to a DOS file extension), such as:

/QDLS/FLR1/DOC1.TXT

46 IBM i: Integrated file system

• The name in each component can be up to 8 characters long, and the extension (if any) can be up to
3 characters long. The maximum length of the path name is 82 characters, assuming an absolute path
name that begins with /QDLS.
• The directory hierarchy within the document library services (QDLS) file system can be 32 levels deep.
If / and QDLS are included as the first two levels, the directory hierarchy can be 34 levels deep.
• The characters in names are converted to the code page of the job when the names are stored unless
data area Q0DEC500 has been created in the QUSRSYS library. If this data area exists, then the
characters in names are converted to code page 500 when the names are stored. This function provides
compatibility with the behavior of the QDLS file system in previous releases. A name may be rejected if
it cannot be converted to the appropriate code page.
For more information about code pages, see the i5/OS globalization topic in the IBM i Information
Center.
Related concepts
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Links in the QDLS file system

Symbolic links cannot be created or stored in the QDLS file system.
The integrated file system handles the relationship between a folder and document library objects in a
folder as the equivalent of one link between the folder and each object in the folder. Thus, it is possible to
link to an object in the QDLS file system from a file system that supports symbolic links.
Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands and displays in the QDLS file system
Many integrated file system commands and displays are valid in the QDLS file system.
The commands listed in “Accessing using CL commands” on page 69 can operate on the QDLS file
system, except for the following commands:
• The ADDLNK command can be used only to link to an object in QDLS from another file system that
supports symbolic links.
• The CHKIN and CHKOUT commands are supported for documents, but not for folders.
• These commands are not supported:
– APYJRNCHG
– CHGJRNOBJ
– DSPJRN
– ENDJRN
– RCLLNK
– RCVJRNE
– RTVJRNE
– SNDJRNE
– STRJRN
The same restrictions apply to the user displays described in “Accessing using menus and displays” on
page 68.

Integrated file system 47

 Use of integrated file system APIs in the QDLS file system
Many integrated file system APIs are valid in the QDLS file system.
The APIs listed in “Performing operations using APIs” on page 115 can operate on the QDLS file system,
except for the following APIs:
• The symlink() function can be used only to link to an object in QDLS from another file system that
supports symbolic links.
• The following functions are not supported:
– givedescriptor()
– ioctl()
– link()
– QjoEndJournal()
– QjoRetrieveJournalEntries()
– QjoRetrieveJournalInformation()
– QJORJIDI()
– QJOSJRNE()
– QjoStartJournal()
– readlink()
– takedescriptor()
Related information
Application programming interfaces (APIs)

Optical file system (QOPT)

The QOPT file system provides access to stream data that is stored on optical media.
In addition, it:
• Provides a hierarchical directory structure similar to PC operating systems such as DOS.
• Is optimized for stream file input and output.
• Supports data stored in stream files.

Accessing QOPT through the integrated file system

The QOPT file system can be accessed through the integrated file system using either the PC server or the
integrated file system commands, user displays, and APIs.
In using the integrated file system interface, you should be aware of the following considerations and
limitations.
Related information
Optical storage

Integrated file system and HFS in the QOPT file system

Operations can be performed on objects in the QOPT file system through either the integrated file system
interface or APIs provided by a hierarchical file system (HFS).
Whereas the integrated file system is based on the Integrated Language Environment (ILE) program
model, HFS is based on the original IBM i program model.
The HFS APIs allow you to perform a few additional operations that the integrated file system does not
support. In particular, you can use HFS APIs to access and change directory extended attributes (also
called directory entry attributes) or to work with held optical files. Be aware that the naming rules for using
HFS APIs are different from the naming rules for APIs using the integrated file system interface.

48 IBM i: Integrated file system

The maximum file size that can be accessed in the QOPT file system is different for HFS than it is for
the integrated file system. The maximum size QOPT file that is accessible through HFS is 4,294,705,152
bytes and through the integrated file system it is 9,999,999,827,968 bytes.
For more information about HFS APIs, see the Optical device programming topic collection.
Related information
Hierarchical file system APIs

Case-sensitivity in the QOPT file system

Depending on the format of the optical media, case may or may not be preserved when creating files or
directories in QOPT. However, file and directory searches are case-insensitive regardless of the optical
media format.

Path names in the QOPT file system

The path name must begin with a slash (/). The path is made up of the file system name, the volume
name, the directory and subdirectory names, and the file name.
• For example:

/QOPT/VOLUMENAME/DIRECTORYNAME/SUBDIRECTORYNAME/FILENAME

• The file system name, QOPT, is required.

• The volume and path name length vary by optical media format.
• You can specify /QOPT in the path name or include one or more directories or subdirectories in the
path name. Directory and file names allow any character except X'00' through X'3F', X'FF'. Additional
restrictions may apply based on the optical media format.
• The file name is the last element in the path name. The file name length is limited by the directory name
length in the path.
For more details on path name rules in the QOPT file system, see the "Path Name Rules" discussion in
Path names.
Related concepts
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Links in the QOPT file system

The QOPT file system supports only one link to an object. Symbolic links cannot be created or stored in
QOPT. However, files in QOPT can be accessed by using a symbolic link from the "root" (/), QOpenSys, or
user-defined file systems.
Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands and displays in the QOPT file system
Many integrated file system commands and displays are valid in the QOPT file system.
Most commands listed in “Accessing using CL commands” on page 69 can operate on the QOPT file
system. There are, however, a few exceptions in the QOPT file system. Keep in mind that it may not
be safe to use these CL commands in a multithread capable process; Certain restrictions may apply,
depending on the optical media format. The same restrictions apply to the user displays described in
“Accessing using menus and displays” on page 68.
The following integrated file system commands are not supported by the QOPT file system:

Integrated file system 49

 • ADDLNK
• APYJRNCHG
• CHGJRNOBJ
• CHKIN
• CHKOUT
• DSPJRN
• ENDJRN
• RCLLNK
• RCVJRNE
• RTVJRNE
• SNDJRNE
• STRJRN
• WRKOBJOWN
• WRKOBJPGP

Use of integrated file system APIs in the QOPT file system

Many integrated file system APIs are valid in the QOPT file system.
All of the APIs listed in “Performing operations using APIs” on page 115 can operate on the QOPT file
system in a threadsafe manner, except for the following APIs:
• QjoEndJournal()
• QjoRetrieveJournalEntries()
• QjoRetrieveJournalInformation()
• QJORJIDI()
• QJOSJRNE()
• QjoStartJournal()
• Qp0lGetPathFromFileID()
Related information
Application programming interfaces (APIs)

IBM i NetClient file system (QNTC)

The QNTC file system provides access to data and objects that are stored on an Integrated xSeries Server
(IXS) running Windows NT 4.0 or later, or Linux operating system. The QNTC file system also provides
access to data and objects that are stored on remote servers running Windows NT 4.0 or later, Linux
Samba 3.0 or later, or supported versions of IBM i NetServer.
The QNTC file system is part of the base IBM i operating system. It is not necessary to have the Integrated
Server Support, option 29 of the operating system, installed to access /QNTC.

Accessing QNTC through the integrated file system interface

By using the IBM i NetServer, IBM Navigator for i, integrated file system commands, user displays, or
APIs, you can access the QNTC file system through the integrated file system interface.
Be aware of the following considerations and limitations.

50 IBM i: Integrated file system

Authorities and ownership in the QNTC file system
The QNTC file system does not support the ownership concept of a file or directory.
Authentication on the server is performed using either the current user profile and password or network
authentication service using kerberos. If the user profile and password are used, then the same user and
password must exist on the server. To use Network Authentication Service see the article on this topic in
this section.
Attempts to use a command or API to change the ownership of files that are stored in QNTC will fail. A
system user profile, called QDFTOWN, owns all of the files and directories in QNTC.
The authority to NT server files and directories is administered from the Windows NT server. QNTC does
not support the WRKAUT and CHGAUT commands.

Case sensitivity in the QNTC file system

The QNTC file system preserves the same uppercase and lowercase form in which object names are
entered, but the QNTC file system itself does not distinguish between uppercase and lowercase in the
names. Therefore, if the server's file system is case sensitive, the path name's case must be specified
appropriately in the QNTC file system.
A search for object names achieves the same result regardless of whether characters in the names are
uppercase or lowercase.

Path names in the QNTC file system

The path consists of the file system name, the server name, the share name, the directory and
subdirectory names, and the object name.
The requirements for a path name are as follows:
• The path name must begin with a slash and can be up to 255 characters long. Path names have the
following form:

/QNTC/Servername/Sharename/Directory/ . . . /Object
(QNTC is a required part of the path name.)

• The server name is a required portion of the QNTC path name. The server name can be a TCP/IP
hostname, a NetBIOS name, or a TCP/IP address. Starting in V6R1, IPv6 addresses are supported in
addition to IPv4 addresses.
• The share name can be up to 255 characters long.
• Each component of the path name after the share name can be up to 255 characters long.
• Within QNTC, 130 levels of hierarchy are generally available. If all components of the path name are
included as hierarchy levels, the directory hierarchy can be as many as 132 levels deep.
• Names are stored in the Unicode CCSID.
• By default, each supported server that is functional in the local subnet automatically appears as a
directory under /QNTC. Use the Create Directory (CRTDIR) command or mkdir() API to add accessible
systems located outside the local subnet.
Related concepts
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.
Related information
Create Directory (MKDIR) command
mkdir()--Make Directory API
i5/OS glossary

Integrated file system 51

 Links in the QNTC file system
The QNTC file system supports only one link to an object. You cannot create or store symbolic links in
QNTC.
You can use a symbolic link from the "root" (/) or QOpenSys file system to access data in QNTC.
Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands and displays in the QNTC file system
Many integrated file system commands and displays are valid in the QNTC file system
The commands listed in “Accessing using CL commands” on page 69 can operate on the QNTC file
system, except for the following commands:
• ADDLNK
• APYJRNCHG
• CHGJRNOBJ
• CHGOWN
• CHGAUT
• CHGPGP
• CHKIN
• CHKOUT
• DSPAUT
• DSPJRN
• ENDJRN
• RCLLNK
• RCVJRNE
• RTVJRNE
• RST (available with Integrated xSeries Servers)
• SAV (available with Integrated xSeries Servers)
• SNDJRNE
• STRJRN
• WRKAUT
• WRKOBJOWN
• WRKOBJPGP
The same restrictions apply to the user displays that are described in “Accessing using menus and
displays” on page 68.

Use of integrated file system APIs in the QNTC file system

Many integrated file system APIs are valid in the QNTC file system.
The APIs listed in “Performing operations using APIs” on page 115 can operate on the QNTC file system,
except for the following APIs:
• The chmod(), fchmod(), utime(), and umask() functions will have no effect on objects in QNTC, but
attempting to use them will not cause an error.
• The QNTC file system does not support the following functions:

52 IBM i: Integrated file system

 – chown()
– fchown()
– fclear()
– fclear64()
– givedescriptor()
– link()
– QjoEndJournal()
– QjoRetrieveJournalEntries()
– QjoRetrieveJournalInformation()
– QJORJIDI()
– QJOSJRNE()
– QjoStartJournal()
– Qp0lGetPathFromFileID()
– readlink()
– symlink()
– takedescriptor()
• The QNTC file system ignores any resource limits set using the setrlimit() API when performing the
following functions:
– write()
– writev()
– pwrite()
– pwrite64()
Related information
Application programming interfaces (APIs)

QNTC environment variables

The network browsing behavior of QNTC can be controlled by two environment variables. Support for
these environment variables began in i5/OS V5R4. Use the ADDENVVAR CL command to create these
environment variables.

QZLC_SERVERLIST
When this environment variable is set to "2", all servers that appear in the /QNTC directory in the
integrated file system can be accessed by QNTC. This was the default behavior before V5R4. When this
variable is not set to "2" or has not been created, some servers that appear in the /QNTC directory might
not be accessible.

QIBM_ZLC_NO_BROWSE
When this environment variable is set to "1", the /QNTC directory will only contain servers that were
created with the CRTDIR CL command or mkdir() API. The performance of many operations against the
QNTC file system will improve when this environment variable has been set. But all /QNTC directories
need to be created using the CL command.

QIBM_ZLC_SMB_VERS
Valid QIBM_ZLC_SMB_VERS settings:
• When the environment variable does not exist or is set to “0”, the QNTC file system will negotiate a
suitable protocol version with the server.

Integrated file system 53

 • When the environment variable is set to "1", the QNTC file system will only use the SMB/CIFS protocol.
• When the environment variable is set to "2", the QNTC file system will only use SMB 2.
• When the environment variable is set to "2.1", the QNTC file system will only use SMB 2.1 and support
multi-credit operations, this can improve the performance of reading/writing operations.
• When the environment variable is set to "3", the QNTC file system will only use SMB 3 and support the
encryption of traffic between client and server, this can improve security of the QNTC file system.
• When the environment variable is set to any other value, the QNTC operation will fail.
• When the environment variable value is changed from 0 to another value, it may break connections if
the support levels of QNTC and server are incompatible.
Note:
When the environment variable is set to "2" or a higher value, any connection attempts to servers only
supporting the SMB/CIFS protocol will fail.

Creating directories in the QNTC file system

You can use the Create Directory (CRTDIR) command or mkdir() API to add a server directory to
the /QNTC directory.
By default, a QNTC directory is automatically created for all functional servers in the IBM i NetServer
domain and the local subnet. Servers outside the local subnet or IBM i NetServer domain must be added
using the CRTDIR command or mkdir() API. For example:

CRTDIR '/QNTC/NTSRV1'

adds the NTSRV1 server into the QNTC file system directory structure to enable accessing of files and
directories on that server.
You can also add a new server to the directory structure by using the TCP/IP address. The server name
can be either an IPv4 address or an IPv6 address. For example:

CRTDIR '/QNTC/9.130.67.24'

or:

CRTDIR '/QNTC/2001:0db8:3c4d:0015:0000:0000:abcd:ef12'

adds the server into the QNTC file system directory structure.
Notes:
• By configuring IBM i NetServer for WINS, it is possible to automatically create directories for servers
beyond your subnet.
• If you use the CRTDIR CL command or the mkdir() API to add directories to the directory structure,
the directories do not remain visible after performing a system IPL or after running a Reclaim Storage
(RCLSTG) command. The CRTDIR command or mkdir() API must be reissued after performing a
system IPL or running a RCLSTG command on the directories.
If you prefer to add directories using the API or CL command, you can improve the performance of these
commands by adding the environment variable QIBM_ZLC_NO_BROWSE, as in the following example:
ADDENVVAR ENVVAR(QIBM_ZLC_NO_BROWSE) VALUE(1) LEVEL(*SYS)
This environment variable causes the file system to bypass all network browsing when performing file
operations.
Related information
Create Directory (MKDIR) command
mkdir()--Make Directory API

54 IBM i: Integrated file system

Use of integrated file system APIs in the Network File System
Many integrated file system APIs are valid in the Network File System (NFS).
All of the APIs that are listed in “Performing operations using APIs” on page 115 can operate on the
network file system, except for the following APIs:
• mkfifo()
• QjoEndJournal()
• QjoRetrieveJournalEntries()
• QjoRetrieveJournalInformation()
• QJORJIDI()
• QJOSJRNE()
• QjoStartJournal()
For a complete description of the C language functions that are related specifically to the Network File

System, see i5/OS Network File System Support .

Note: A Network File System must be mounted before any APIs can be used on it.
Related information
Application programming interfaces (APIs)

Enabling QNTC file system for Network Authentication Service

The QNTC file system enables IBM i platform access to Common Integrated File System (CIFS) servers
that support the Kerberos V5 authentication protocol.
Rather than using a LAN manager type password to authenticate with each server, a properly configured
IBM i platform will now be able to access supported CIFS servers with a single logon transaction.
To enable the Network Authentication Service (NAS) for use with QNTC, you must configure these items:
• Network Authentication Service (NAS)
• Enterprise Identity Mapping (EIM)
Once the above items have been configured, you can then enable a user to use NAS with the QNTC file
system. The following steps are needed to allow a user to take advantage of the QNTC NAS support.
• The user's IBM i user profile must have the local password management (LCLPWDMGT) parameter
set to *NO. By specifying *NO, the user does not have a password to the server and cannot sign on
to a 5250 session. The only access to the server is through NAS-enabled applications, such as IBM
Navigator for i or IBM i Access 5250 Display Emulator.
If the user specifies *YES, the password is managed by the server and the user is authenticated without
NAS.
• You must have a Kerberos ticket and IBM Navigator for i connection.
• The Kerberos ticket for the IBM i platform you are using must be forwardable. To make a ticket
forwardable, follow these steps:
1. Access the Active Directory Users and Computers tool on the KDC for your NAS realm.
2. Select users.
3. Select the name that corresponds to the service principal name.
4. Select Properties.
5. Select the Account tab.
6. Select Account is trusted for delegation.
Related information
Network authentication service

Integrated file system 55

 Enterprise Identity Mapping (EIM)

IBM i file server file system (QFileSvr.400)

The QFileSvr.400 file system provides transparent access to other file systems that reside on remote IBM
i platforms. It is accessed through a hierarchical directory structure.
The QFileSvr.400 file system can be thought of as a client that acts on behalf of users to perform file
requests. QFileSvr.400 interacts with the IBM i file server on the target system to perform the actual file
operation.

Accessing QFileSvr.400 through the integrated file system interface

The QFileSvr.400 file system can be accessed through the integrated file system interface using either the
IBM i file server or the integrated file system commands, user displays, and APIs.
In using the integrated file system interfaces, you should be aware of the following considerations and
limitations.
Note: The characteristics of the QFileSvr.400 file system are determined by the characteristics of the file
system being accessed on the target server.

Case-sensitivity in the QFileSvr.400 file system

For a first-level directory, which actually represents the "root" (/) directory of the target system, the
QFileSvr.400 file system preserves the same uppercase and lowercase form in which object names are
entered.
However, no distinction is made between uppercase and lowercase when QFileSvr.400 searches for
names.
For all other directories, case-sensitivity is dependent on the specific file system being accessed.
QFileSvr.400 preserves the same uppercase and lowercase form in which object names are entered when
file requests are sent to the IBM i file server.

Path names in the QFileSvr.400 file system

Path names have specific form in the QFileSvr.400 file system.
• The form is:

/QFileSvr.400/RemoteLocationName/Directory/Directory . . . /Object

The first-level directory (that is, RemoteLocationName in the previous example) represents both of the
following attributes:
– The name of the target system that is used to establish a communications connection. The target
system name can be either of the following names:
- A TCP/IP host name (for example, beowulf.newyork.corp.com)
Note: The host name can be one that resolves to either an IPv4 or an IPv6 address provided
that the target system is also at V6R1. For releases earlier than V6R1, only IPv4 addresses are
supported.
- An SNA LU 6.2 name (for example. appn.newyork)
– The "root" (/) directory of the target system
Because of this representation, any attributes specified when the first-level directory is created are
ignored.
To use this file system, the first-level directory must be created. This can be done by using any
integrated file system interface that creates directories.

56 IBM i: Integrated file system

 Note: First-level directories are not persistent across IPLs. That is, the first-level directories must be
created again after each IPL.
• Each component of the path name can be up to 255 characters long. The full path name can be up to 16
megabytes long.
Note: The file system in which the object resides may restrict the component length and path name
length to less than the maximum allowed by QFileSvr.400.
• There is no limit to the depth of the directory hierarchy, other than program and system limits and any
limits imposed by the file system being accessed.
• The characters in names are converted to Unicode form when the names are stored.
Related concepts
Name continuity
When you use the "root" (/), QOpenSys, and user-defined file systems, you can take advantage of system
support that ensures characters in object names remain the same.
Path name
A path name (also called a pathname on some systems) tells the system how to locate an object.

Communications in the QFileSvr.400 file system

The QFileSvr.400 file system communicates in the following ways.
• TCP connections with the file server on a target server can be established only if the QSERVER
subsystem on the target server is active.
• SNA LU 6.2 connections are attempted only if there is a locally controlled session that is not in use
(for example, a session specifically established for use by the LU 6.2 connection). When establishing
LU 6.2 connections, the QFileSvr.400 file system uses a mode of BLANK. On the target system, a job
named QPWFSERV is submitted to the QSERVER subsystem. The user profile of this job is defined by the
communications entry for the BLANK mode. For more information about LU 6.2 communications, see

APPC Programming .
• File server requests that use TCP as the communications protocol are performed within the context of
the job that is issuing the request. File server requests that use SNA as the communications protocol are
performed by the IBM i system job Q400FILSVR.
• If a connection is not yet established with the target server, the QFileSvr.400 file system assumes that
the first-level directory represents a TCP/IP host name. The QFileSvr.400 file system goes through the
following steps to establish a connection with the target server:
1. Resolve the remote location name to an IP address.
Note: The remote location name can be one that resolves to either an IPv4 or an IPv6 address
provided that the target system is also at V6R1. In releases earlier than V6R1, only IPv4 addresses
are supported.
2. Connect to the host server's server mapper on well-known port 449 using the resolved IP address.
Then send a query to the server mapper for the service name "as-file." One of the following occurs as
result of the query:
– If "as-file" is in the service table on the target server, the server mapper returns the port on which
the IBM i file server daemon is listening.
– If the server mapper is not active on the target server, the default port number for "as-file" (8473)
is used.
The QFileSvr.400 file system then tries to establish a TCP connection with the IBM i file server
daemon on the target server. When the connection is established, QFileSvr.400 exchanges requests
and replies with the file server. Within the QSERVER subsystem, the QPWFSERVSO prestart requests
take control of the connection. Each prestart job runs under its own user profile.

Integrated file system 57

 3. If the remote location name is not resolved to an IP address, the first-level directory is assumed to
be an SNA LU 6.2 name. Therefore, an attempt is made to establish an APPC connection with the
IBM i file server.
• The QFileSvr.400 file system supports two different TCP communication policies, single connection
policy and multiple connection policy. For single connection policy, all of the jobs accessing one
remote server will share one connection. For multiple connection policy, each job owns one
separate connection. Single connection is the default policy. Multiple connection policy allows better
concurrency.
The QFileSvr.400 connection policy could be switched using the following steps:
1. Add/set the system level environment variable QIBM_RFS_CONNECTION_POLICY to the desired
value. The value '1' means multiple connection policy, other values are single connection policy.
There is no default value.
2. Remove all of the child directories under the '/QFileSvr.400' directory. If they are not all removed,
the policy switch will fail.
3. You must log out and log back on to enable the connection policy. The new connection policy will be
enabled when you try to access the first remote server.
Note: SNA only support single connection policy.
• The QFileSvr.400 file system periodically (every 2 hours) checks to determine if there are any
connections that are not being used (for example, no opened files associated with the connection)
and for connections that have had no activity during a 2-hour period. If such a connection is found, the
connection is ended.
• The QFileSvr.400 file system cannot detect loops. The following path name is an example of a loop:

/QFileSvr.400/Remote2/QFileSvr.400/Remote1/QFileSvr.400/Remote2/...

where Remote1 is the local system. When the path name that contains a loop is specified, the
QFileSvr.400 file system returns an error after a brief period of time. The error indicates that a time-out
has occurred.
• The QFileSvr.400 file system will use an existing free session when communicating over SNA. It is
necessary to start the mode and establish a session for the QFileSvr.400 to successfully connect to the
remote communications system.
• QFileSvr.400 connections may be secured with Transport Layer Security. See "Securing QFileSvr.400
using Transport Layer Security " for additional information and configuration details.
You can use Transport Layer Security (TLS) connections to encrypt data transferred through the
QFileSvr.400 file system to IBM i hosting server by performing a few tasks.
– Use Digital Certificate manager (DCM) to assign a certificate to the server application.
– Import the Certificate Authority (CA) certificate that issued the server certificate into the QFileSvr.400
client system certificate store.
– Since the secure environment cannot be shared between jobs, you must define and set the
QIBM_P0A_FILESYS_SEC_ENABLE environment variable to a value of ‘1’.
– Set the QIBM_RFS_CONNECTION_POLICY environment variable to value of ‘1’ to use the multiple
connection policy.
Once configured, all users of QFileSvr.400 file system will be using secure connections. See section
“Configuration details for securing the QFileSvr.400 file system” for additional information.
Note: The file server should be set up before using secure connections. For more detail on
configuring the file server see section https://www.ibm.com/support/knowledgecenter/ssw_ibm_i_74/
rzahh/sslreq.htm

58 IBM i: Integrated file system

Security and object authority in the QFileSvr.400 file system
If both of the systems have Network Authentication Service and Enterprise Identity Mapping (EIM)
configured, and the user has authenticated with Kerberos, then Kerberos can be used to authenticate to
access a file system that resides on a target IBM i platform.
• If the Kerberos authentication fails, then the user ID and password can be used to verify access.
Note: If the ticket-granting ticket or the IBM i ticket expires after the target system has verified your
access, the expiration will not be effective until the connection to the target system has ended.
• To access a file system that resides on a target IBM i platform, you must have a user ID and password
on the target system that matches the user ID and password on the local system if Kerberos is not used
to authenticate.
Note: If your password on the local or target system is changed after the target system has verified your
access, then the change is not reflected until the connection to the target system has ended. However,
there is no delay if your user profile on the local system is deleted and another user profile is created
with the same user ID. In this case, the QFileSvr.400 file system verifies that you have access to the
target system.
• Object authority is based on the user profile that resides on the target system. That is, you are allowed
to access an object in the file system on the target system only if your user profile on the target system
has the appropriate authority to the object.
Related information
Network authentication service
Enterprise Identity Mapping (EIM)

Securing QFileSvr.400 file system connections

Securing the QFileSvr.400 file system connections.
QFileSvr.400 supports secure connections. Users can use secure connections to communicate to the
server once the environment is successfully configured. Follow the process to configure the secure
connection environment in Digital Certificate Manager (DCM):
1. Transport Layer Security on QFileSvr.400 client
To use secure connections, both environment variables QIBM_P0A_FILESYS_SEC_ENABLE and
QIBM_RFS_CONNECTION_POLICY must be to ‘1’ at the system level on the client.
2. To start DCM for the QFileSvr.400 client in your browser
• 5770SS1 Option 34 must be installed.
• In your browser, go to http://system:2001
• Log in to the system (your userid must have *SECOFR authority)
• Select Digital Certificate Manager in IBM i tasks
3. Create the *SYSTEM certificate store on QFileSvr.400 client
To participate in a Transport Layer Security (TLS) session, the QFileSvr.400 client must be able to
accept and authenticate the certificate that the file server presents. To authenticate the certificate,
the QFileSvr.400 client must have a copy of the issuing Certificate Authority (CA) certificate in the
*SYSTEM certificate store.
Assuming the Digital Certificate Manager (DCM) has not been previously used to create or manage
certificates, the QFileSvr.400 client must create the *SYSTEM certificate store by following these
steps:
• In the DCM navigation frame, select Create New Certificate Store and select *SYSTEM as the
certificate store to create and click Continue.
• Select No to create a certificate as part of creating the *SYSTEM certificate store and click Continue.
• Specify a password for the new certificate store and click Continue to display a confirmation page.

Integrated file system 59

 • Click OK
4. Importing the server’s CA certificate into QFileSvr.400 client’s *SYSTEM certificate store
The QFileSvr.400 client's *SYSTEM certificate store must contain a copy of the server’s CA certificate in
the client’s *SYSTEM certificate store. The certificate file needs to reside in an IFS directory accessible
to DCM.
To import the local certificate authority certificate into the *SYSTEM certificate store :
a. In the DCM navigation frame, click Select a Certificate Store and select *SYSTEM as the certificate
store to open.
b. When the Certificate Store and Password page displays, provide the password that was specified
for the certificate store when it was created, and click Continue.
c. After the navigation frame refreshes, select Manage Certificates to display a list of tasks.
d. From the task list, select Import certificate.
e. Select Certificate Authority (CA) as the certificate type and click Continue.
f. Specify the fully qualified path and file name for the CA certificate file and click Continue.
A message displays that either confirms that the import process succeeded or provide error
information if the process failed.
5. Specify the server’s CA as a trusted CA for QFileSvr.400 client
Before the QFileSvr.400 client can make secure connections to the server, QFileSvr.400 client must
use Digital Certificate Manager (DCM) to specify which certificate authorities (CA) the client should
trust.
a. In the DCM navigation frame, click Select a Certificate Store and select *SYSTEM as the certificate
store to open.
b. When the Certificate Store and Password page displays, provide the password that was specified
for the certificate store when it was created, and click Continue.
c. After the navigation frame refreshes, select Manage Applications to display a list of tasks.
d. From the task list, select Define CA trust list.
e. Select Client as the type of application for which you want to define the list and click Continue.
f. Select the QFileSvr.400 client application (QIBM_QP0A_FILESYS _CLIENT) from the list and click
Continue to display a list of CA certificates.
g. Select server’s CA certificate that was imported previously and click OK. DCM displays a message
to confirm the trust list selection.
With these steps complete, the server can establish a secure session with QFileSvr.400 client.

Links in the QFileSvr.400 file system

The QFileSvr.400 file system supports only one link to an object.
Symbolic links cannot be created or stored in QFileSvr.400. However, files in QFileSvr.400 can be
accessed by using a symbolic link from the "root" (/), QOpenSys, or user-defined file systems.
Related concepts
Link

60 IBM i: Integrated file system

A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands and displays in the QFileSvr.400 file
system
Many integrated file system commands and displays are valid in the QFileSvr.400 file system.
The commands listed in “Accessing using CL commands” on page 69 can operate on the QFileSvr.400
file system, except for the following commands:
• ADDLNK
• APYJRNCHG
• CHGAUT
• CHGJRNOBJ
• CHGOWN
• DSPAUT
• DSPJRN
• ENDJRN
• RCLLNK
• RCVJRNE
• RST
• RTVJRNE
• SAV
• SNDJRNE
• STRJRN
• WRKOBJOWN
• WRKOBJPGP
The same restrictions apply to the user displays described in “Accessing using menus and displays” on
page 68.

Use of integrated file system APIs in the QFileSvr.400 file system

Many integrated file system APIs are valid in the QFileSvr.400 file system.
The APIs listed in “Performing operations using APIs” on page 115 can operate on the QFileSvr.400 file
system, except for the following APIs:
• chown()
• fchown()
• givedescriptor()
• link()
• mkfifo()
• QjoEndJournal()
• QjoRetrieveJournalEntries()
• QjoRetrieveJournalInformation()
• QJORJIDI()
• QJOSJRNE
• QjoStartJournal

Integrated file system 61

 • Qp0lGetPathFromFileID()
• symlink()
• takedescriptor()
Related information
Application programming interfaces (APIs)

Network File System (NFS)

The Network File System (NFS) provides the user with access to data and objects that are stored on a
remote NFS server.
An NFS server can export a Network File System that NFS clients will then mount dynamically.
In addition, any file system mounted locally through the Network File System will have the features,
characteristics, limitations, and dependencies of the directory or file system it was mounted from on the
remote server. Operations on mounted file systems are not performed locally. Requests flow through the
connection to the server and must obey the requirements and restrictions of the type of file system on the
server.

Accessing NFS file systems through the integrated file system interface
The NFS is accessible through the integrated file system interface. Be aware of these considerations and
limitations.

Characteristics of the Network File System

The characteristics of any file system mounted through NFS are dependent on the type of file system that
was mounted from the server.
It is important to realize that requests performed on what appears to be a local directory or file system
are really operating on the server through the NFS connection.
This client/server relationship can be confusing. Consider, for example, that you mounted the QDLS file
system from the server on top of a branch of the "root" (/) directory of your client. Although the mounted
file system appears to be an extension of the local directory, it actually functions and performs as the
QDLS file system.
Realizing this relationship for file systems mounted through NFS is important for processing requests
locally and through the server connection. Just because a command processes correctly on the local level
does not mean that it will work on the directory mounted from the server. Each directory mounted on the
client will have the properties and characteristics of the server file system.

Variations of servers and clients in the Network File System

There are three major possibilities for client/server connections that can affect how the Network File
System (NFS) will function and what its characteristics will be.
The possibilities are:
• The user mounts a file system from a IBM i platform on a client.
• The user mounts a file system from a UNIX platform on a client.
• The user mounts a file system on a client from a system that is neither a IBM i platform nor a UNIX
platform.
In the first scenario, the mounted file system behaves on the client similarly to how it behaves on the
IBM i platform. However, both the characteristics of the Network File System and the file system being
served need to be taken into account. For example, if you mount the QDLS file system from the server
to the client, it has the characteristics and limitations of the QDLS file system. For instance, in the QDLS
file system, path name components are limited to 8 characters plus a 3-character extension. However,

62 IBM i: Integrated file system

the mounted file system also has NFS characteristics and limitations. For example, you cannot use the
CHGAUD command to change the auditing value of an NFS object.
In the second scenario, it is important to realize that any file system mounted from a UNIX server behaves
most similarly to the IBM i QOpenSys file system.
In the third scenario, you will need to review the documentation for the file system associated with the
operating system.
Related reference
Open systems file system (QOpenSys)
The QOpenSys file system is compatible with open system standards based on UNIX, such as POSIX and
X/Open Portability Guide (XPG). Like the "root" (/) file system, this file system takes advantage of the
stream file and directory support that is provided by the integrated file system.

Links in the Network File System

Generally, multiple hard links to the same object are allowed in the Network File System.
Symbolic links are fully supported. A symbolic link can be used to link from a Network File System to an
object in another file system. The capabilities for multiple hard links and symbolic links are completely
dependent on the file system that is being mounted with NFS.
Related concepts
Link
A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.

Use of integrated file system commands in the Network File System

Many integrated file system commands are valid in the Network File System (NFS).
All of the commands that are listed in “Accessing using CL commands” on page 69 and the displays
that are described in “Accessing using menus and displays” on page 68 can operate on the Network File
System, except for the following commands:
• APYJRNCHG
• CHGJRNOBJ
• CHGAUD
• CHGATR
• CHGAUT
• CHGOWN
• CHGPGP
• CHKIN
• CHKOUT
• DSPJRN
• ENDJRN
• RCLLNK
• RCVJRNE
• RTVJRNE
• SNDJRNE
• STRJRN
There are some CL commands that are specific to the Network File System and other mounted file
systems in general. However, it might not be safe to use these commands in a multithread capable
process. The following table describes these commands.

Integrated file system 63

 Table 6. Network file system CL commands
Command Description
ADDMFS Add Mounted File System. Places exported, remote server file systems over
local client directories.
CHGNFSEXP Change Network File System. Export Adds or removes directory trees to the
export table of file systems that are exported to Network File System clients.
DSPMFSINF Display Mounted File System Information. Displays information about a
mounted file system.
ENDNFSSVR End Network File System Server. Ends one or all of the Network File System
daemons on the server.
EXPORTFS Export a File System. Adds or removes directory trees to the export table of
file systems that are exported to Network File System clients.
MOUNT Mount a File System. Places exported, remote server file systems over local
client directories. This command is an alias for the ADDMFS command.
RLSIFSLCK Release Integrated File System Locks. Releases all Network File System
byte-range locks held by a client or on an object.
RMVMFS Remove Mounted File System. Removes exported, remote server file
systems from the local client namespace.
STRNFSSVR Start Network File System Server. Starts one or all of the Network File
System daemons on the server.
UNMOUNT Unmount a File System. Removes exported, remote server file systems
from the local client namespace. This command is an alias for the RMVMFS
command.

Note: A Network File System must be mounted before any commands can be used on it.
Related information
i5/OS Network File System Support PDF

Use of integrated file system APIs in the Network File System

Many integrated file system APIs are valid in the Network File System (NFS).
All of the APIs that are listed in “Performing operations using APIs” on page 115 can operate on the
network file system, except for the following APIs:
• mkfifo()
• QjoEndJournal()
• QjoRetrieveJournalEntries()
• QjoRetrieveJournalInformation()
• QJORJIDI()
• QJOSJRNE()
• QjoStartJournal()
For a complete description of the C language functions that are related specifically to the Network File

System, see i5/OS Network File System Support .

Note: A Network File System must be mounted before any APIs can be used on it.
Related information
Application programming interfaces (APIs)

64 IBM i: Integrated file system

Comparison of Network File System version 4 to Prior Versions
Summary of differences between NFSv4 and NFS versions 2 and 3.
Beginning with IBM i 6.1, the Network File System version 4 (NFSv4) is included in the base operating
system. Here is a short summary of the differences between NFSv4 and NFS versions 2 and 3:
• The NFSv4 protocol introduces state, whereas the NFS version 2 and 3 protocols are stateless. The
usage information of an object by an NFSv4 client is maintained by the server. Operations through the
NFSv4 protocol, such as opening, locking, reading, and writing, carry state information that notify the
server of the intentions on the object by the client. The server can then return information to a client
about other clients that also have intentions on the same object. The use of a persistent open on the
server by the NFSv4 protocol avoids some situations where an NFS version 2 or 3 client could become
locked out of a file while writing to it.
• The NFSv4 protocol includes support for byte-range locking and share modes as part of the base
protocol. Locking in NFSv4 is lease-based, which requires the NFSv4 client to maintain contact with the
server to preserve open and lock state owned by the client.
• The NFSv4 protocol introduces a compound request format. An NFSv4 client can combine several
simple operations (for example, LOOKUP, OPEN, and READ) into a single request to the server. The
single request allows NFSv4 to perform a complex operation in one network exchange.
• The NFSv4 protocol specifies improved security mechanisms over the others required by previous
protocol versions. IBM i provides support for Kerberos 5 authentication and data protection in addition
to the traditional AUTH_SYS security. The security API used by NFSv4 allows for easy addition of new
security mechanisms in the future.
• The NFSv4 protocol standardizes the representation of string data. All string data used by the protocol
is represented in UTF-8 as it crosses the network. User and group information is passed between the
client and server in string form, not as numeric values as in previous versions.
• The NFSv4 protocol combines the separate component protocols of previous NFS versions into a single
protocol specification. The single point of contact for the NFSv4 protocol allows for better compatibility
with network fire walls.
• NFS Version 4 requires support of RPC over streaming network transport protocols such as TCP. The
NFSv4 support provided by IBM i uses TCP exclusively.
For more in-depth information about the NFSv4 protocol, read RFC 3530 http://www.ietf.org/rfc/
rfc3530.txt.

Setting up a network for RPCSEC-GSS

Setting up a network for RPCSEC-GSS.
The network that is being set up in this scenario contains five servers and is configured for RPCSEC-GSS.
About this task
The five servers on the network are as follows:
• kdc.rochester.ibm.com
• alpha.rochester.ibm.com
• beta.rochester.ibm.com
• gamma.rochester.ibm.com
• zeta.rochester.ibm.com
The system kdc.rochester.ibm.com is configured as the Key Distribution Center (KDC) server, and the
Kerberos realm ROCHESTER.IBM.COM is created, on which all the systems except kdc.rochester.ibm.com
and zeta.rochester.ibm.com are NFS servers offering file systems exported with RPCSEC-GSS.
In addition, this network has the following users, which have been configured on some of the systems:
• adam

Integrated file system 65

 • brian
• charlie
• dave
• eric
Note: The following setup is provided only as an example, and is not appropriate for all environments. See
Network authentication service before attempting to set up a new Kerberos realm.
Note: Kerberos requires that the system time is reasonably the same throughout the network. Before
beginning this procedure, you can set up a mechanism to automatically synchronize time throughout the
network, such as NTP.
1. Set up the KDC server as described in the Configuring network authentication service topic.
Note: The KDC server is not ideally used for any other purpose. If the KDC is compromised, all
Kerberos principals are compromised.
2. Create principals for each user and host. This example creates Kerberos principals that match the
user profile name of the associated user. The principal name is mapped to the user name by NFS
to determine the local credential associated with the principal. For a description of how to use more
general mappings between principals and user names, see Identity mapping. For this network, we
created the following principals:
• adam
• brian
• charlie
• dave
• eric
• nfs/alpha.rochester.ibm.com
• nfs/beta.rochester.ibm.com
• nfs/gamma.rochester.ibm.com
Note: For this simplified scenario, the chosen user principal names must match the corresponding user
names on the IBM i. NFS uses the principal name as a user name to obtain user and group IDs on the
local system. If the names do not match, the access is treated as an anonymous access.
The KDC is now configured.
3. Each NFS client and server is now configured as Kerberos clients by using the IBM i Network
Authentication Service configuration wizard. If the machine is to act as a server, be sure to select
the option in the configuration wizard to create the NFS service principals. See Configuring network
authentication service for more information.
It is now possible to kinit as any of the user principals on any of the configured systems. For example,
to kinit as user adam, run the following command in a Qshell environment:
kinit adam
You must specify adam's Kerberos, not IBM i, password.
4. Repeat this setup for each system.
5. Ensure that the NFS GSS Daemon (QNFSGSSD) is started on both client and server machines.
6. Now the NFS server is ready to work, although, all users are mapped to the NFS anonymous profile
(QNFSANON). It is advisable to have all the users exist on all servers with the same uid and gid;
any users that do not exist now have access to the exported directory only as QNFSANON. To get
user names to map properly, you must configure the NFS registry daemon. See Identity Mapping for
information about configuring the NFS registry daemon.
Note: Before attempting any Kerberos client operations, the user must use kinit to obtain valid
credentials.

66 IBM i: Integrated file system

 Identity mapping
Identity mapping provides a method for the local NFS server and client to translate foreign users and
groups to local users and groups.
IBM i uses Enterprise Identity Mapping (EIM) technology, which is based on LDAP, to perform its identity
mapping. All NFS identity mapping data is stored on an LDAP server.
For simple environments where all clients and servers exist in a single NFS domain namespace that
matches the DNS suffix configured for the machine under CFGTCP option 12, EIM configuration is not
necessary. In that case, IBM i uses local name resolution to convert string representations of users and
groups to native user identifiers.
For environments where the client and server do not participate in the same NFS domain namespace or
where Kerberos 5 is used, the EIM service must be configured.
If the IBM i machine is not currently part of an EIM domain, the system must either be joined to an
existing EIM domain or a new one must be created. See Configuring Enterprise Identity Mapping.
Once the initial EIM configuration is complete or if the IBM i machine is already part of an EIM domain, it
is necessary to add the correct NFS registries to the domain.
For NFSv4 on IBM i, user name mappings must be located in registries with the ‘NFS_' prefix. For
example, when searching for a user mapping for the ‘rochester.ibm.com' namespace, IBM i expects the
registry name to be ‘NFS_rochester.ibm.com'.
Group name mappings must be located in registries with the ‘NFSGR_' prefix. For example, when
searching for a group mapping for the ‘rochester.ibm.com' namespace, IBM i expects the registry name to
be ‘NFSGR_rochester.ibm.com'.
See Managing Enterprise Identity Mapping registry definitions for information about how to add the
appropriate registries to the EIM domain.
After the configuration step is complete, the EIM administrator can populate the LDAP server with NFS
identity mapping data. See Managing Enterprise Identity Mapping identifiers for information about how to
work with EIM identifiers.
After IBM i has been configured to use the mapping data in EIM, the NFS registry daemon (QNFSRGYD)
needs to be restarted. The NFS registry daemon checks for the availability of an EIM server upon startup,
and if one is found, mapping functions use the EIM.

Accessing the integrated file system

All of the user interfaces, such as menus, commands, and displays, that are used to work with your
system's libraries, objects, database files, folders, and documents still operate as they did before the
introduction of the integrated file system.
These interfaces, however, cannot be used to work with the stream files, directories, and other objects
supported by the integrated file system.
A separate set of user interfaces is provided for the integrated file system. These interfaces can be used
on objects in any file system that can be accessed through the integrated file system directories.
You can interact with the directories and objects of the integrated file system from your system by using
menus and displays or by using control language (CL) commands. There are many IBM i (SQL) services
that work with the integrated file system. Additionally, you can use application programming interfaces
(APIs) to take advantage of the stream files, directories, and other support of the integrated file system.
You can also interact with the integrated file system through IBM Navigator for i, a web interface used for
managing and administering your system.

Integrated file system 67

Accessing using menus and displays
You can perform operations on files and other objects in the integrated file system by using a set of menus
and displays provided by your system.

About this task

To display integrated file system menus:

Procedure
1. Sign on to your system.
2. Press Enter to continue.
3. From the main menu, select the Files, Libraries, and Folders option.
4. From the Files, Libraries, and Folders menu, select the Integrated File System option.

Results
From here, you can work with Directory commands, Object commands, or Security commands in the
integrated file system, depending on your needs. However, if you know the CL command you will be using,
you can type it at the command line at the bottom of the screen and press Enter, bypassing the menu of
options.
Additionally, you can access the integrated file system from any menu on your system by performing the
following steps:
1. Type GO DATA on any command line to display the Files, Libraries, and Folders menu.
2. Select Integrated file system.
To see a menu of Network File System commands, type GO CMDNFS on any command line. To see a menu
of user-defined file system commands, type GO CMDUDFS on any command line.
From the integrated file system menus, you can request displays or commands on which you can do the
following operations:
• Create, convert, and remove a directory
• Display and change the name of the current directory
• Add, display, change, and remove object links
• Copy, move, and rename objects
• Check out and check in objects
• Save (back up) and restore objects
• Display and change object owners and user authorities
• Display and change attributes of objects
• Copy data between stream files and database file members
• Create, delete, and display the status of user-defined file systems
• Export file systems from a server
• Mount and unmount file systems on a client
Some file systems do not support all of these operations.
Related concepts
File systems
A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.
Related reference
Path name rules for CL commands and displays

68 IBM i: Integrated file system

 When using a CL command or display to operate on an object, you identify the object by supplying its path
name.
Accessing using CL commands
All of the operations that you can do through the integrated file system menus and displays can be done
by entering control language (CL) commands. These commands can operate on files and other objects in
any file system that are accessible through the integrated file system interface.

Accessing using CL commands

All of the operations that you can do through the integrated file system menus and displays can be done
by entering control language (CL) commands. These commands can operate on files and other objects in
any file system that are accessible through the integrated file system interface.
Table 1 summarizes the integrated file system commands. For more information about CL commands that
are specifically related to user-defined file systems, the Network File System, and mounted file systems
in general, see “User-defined file systems (UDFSs)” on page 32 and “Network File System (NFS)” on
page 62. Where a command performs the same operation as a DOS command, an alias (an alternative
command name) is provided for the convenience of DOS users.

Table 7. Integrated file system commands

Command Description Alias
ADDLNK3 Add Link. Adds a link between a directory and an object.
ADDMFS3 Add Mounted File System. Places exported, remote server MOUNT
file systems over local client directories.
APYJRNCHG2 3 Apply Journaled Changes. Uses journal entries to apply
changes that have occurred since a journaled object was
saved or to apply changes up to a specified point.
CHGATR3 Change Attribute. Changes an attribute for a single object, a
group of objects, or a directory tree where the directory, its
contents, and the contents of all of its subdirectories have
the attribute changed.
CHGAUD3 Change Auditing Value. Turns auditing on or off for an
object.
CHGAUT3 Change Authority. Gives specific authority for an object to a
user or group of users.
CHGCURDIR3 Change Current Directory. Changes the directory to be used CD, CHDIR
as the current directory.
CHGJRNOBJ2 3 Change Journaled Objects. Changes the journaling
attributes of an object or list of objects without the need
to end and restart journaling for the object.
CHGNFSEXP Change Network File System Export. Adds directory trees to EXPORTFS
or removes them from the export table that is exported to
NFS clients.
CHGOWN3 Change Owner. Transfers object ownership from one user to
another.
CHGPGP3 Change Primary Group. Changes the primary group from one
user to another.
CHKIN3 Check In. Checks in an object that was previously checked
out.

Integrated file system 69

 Table 7. Integrated file system commands (continued)
Command Description Alias
CHKOBJITG3 Check Object Integrity. Checks objects for any integrity
violations.
CHKOUT3 Check Out. Checks out an object, which prevents other
users from changing it.
CPY3 Copy. Copies a single object or a group of objects. COPY
CPYFRMSTMF3 Copy from Stream File. Copies data from a stream file to a
database file member.
CPYTOSTMF3 Copy to Stream File. Copies data from a database file
member to a stream file.
CRTDIR3 Create Directory. Adds a new directory to the system. MD, MKDIR
CRTUDFS3 Create UDFS. Creates a User-Defined File System.
CVTDIR Convert directory. Provides information about converting
integrated file system directories from *TYPE1 format to
*TYPE2 format.
CVTRPCSRC Convert RPC Source. Generates C code from an input file RPCGEN
written in the Remote Procedure Call (RPC) language.
DLTUDFS3 Delete UDFS. Deletes a User-Defined File.
DSPATR Display Attributes. Displays the attributes of a single object,
all of the objects matching a specified pattern, or all of the
objects in a specified subtree.
DSPAUT Display Authority. Shows a list of authorized users of an
object and their authorities for the object.
DSPCURDIR Display Current Directory. Shows the name of the current
directory.
DSPJRN2 3 Display Journal. Converts journal entries (contained in
one or more receivers) into a form suitable for external
representation.
DSPLNK Display Object Links. Shows a list of objects in a directory
and provides options to display information about the
objects.
DSPF Display Stream File. Displays a stream file or a database file.
DSPMFSINF Display Mounted File System Information. Displays STATFS
information about a mounted file system.
DSPUDFS Display UDFS. Displays User-Defined File System.
EDTF Edit Stream File. Edits a stream file or a database file.
ENDJRN2 3 End Journal. Ends the journaling of changes for an object or
list of objects.
ENDNFSSVR End Network File System Server. Ends one or all of the NFS
daemons on the server and the client.
ENDRPCBIND End RPC Binder Daemon. Ends the Remote Procedure Call
(RPC) RPCBind daemon.

70 IBM i: Integrated file system

Table 7. Integrated file system commands (continued)
Command Description Alias
MOV3 Move. Moves an object to a different directory. MOVE
PRTDIRINF Print Directory Information. Used to print directory
information for objects in the integrated file system
that was collected by the Retrieve Directory Information
(RTVDIRINF) command.
RCLLNK3 Reclaim Object Links. Identifies and, if possible, corrects
problems in mounted file systems that are in use.
RCVJRNE2 3 Receive Journal Entry. Allows a specified user exit program
to continuously receive journal entries.
RLSIFSLCK3 Release Integrated File System Locks. Releases all byte-
range locks held by an NFS client or on an object.
RMVDIR3 Remove Directory. Removes a directory from the system. RD, RMDIR
RMVLNK3 Remove Link. Removes the link to an object. DEL, ERASE
RMVMFS3 Remove Mounted File System. Removes exported, remote UNMOUNT
server file systems from the local client directories.
RNM3 Rename. Changes the name of an object in a directory. REN
RPCBIND Start RPC Binder Daemon. Starts the Remote Procedure Call
(RPC) RPCBind Daemon.
RST3 Restore. Copies an object or group of objects from a backup
device to the system.
RTVCURDIR Retrieve Current Directory. Retrieves the name of the
current directory and puts it into a specified variable (used
in CL programs)
RTVDIRINF Retrieve Directory Information. Collects attributes for
objects in the integrated file system.
RTVJRNE2 3 Retrieve Journal Entry. Gets a particular journal entry and
places the results in CL variables.
SAV3 Save. Copies an object or group of objects from the system
to a backup device.
SNDJRNE2 3 Send Journal Entry. Adds user journal entries, which are
optionally associated with a journaled object, to a journal
receiver.
STRJRN2 3 Start Journal. Starts journaling changes (made to an object
or list of objects) to a specific journal.
STRNFSSVR Start Network File System Server. Starts one or all of the
NFS daemons on the server and client.
WRKAUT Work with Authority. Shows a list of users and their
authorities and provides options for adding a user, changing
a user authority, or removing a user.
WRKLNK Work with Object Links. Shows a list of objects in a directory
and provides options for performing actions on the objects.

Integrated file system 71

 Table 7. Integrated file system commands (continued)
Command Description Alias
WRKOBJOWN1 Work with Objects by Owner. Shows a list of objects owned
by a user profile and provides options for performing actions
on the objects.
WRKOBJPGP1 Work with Objects by Primary Group. Shows a list of objects
controlled by a primary group and provides options for
performing actions on the objects.

Notes:
1. The WRKOBJOWN and WRKOBJPGP commands can display all object types but may not be fully
functional in all file systems.
2. See Journal management in the IBM i Information Center for more information.
3. These commands are Unicode-enabled. See Unicode support in control language in the IBM i
Information Center for more information.
Related concepts
File systems
A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.
Related tasks
Accessing using menus and displays
You can perform operations on files and other objects in the integrated file system by using a set of menus
and displays provided by your system.
Related information
Control language (CL)

Path name rules for CL commands and displays

When using a CL command or display to operate on an object, you identify the object by supplying its path
name.
The following is a summary of rules to keep in mind when specifying these path names. The term object in
these rules refers to any directory, file, link, or other object:
• Normally, all characters in the path name must be valid characters in the job’s coded character set
identifier (CCSID) or, if the job CCSID is 65535, the job default CCSID. Some commands are “Unicode-
enabled”, meaning you can specify the command and path name information in a Unicode CCSID
through interfaces such as the QCAPCMD API. See Unicode support in control language in the IBM i
information center for more information.
• Path names are specified in hierarchical order, beginning with the highest-level directory and ending
with the name of the object to be operated on by the command. A delimiter character separates each
component of the path name; for example:
Dir1/Dir2/Dir3/UsrFile
• All commands and displays allow the slash (/) character to be used as the delimiter. Most commands
and displays also allow the backslash (\) character to be used as the delimiter. However, some
commands and displays may treat the backslash (\) no differently than any other character. Therefore,
the backslash (\) character must be used with caution.
• An absolute path name begins with a delimiter. Path name resolution starts at the "root" (/) directory;
for example:
/Dir1/Dir2/Dir3/UsrFile
where Dir1 is a subdirectory of the “root” (/) directory.

72 IBM i: Integrated file system

• If a path name begins with a tilde (~) character followed by a delimiter, then path name resolution will
start at the home directory of the user entering the command; for example:
'~user-profile-name/UsrDir/UsrObj'
• A tilde (~) character appearing anywhere other than the very beginning of a specified path name has no
special significance.
• A relative path name does not begin with a delimiter or a tilde (~) character. Path name resolution starts
at the current directory; for example:
‘MyDir/MyFile’
where MyDir is a subdirectory of the current directory. When using relative path names you must be very
careful to ensure that the current directory is set correctly.
• Path names containing delimiters, tildes, or other special characters (see below) must be enclosed in
single quotation marks (') when entered on a command line. These quotation marks are optional when
path names are entered on displays.
• The slash (/) character and the null character can not be used in the individual components of a path
name. The backslash (\) character can not be used in a component for commands and displays that
treat the backslash (\) as a delimiter.
• Characters with code points below hexadecimal 40 can be specified on the commands, but only if the
entire path name is specified in hexadecimal format; for example:
CRTDIR DIR(X’02F0F1F3’)
Use of such characters is not recommended.
• Object names must be unique within each directory.
• The commands accept component names up to 255 characters long and path names up to 5000
characters long. However, some file systems only allow much smaller component and path names. They
may also limit the characters that are allowed in component names or require a particular component
name format. Also, the path name will not normally be converted to uppercase by the commands or
displays themselves, but the file system may do so before creating or searching for the specified object.
See the individual file system descriptions in the "File systems" topic of the IBM i Information Center for
more information about these considerations.
• Some commands allow you to specify a pattern that allows a single command to operate on multiple
objects that match the specified pattern. One or more asterisks (*) or question marks (?) can be used in
the last component of a path name. An asterisk (*) matches any number of characters. A question mark
(?) matches any single character. The following example searches for all objects in ‘/Dir1/Dir2/Dir3’
whose names begin with ‘d’ and end with ‘txt’. It could find objects named ‘dshorttxt’, ‘d1234.txt’, and
even ‘dtxt’:
'/Dir1/Dir2/Dir3/d*txt'
The following example searches for objects whose names begin with ‘d’ followed by any single
character and end with ‘txt’. It could find objects named ‘d1txt’ or ‘d.txt’, but not ‘d12txt’ or ‘dtxt’:
'/Dir1/Dir2/Dir3/d?txt'
More complex patterns can be specified with multiple asterisks, multiple question marks, or
combinations of the two. However, to avoid confusion with i5/OS special values, path names can not
start with a single asterisk (*) character. To perform a pattern match at the very beginning of a path
name (as in a relative path name), use two asterisks (*); for example:
'**.file'
• The path name must be enclosed in additional sets of single quotation marks (') or double quotation
marks (") if any of the following characters is used in an actual component name:
– Asterisk (*)
– Question mark (?)
– Single quotation mark (')

Integrated file system 73

 – Double quotation mark (")
– Tilde (~), if used as the first character in the first component name of the path name. If used in any
other position, the tilde is interpreted as just another character.
For example, if a component contains an asterisk, the name must be specified within additional
quotation marks so that the asterisk is not interpreted as part of a pattern:

'"/Dir1/Dir/A*Smith"'

or

'''/Dir1/Dir/A*Smith'''

If the component name itself contains single quotation marks or double quotation marks, you must
specify two such marks in a row to represent them; for example:

‘”/Dir1/Dir/Smith’’s Directory”’

or

‘”/Dir1/Dir/This is an “”example”””’

• Even though the various file systems, commands, and displays may allow certain special characters in
component names, you should avoid using those characters because they make it difficult or impossible
to work with objects using other interfaces. The slash (/), backslash (\), asterisk (*), question mark (?),
single quotation mark ('), double quotation mark ("), tilde (~), and colon (:) all have special significance
to certain interfaces. For example, the CL commands treat both the slash (/) and backslash (\) as
delimiters and can not properly interpret a path where component names contain either of those
characters. Similarly, the presence of backslash (\) or colon (:) characters may make some objects
unusable from PC client interfaces.
Related concepts
File systems
A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.
Related information
Control language (CL)

Working with output of the RTVDIRINF and PRTDIRINF commands

The Retrieve Directory Information (RTVDIRINF) command is used to collect attributes for
objects in the integrated file system. The collected information is stored in database files (tables) that are
named using the information file prefix specified by the INFFILEPFX parameter. The tables are created in
the library specified by the INFLIB parameter.
Three tables are created as a result of the RTVDIRINF command. One table stores object attributes, one
is for directories, and the last table is used to determine which files were used to store object attributes.
Starting in V6R1, you can also create these three tables with System i® Navigator. See “Collecting and
analyzing folder attributes with IBM Navigator for i” on page 90 for more information.
The following table describes the fields that are provided for the table that stores the object attributes.
If *GEN is specified in the information file prefix (INFFILEPFX) parameter, the database files are created
with a unique prefix generated by this command. The prefix begins with QAEZD followed by four digits.
The files created to store the collected information are named using this prefix followed by either
the letter D (for the file that contains directory information) or the letter O (for the file that contains
information about objects in directories). For example, the first time that the command is run with *GEN
specified, files QAEZD0001D and QAEZD0001O are created in the library specified by the Information
library (INFLIB) parameter. Users can specify a file prefix to use for naming this database, which can be
up to nine characters long.

74 IBM i: Integrated file system

Table 8. QAEZDxxxxO (store object's attributes)
Field name Field type Field description
QEZDIRIDX INTEGER Relational identifier between directory and object tables. It can be
used to join the tables to get complete path name information. The
QEZDIRIDX field value in the object table matches the QEZDIRIDX
value in the directories table for the object's parent directory.
Note: When the "root" directory ( /) is specified as input to
RTVDIRINF, the QEZDIRIDX value will be 1 even though the "root"
directory (/) does not have a parent directory.

QEZOBJNAM VARGRAPHIC The object name. 2

1 (1024)
QEZOBJLEN INTEGER Number of bytes contained in the object name (field QEZOBJNAM).
QEZNMCCSID INTEGER The CCSID in which the object name (field QEZOBJNAM) is
represented.
QEZREGION GRAPHIC (2) A two-character ID representing the country of the object name (field
QEZOBJNAM). This ID affects actions that tend to be defined by the
location of the action, such as collating sequence.
QEZLANGID GRAPHIC (3) A three character ID representing the language that the object name
(field QEZOBJNAM) is in.
QEZMODE INTEGER The file access mode and type. For more information about mode,
see Open file open() API.
QEZOBJTYPE GRAPHIC (10) The object type.
1

QEZCCSID INTEGER The CCSID of the data and extended attributes of the object.
QEZALCSIZE 1 BIGINT The number of bytes that have been allocated for this object.
QEZDTASIZE BIGINT The size in bytes of the data in this object. This size does not include
object headers or the size of extended attributes associated with the
object.
QEZEAS BIGINT Number of extended attributes associated with this object.
QEZCEAS BIGINT Number of critical extended attributes associated with this object.
QEZEXTATRS BIGINT Total number of bytes for all the extended attribute data.
QEZCRTTIM TIMESTAMP The date and time at which the object was created.
QEZACCTIM TIMESTAMP The date and time that the object's data was last accessed.
QEZCHGTIMA TIMESTAMP The date and time that the object's attributes were last changed.
1

QEZCHGTIMD TIMESTAMP The date and time that the object's data was last changed.
QEZSTGFREE SMALLINT Whether the object's data has been moved offline, freeing its online
1 storage. Valid values are:
0 - The object's data is not offline.
1 - The object's data is offline.

Integrated file system 75

 Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZCHKOUT 1 SMALLINT An indicator as to whether an object is checked out. Valid values are:
0 - The object is not checked out.
1 - The object is checked out.

QEZCHKOWN GRAPHIC (10) The user who has the object checked out. This field is blank if it is not
checked out.
QEZCHKTIM TIMESTAMP The date and time the object was checked out. If the object is not
checked out, this field will have NULL as its value.
QEZLOCAL SMALLINT Whether an object is stored locally or stored on a remote system. The
decision of whether an object is local or remote varies according to
the respective file system rules. Objects in file systems that do not
carry either a local or remote indicator are treated as remote. Valid
values are:
1 - The object's data is stored locally.
2 - The object's data is on a remote system.

QEZOWN 1 GRAPHIC (10) The name of the user profile that is the owner of the object or the
following special value:
*NOUSRPRF - This special value is used by the Network File System
to indicate that there is no user profile on the local iSeries server with
a user ID (UID) matching the UID of the remote object.

QEZUID INTEGER Each user on the system must have a unique numeric user
identification number (UID).
QEZOWNPGP GRAPHIC (10) The name of the user profile that is the primary group of the object or
the following special values:
*NONE - The object does not have a primary group
*NOUSRPRF - This special value is used by the Network File System
to indicate that there is no user profile on the local server with a
group ID (GID) matching the GID of the remote object.

QEZGID INTEGER Group profiles are identified by a unique numeric group identification
number (GID).
QEZAUTLST GRAPHIC (10) The name of the authorization list that is used to secure the named
object. The value *NONE indicates that no authorization list is used in
determining authority to the object.
QEZASP SMALLINT The auxiliary storage pool in which the object is stored.
QEZJRNSTS 1 SMALLINT Current journaling status of the object. This field will be one of the
following values:
0 (NOT_JOURNALED) - The object is currently not being journaled.
1 (JOURNALED) - The object is currently being journaled.

76 IBM i: Integrated file system

Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZJSUBTRE SMALLINT When this flag is returned, this object is a directory with integrated
file system journaling subtree semantics.
0 - The object is not journaled with subtree semantics.
1 - The object is journaled with subtree semantics. New objects
created within this directory's subtree will inherit the journaling
attributes and options from this directory.

QEZJOPTENT SMALLINT When journaling is active, entries that are considered optional are
journaled. The list of optional journal entries varies for each object
type.
0 - Object is not journaled with optional entries.
1 - Object is journaled with optional entries.

QEZJAFTERI SMALLINT When journaling is active, the image of the object after a change is
journaled.
0 - Object is not journaled with after images.
1 - Object is journaled with after images.

QEZJBEFORI SMALLINT When journaling is active, the image of the object before a change is
journaled.
0 - Object is not journaled with before images.
1 - Object is journaled with before images.

QEZJRNID GRAPHIC (10) This field associates the object being journaled with an identifier that
can be used on various journaling-related commands and APIs. This
field is blank if the object has never been journaled.
QEZJRNNAM GRAPHIC (10) If the value of the journaling status is JOURNALED, then this field
contains the name of the journal currently being used. If the value of
the journaling status is NOT_JOURNALED, then this field contains the
name of the journal last used for this object. All bytes in this field will
be set to binary zero if this object has never been journaled. This field
is blank if the object has never been journaled.
QEZJRNLIB GRAPHIC (10) If the value of the journaling status is JOURNALED, then this field
contains the name of the library containing the currently used
journal. If the value of the journaling status is NOT_JOURNALED, then
this field contains the name of the library containing the last used
journal. All bytes in this field will be set to binary zero if this object
has never been journaled. This field is blank if the object has never
been journaled.
QEZJRNSTR TIMESTAMP The number of seconds since the Epoch that corresponds to the last
date and time for which the object had journaling started for it. This
field will be set to binary zero if this object has never been journaled.
This field has NULL as its value if the object has never been journaled.

Integrated file system 77

 Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZAUDT GRAPHIC (10) The auditing value associated with the object. Valid values are:
*NONE - No auditing occurs for this object when it is read or changed
regardless of the user who is accessing the object.
*USRPRF - Audit this object only if the current user is being audited.
The current user is tested to determine if auditing should be done
for this object. The user profile can specify if only change access
is audited or if both read and change accesses are audited for this
object.
*CHANGE - Audit all change access to this object by all users on the
system.
*ALL - Audit all access to this object by all users on the system. All
access is defined as a read or change operation.
*NOTAVL - The user performing the operation is not allowed to
retrieve the current object auditing value.

QEZBLKSIZ INTEGER The block size of an object.

QEZNLNK INTEGER The number of hard links to the object.
QEZFILEID 1 GRAPHIC (16) The file ID for the object. An identifier associated with the object. A
file ID can be used with Qp0lGetPathFromFileID() to retrieve an
object's path name.
QEZFILEIDS INTEGER The 4 byte file ID of the file. This number uniquely identifies the
object within a file system. This number cannot identify the object in
the whole system.
QEZGENID BIGINT The generation ID associated with the file ID.
QEZFSID BIGINT The file system ID to which the object belongs. This number uniquely
identifies the file system to which the object belongs.
QEZRDEV BIGINT If the object represents a device special file, the real device it
represents.
QEZDOM GRAPHIC (10) The domain of the object. Valid values are:
*SYSTEM - Object exists in system domain.
*USER - Object exists in user domain.

78 IBM i: Integrated file system

Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZCRTAUD GRAPHIC (10) The auditing value associated with an object created in this directory.
Valid values are:
*NONE - No auditing occurs for this object when it is read or changed
regardless of the user who is accessing the object.
*USRPRF - Audit this object only if the current user is being audited.
The current user is tested to determine if auditing should be done
for this object. The user profile can specify if only change accesses
are audited or if both read and change accesses are audited for this
object.
*CHANGE - Audit all change accesses to this object by all users on
the system.
*ALL - Audit all accesses to this object by all users on the system. All
accesses are defined as a read or change operation.
*NOTAVL - The user performing the operation is not allowed to
retrieve the current create object auditing value.
*SYSVAL - The object auditing value for the objects created in the
directory is determined by the system auditing value (QCRTOBJAUD).

QEZSCN GRAPHIC (1) Whether the object will be scanned when exit programs are
registered with any of the integrated file system scan-related exit
points.
Valid values are:
x'00' (SCANNING_NO) - The object will not be scanned according to
the rules described in the scan-related exit programs.
Note: If the Scan file systems control (QSCANFSCTL) value
*NOPOSTRST is not specified when an object with this attribute is
restored, the object will be scanned at least once after the restore.
x'01' (SCANNING_YES) - The object will be scanned according to the
rules described in the scan-related exit programs if the object has
been modified or if the scanning software has been updated since the
last time the object was scanned.
x'02' (SCANNING_CHGONLY) - The object will be scanned according
to the rules described in the scan-related exit programs only if the
object has been modified since the last time the object was scanned.
It will not be scanned if the scanning software has been updated.
This attribute only takes effect if the Scan file systems control
(QSCANFSCTL) system value has *USEOCOATR specified. Otherwise,
it will be treated as if the attribute is SCANNING_YES.
Note: If the Scan file systems control (QSCANFSCTL) value
*NOPOSTRST is not specified when an object with this attribute is
restored, the object will be scanned at least once after the restore.

Integrated file system 79

 Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZINHSCN GRAPHIC (1) Whether the objects created in a directory will be scanned when
exit programs are registered with any of the integrated file system
scan-related exit points.
Valid values are:
x'00' - After an object is created in the directory, the object will not
be scanned according to the rules described in the scan-related exit
programs.
Note: If the Scan file systems control (QSCANFSCTL) value
*NOPOSTRST is not specified when an object with this attribute is
restored, the object will be scanned at least once after the restore.
x'01' - After an object is created in the directory, the object will be
scanned according to the rules described in the scan-related exit
programs if the object has been modified or if the scanning software
has been updated since the last time the object was scanned.
x'02' - After an object is created in the directory, the object will
be scanned according to the rules described in the scan-related
exit programs only if the object has been modified since the
last time the object was scanned. It will not be scanned if the
scanning software has been updated. This attribute only takes
effect if the Scan file systems control (QSCANFSCTL) system value
has *USEOCOATR specified. Otherwise, it will be treated as if the
attribute is SCANNING_YES.
Note: If the Scan file systems control (QSCANFSCTL) value
*NOPOSTRST is not specified when an object with this attribute is
restored, the object will be scanned at least once after the restore.

80 IBM i: Integrated file system

Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZSSTATUS GRAPHIC (1) The scan status associated with this object. This field will be one of
the following values:
x'00' (SCAN_REQUIRED) - A scan is required for the object either
because it has not yet been scanned by the scan-related exit
programs, or because the objects data or CCSID has been modified
since it was last scanned. Examples of object data or CCSID
modifications are: writing to the object, directly or through memory
mapping; truncating the object; clearing the object; and changing the
objects CCSID attribute.
x'01' (SCAN_SUCCESS) - The object has been scanned by a scan-
related exit program, and at the time of that last scan request, the
object did not fail the scan.
x'02' (SCAN_FAILURE) - The object has been scanned by a scan-
related exit program, and at the time of that last scan request, the
object failed the scan and the operation did not complete. Once
an object has been marked as a failure, it will not be scanned
again until the object's scan signature is different than the global
scan key signature or independent ASP group scan key signature as
appropriate. Therefore, subsequent requests to work with the object
will fail with a scan failure indication. Examples of requests which will
fail are opening the object, changing the CCSID of the object, copying
the object.
x'05' (SCAN_PENDING_CVN) - The object is not in a *TYPE2
directory, and therefore will not be scanned until the directory is
converted.
x'06' (SCAN_NOT_REQUIRED) - The object does not require any
scanning because the object is marked to not be scanned.

QEZSSIGDF GRAPHIC (1) The scan signatures give an indication of the level of the scanning
software support.
When an object is in an independent ASP group, the object scan
signature is compared to the associated independent ASP group scan
signature. When an object is not in an independent ASP group, the
object scan signature is compared to the global scan signature value.
This field will be one of the following values:
x'00' - The compared signatures are not different.
x'01' - The compared signatures are different.

QEZSBINARY GRAPHIC (1) This indicates if the object has been scanned in binary mode when it
was previously scanned. This field will be one of the following values:
x'00' - The object was not scanned in binary mode.
x'01' - The object was scanned in binary mode. If the object scan
status is SCAN_SUCCESS, then the object was successfully scanned
in binary. If the object scan status is SCAN_FAILURE, then the object
failed the scan in binary.

Integrated file system 81

 Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZSCCSID1 INTEGER This indicates if the object has been scanned in the listed CCSID
when it was previously scanned. If the object scan status is
SCAN_SUCCESS, then the object was successfully scanned in this
CCSID. If the object scan status is SCAN_FAILURE, then the object
failed the scan in this CCSID. A value of 0 means this field does not
apply.
QEZSCCSID2 INTEGER This indicates if the object has been scanned in the listed CCSID
when it was previously scanned. If the object scan status is
SCAN_SUCCESS, then the object was successfully scanned in this
CCSID. If the object scan status is SCAN_FAILURE, then this field will
be 0. A value of 0 means this field does not apply.
QEZUDATE TIMESTAMP The number of seconds since the Epoch that corresponds to the date
the object was last used. This field is zero when the object is created.
If usage data is not maintained for theIBM i type or the file system to
which an object belongs, this field is zero.
QEZUDCOUNT INTEGER The number of days an object has been used. Usage has different
meanings according to the specific file system and according to
the individual object types supported within a file system. Usage
can indicate the opening or closing of a file or can refer to adding
links, renaming, restoring, or checking out an object. This count is
increased each day that an object is used and is reset to zero by
calling the Qp0lSetAttr() API.
QEZURESET INTEGER The number of seconds since the Epoch that corresponds to the date
the days used count was last reset to zero (0). This date is set to the
current date when the Qp0lSetAttr() API is called to reset the Days
used count to zero.
QEZPRMLNK SMALLINT When an object has several names, this field will be set only for the
first name found.
QEZALWCKP SMALLINT Whether a stream file (*STMF) can be shared with readers and
W writers during the save-while-active checkpoint processing. Valid
values are:
0 - The object can be shared with readers only.
1 - The object can be shared with readers and writers.

QEZSIG 1 SMALLINT Whether an object has anIBM i digital signature. Valid values are:
0 - The object does not have anIBM i digital signature.
1 - The object does have anIBM i digital signature.

QEZSYSSIG SMALLINT Whether the object was signed by a source that is trusted by the
system. Valid values are:
0 - None of the signatures came from a source that is trusted by the
system.
1 - The object was signed by a source that is trusted by the system.
If the object has multiple signatures, at least one of the signatures
came from a source that is trusted by the system.

82 IBM i: Integrated file system

Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZMLTSIG SMALLINT Whether an object has more than oneIBM i digital signature. Valid
values are:
0 - The object has only one digital signature.
1 - The object has more than one digital signature. If the QEZSYSSIG
field has the value 1, at least one of the signatures is from a source
trusted by the system.

QEZDSTGOPT SMALLINT This option should be used to determine how auxiliary storage is
allocated by the system for the specified object. This option can
only be specified for stream files in the "root" (/), QOpenSys and
user-defined file systems. This option will be ignored for *TYPE1 byte
stream files. Valid values are:
0 - The auxiliary storage will be allocated normally. That is, as
additional auxiliary storage is required, it will be allocated in logically
sized extents to accommodate the current space requirement, and
anticipated future requirements, while minimizing the number of disk
I/O operations.
1 - The auxiliary storage will be allocated to minimize the space used
by the object. That is, as additional auxiliary storage is required, it
will be allocated in small sized extents to accommodate the current
space requirement. Accessing an object composed of many small
extents may increase the number of disk I/O operations for that
object.
2 - The system will dynamically determine the optimum auxiliary
storage allocation for the object, balancing space used versus disk
I/O operations. For example, if a file has many small extents, yet
is frequently being read and written, then future auxiliary storage
allocations will be larger extents to minimize the number of disk I/O
operations. Or, if a file is frequently truncated, then future auxiliary
storage allocations will be small extents to minimize the space used.
Additionally, information will be maintained on the stream file sizes
for this system and its activity. This file size information will also be
used to help determine the optimum auxiliary storage allocations for
this object as it relates to the other objects sizes.

Integrated file system 83

 Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZMSTGOPT SMALLINT This option should be used to determine how main storage is
allocated and used by the system for the specified object. This option
can only be specified for stream files in the "root" (/), QOpenSys and
user-defined file systems. Valid values are:
0 - The main storage will be allocated normally. That is, as much
main storage as possible will be allocated and used. This minimizes
the number of disk I/O operations since the information is cached in
main storage.
1 - The main storage will be allocated to minimize the space used by
the object. That is, as little main storage as possible will be allocated
and used. This minimizes main storage usage while increasing the
number of disk I/O operations since less information is cached in
main storage.
2 - The system will dynamically determine the optimum main storage
allocation for the object depending on other system activity and
main storage contention. That is, when there is little main storage
contention, as much storage as possible will be allocated and used
to minimize the number of disk I/O operations. And when there
is significant main storage contention, less main storage will be
allocated and used to minimize the main storage contention. This
option only has an effect when the storage pool's paging option is
*CALC. When the storage pool's paging option is *FIXED, the behavior
is the same as STG_NORMAL. When the object is accessed through a
file server, this option has no effect. Instead, its behavior is the same
as STG_NORMAL.

QEZDIRTYP2 SMALLINT The format of the specified directory object. Valid values are:
0 - The directory format is *TYPE1.
1 - The directory format is *TYPE2.

QEZFILTYP2 1 SMALLINT The format of the stream file (*STMF). Valid values are:
0 - The stream file format is *TYPE1.
1 - The stream file format is *TYPE2.

QEZUDFTYP2 SMALLINT The default file format of stream files (*STMF) created in the user-
defined file system. Valid values are:
0 - The stream file format is *TYPE1.
1 - The stream file format is *TYPE2.

QEZNONSAV SMALLINT Whether the object can be saved or not. Valid values are:
0 - Object will be saved.
1 - Object will not be saved. Additionally, if this object is a
directory, none of the objects in the directory's subtree will be
saved unless they were explicitly specified as an object to be saved.
The subtree includes all subdirectories and the objects within those
subdirectories.

84 IBM i: Integrated file system

Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZCLSTRSP SMALLINT The object is the storage which was allocated for Integrated xSeries
servers to use as virtual disk drives for the xSeries servers. From the
perspective of the iSeries server, virtual drives appear as byte stream
files within the integrated file system.
0 - Object is not virtual disk storage.
1 - Object is virtual disk storage.

QEZCASE SMALLINT Indicates the Case Sensitivity of the file system that contains this
object.
0 - File system is not case sensitive.
1 - File system is case sensitive.

QEZOFLOW SMALLINT Indicates if the object has overflowed the auxiliary storage pool it
resides in. Valid values are:
0 - Auxiliary storage pool is not overflowed.
1 - Auxiliary storage pool is overflowed.

QEZPCREAD SMALLINT Whether the object can be written to or deleted, have its extended
attributes changed or deleted, or have its size changed. Valid values
are:
0 - The object can be changed.
1 - The object cannot be changed.

QEZPCHID 1 SMALLINT Whether the object can be displayed using an ordinary directory
listing.
0 - The object is not hidden.
1 - The object is hidden.

QEZPCSYS SMALLINT Whether the object is a system file and is excluded from normal
directory searches.
0 - The object is not a system file.
1 - The object is a system file.

QEZPCARC SMALLINT Whether the object has changed since the last time the object was
examined.
0 - The object has not changed.
1 - The object has changed.

QEZSYSARC SMALLINT Whether the object has changed and needs to be saved. It is set on
when an object's change time is updated, and set off when the object
has been saved.
0 - The object has not changed and does not need to be saved.
1 - The object has changed and does need to be saved.

Integrated file system 85

 Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZJRCVNAM GRAPHIC(10) The oldest journal receiver needed to successfully apply
journal changes. When the Apply information field is set to
PARTIAL_TRANSACTION the journal receiver contains the start of the
partial transaction. Otherwise; the journal receiver contains the start
of the save operation. This field is blank if the object has never been
journaled.
QEZJRCVLIB GRAPHIC(10) The name of the library that contains the journal receiver needed to
successfully apply journal changes. This field is blank if the object
has never been journaled.
QEZJRCVASP GRAPHIC(10) The name of the ASP that contains the journal receiver needed to
successfully apply journal changes. The valid values are:
*SYSBAS - The journal receiver resides in the system or user ASP.
ASP device - The name of the ASP device that contains the journal
receiver.

QEZJTRNI GRAPHIC(1) This field describes information about the current state of the object
as it relates to commitment control boundaries. The valid values are:
x'00' (NONE) - There are no partial transactions.
x'01' (PARTIAL_TRANSACTION) - The object was restored with
partial transactions. This object cannot be used until the Apply
Journaled Changes (APYJRNCHG) or Remove Journaled
Changes (RMVJRNCHG) command is used to complete or rollback
the partial transactions.
x'02' (ROLLBACK_ENDED) - The object had a rollback operation
ended using the "End Rollback" option on the Work with Commitment
Definition (WRKCMTDFN) screen. It is recommended that the object
be restored as it cannot be used. As a last option, the Change
Journaled Object (CHGJRNOBJ) command can be used to allow
the object to be used. Doing this, however, can leave the object in an
inconsistent state.

QEZTMPOBJ SMALLINT Whether the object is a temporary object. Possible values are:
• 0 - The object is a permanent object.
• 1 - The object is a temporary object.

QEZTMPUDFS SMALLINT Whether the objects in the user-defined file system (UDFS) are
temporary objects. Possible values are:
• 0 - The objects in the UDFS are permanent objects.
• 1 - The objects in the UDFS are temporary objects.

QEZUNIT GRAPHIC(10) The preferred storage media for the objects in the UDFS. Possible
values are:
• *SSD - Storage should be allocated from solid state drive storage
media, if available.
• *ANY - Storage will be allocated from any available storage media.
• *NOTAVL - The storage media preference can not be determined.

86 IBM i: Integrated file system

Table 8. QAEZDxxxxO (store object's attributes) (continued)
Field name Field type Field description
QEZINALCPW SMALLINT Whether new stream file objects created or linked within a directory
should inherit the save-while-active checkpoint processing options of
its parent directory. Possible values are:
• 0 - New stream file objects created within this directory will not
inherit the save-while-active checkpoint processing options of the
parent directory.
• 1 - New stream file objects created within this directory will inherit
the save-while-active checkpoint processing options of the parent
directory.

QEZSYSRSSV SMALLINT Whether the system prevents the object from being saved. Possible
values are:
• 0 - The system does not prevent the object from being saved.
• 1 - The system prevents the object from being saved.

Notes:
1. This field is included in the subset of fields used by the PRTDIRINF command.
2. In this field, only the object name is stored. The rest of the path name is stored in field QEZDIRNAM1
if the length of the directory name is below 1 KB or in QEZDIRNAM2 if the directory names are above
1 KB.

The following table is an example of a table that lists the directories processed by the RTVDIRINF
command.

Table 9. QAEZDxxxD (store directory's attributes)

Field name Field type Field description
QEZDIRIDX INTEGER Relational identifier between directory and object tables. It can be
used to join the tables to get complete path name information. The
QEZDIRIDX field value in the object table matches the QEZDIRIDX
value in the directories table for the object's parent directory.
QEZDIRNAM VARGRAPHIC The parent directory path. Only used when path length is below 1
11 (1024) KB (1024 bytes).
QEZDIRNAM DBCLOB (16M) The parent directory path. Only used when path length is above 1 KB
21 (1024 bytes) long. Can store paths up to 16 MB long.
QEZDRCCSID INTEGER The directory CCSID.
QEZDREGIO GRAPHIC (2) The directory path region ID.
N
QEZLANGID GRAPHIC (3) The directory path language ID.
QEZDIRLEN 1 INTEGER Length of directory's path name.
QEZDIRFID GRAPHIC (16) The file ID of the directory. An identifier associated with the object.
A file ID can be used with Qp0lGetPathFromFileID() to retrieve
an object's path name.
QEZDFID INTEGER The file ID of the directory.
QEZDIRFSID BIGINT The file system ID of the directory.

Integrated file system 87

 Table 9. QAEZDxxxD (store directory's attributes) (continued)
Field name Field type Field description
QEZDIRGID BIGINT The generation ID.
QEZPARDIR INTEGER The parent directory index.

Note:
1. This field is included in the subset of fields used by the PRTDIRINF command.

The following table shows the information that the RTVDIRINF command stores regarding the files it has
created when it runs. If the file that contains this information does not exist, the RTVDIRINF command
creates it; when the command runs on subsequent occasions, the information is appended to the existing
file. The PRTDIRINF command uses this information to determine which database files were used to
store information that was retrieved by different instances of the RTVDIRINF command.

Table 10. QUSRSYS/QAEZDBFILE (store files created)

Field name Field type Field description
QEZDIRSRC VARGRAPHIC Path specified in DIR parameter (RTVDIRINF).
(5000)
QEZPRCCSID INTEGER The path CCSID.
QEZPREGIO GRAPHIC (2) The path region ID.
N
QEZPLANGI GRAPHIC (3) The path language ID
D
QEZOBJFILE VARGRAPHIC (20) The name of the file generated to store the object's attributes.
1

QEZDIRFILE VARGRAPHIC (20) The name of the file generated to store the directory's indexes.
1

QEZLIB 1 VARGRAPHIC (20) Library where both generated files reside.

QEZSTRTIME TIMESTAMP Date/Time when RTVDIRINF was submitted.
QEZENDTIM TIMESTAMP Date/Time when RTVDIRINF was completed.
E

Note:
1. This field is included in the subset of fields used by the PRTDIRINF command.

Related tasks
Collecting and analyzing folder attributes with IBM Navigator for i
You can collect and analyze attributes for objects in the integrated file system with IBM Navigator for
i. This easy-to-use interface provides the same function as the Retrieve Directory Information
(RTVDIRINF) command does. You can examine and query data collected by this interface as well as that
collected by the RTVDIRINF command.
Related information
Retrieve Directory Information (RTVDIRINF) command
Qp0lGetPathFromFileID()--Get Path Name of Object from Its File ID API
Qp0lSetAttr()--Set Attributes API
Apply Journaled Changed (APYJRNCHG) command
Remove Journaled Changes (RMVJRNCHG) command

88 IBM i: Integrated file system

Change Journaled Object (CHGJRNOBJ) command
Print Directory Information (PRTDIRINF) command

Accessing the data of RTVDIRINF

There exist several options for accessing the data in the tables.
Listed as follows are ways that you can access the data created by the Retrieve Directory
Information (RTVDIRINF) command:
• Using the Print Directory Information (PRTDIRINF) command
This command is used to print directory information about objects and directory information in the
integrated file system. The information it will print is already stored in the database file specified by the
user in the RTVDIRINF command.
• Using any program or command provided by IBM that can run queries over a DB2® table on the IBM i
operating system.
Some of the more common tools are the Start SQL Interactive Session (STRSQL) command
and the IBM Navigator for i.
For example, if you want to select objects in a specific path (previously collected by the RTVDIRINF
command) that have an allocation size larger than 10 KB, you can run a query like this:

SELECT QEZOBJNAM, QEZALCSIZE FROM library_name/QAEZDxxxxO WHERE

QEZALCSIZE > 10240

• You can make your own programs and access the database tables by using any valid DB methods.
• Instead of running various queries by issuing commands, you can easily use the IBM Navigator for i to
retrieve, display, and analyze the directory information data (which is known as folder attribute data in
IBM Navigator for i). See Collecting and analyzing folder attributes with IBM Navigator for i for more
information.
Related tasks
Collecting and analyzing folder attributes with IBM Navigator for i
You can collect and analyze attributes for objects in the integrated file system with IBM Navigator for
i. This easy-to-use interface provides the same function as the Retrieve Directory Information
(RTVDIRINF) command does. You can examine and query data collected by this interface as well as that
collected by the RTVDIRINF command.
Related information
Print Directory Information (PRTDIRINF) command
Start SQL Interactive Session (STRSQL) command
Embedded SQL programming
SQL call level interface

Using the data of RTVDIRINF

Here are examples that display why the data is important or how you can use the data that is produced
from each of the three tables.
• For Table 8 on page 75, you can make queries to create reports or statistics based on any of the fields
within this table. PRTDIRINF does not include reports based on all of the fields. Instead, a subset will
be used.
• The data from Table 9 on page 87 contains all the directories within the path specified in DIR parameter
of the RTVDIRINF command. If you want to know specific attributes about the path name, for example
the CCSID, language ID, or length, then this data is useful. Also, each directory stored in this table
has a unique value, or index that identifies it. In Table 8 on page 75, you can find the same field,
QEZDIRIDX, that will tell you which objects belong to which directory. To find which objects belong to

Integrated file system 89

 which directory, you can make a query using joins. For example, the following query statement selects
the names of all objects existing in directory "/MYDIR":

SELECT QAEZDxxxxO.QEZOBJNAM FROM library_name/QAEZDxxxxD,

library_name/QAEZDxxxxO WHERE QAEZDxxxxD.QEZDIRNAM1 = '/MYDIR' AND
QAEZDxxxxD.QEZDIRIDX = QAEZDxxxxO.QEZDIRIDX

• Table 10 on page 88 is mostly used by the PRTDIRINF command to obtain specific data about
RTVDIRINF runs. Examples of this are: The names of the tables created, the library where the tables
reside, and the starting and ending time of processing. You might use this table to know when a
RTVDIRINF was issued or what tables must be searched in order to query them.

Collecting and analyzing folder attributes with IBM Navigator for i

You can collect and analyze attributes for objects in the integrated file system with IBM Navigator for
i. This easy-to-use interface provides the same function as the Retrieve Directory Information
(RTVDIRINF) command does. You can examine and query data collected by this interface as well as that
collected by the RTVDIRINF command.

About this task

Note: This method is available in IBM Navigator for i V6R1, or later.
To generate a report for the attributes of certain objects in the integrated file system, follow these steps:

Procedure
1. In IBM Navigator for i, expand File Systems > Integrated File System.
2. Right-click the folder containing the objects of interest, and select Folder Attribute Information >
Collect Attributes.
3. In the Collect Attributes window, specify your preferences. Select Include the contents of subfolders
contained in this folder if you want to collect the attributes of subfolders as well. Optionally, you can
specify the File prefix and Library. Click OK to start collecting object attributes.
This data collection process might take a while. You need to wait a few seconds before the Display
Collected Attributes windows is displayed.
4. In the Display Collected Attributes window, clicking the arrowhead next to the entry you want to
analyze and select Analyze information.
Note: If you have performed the Collect Attributes operation from IBM Navigator for i or run the
Retrieve Directory Information (RTVDIRINF) command before, you can enter Display
Collected Attributes directly by right-clicking on the folder and selecting Folder Attribute Information
> Display Collect Attributes.
5. In the Analyze Folder Information window, customize the attributes you want to view on the Columns,
Filter, and Order tabs. Then click OK to generate the Folder Attribute Information Report.
Here is an example. Suppose you want to display the files that are larger than 10 MB in size and their
owners, with the results sorted first by size and then by owner.
• On the Columns tab, select Owner and click Add. Select Parent folder path and click Add. Select
Object name and click Add. Select Allocated size and click Add.
• On the Filter tab, select Allocated size in Field, select Sizes greater than in Condition, type 10 in
Size, and select Megabytes in Unit. Click Add to create the filter.
• On the Order tab, select Allocated size and Descending for First sort, and select Owner and
Descending for Second sort.
• Click OK. Then the Folder Attribute Information Report displays the tailored information.
6. In the Folder Attribute Information Report Window, you could select the delete action to remove
Integrated File System objects listed in the report without leaving the report window.

90 IBM i: Integrated file system

 Related reference
Working with output of the RTVDIRINF and PRTDIRINF commands
The Retrieve Directory Information (RTVDIRINF) command is used to collect attributes for
objects in the integrated file system. The collected information is stored in database files (tables) that are
named using the information file prefix specified by the INFFILEPFX parameter. The tables are created in
the library specified by the INFLIB parameter.
Accessing the data of RTVDIRINF
There exist several options for accessing the data in the tables.

Accessing using APIs

You can use application programming interfaces (APIs) to access the integrated file system.
Related reference
Performing operations using APIs
Many of the application programming interfaces (APIs) that perform operations on integrated file system
objects are in the form of C language functions.

Accessing using IBM i (SQL) services

You can use SQL procedures, table functions, and views to access the integrated file system.
Related information
IFS Services

Accessing using a PC
If your PC is connected to an IBM iproduct, you can interact with the directories and objects of the
integrated file system as if they were stored on your PC.
You can copy objects between directories by using the drag-and-drop capability of Windows Explorer. As
needed, you can actually copy an object from your system to the PC by selecting the object in the system
drive and dragging the object to the PC drive.
Any objects that are copied between an IBM i product and PCs by using the Windows interface can
be automatically converted between EBCDIC (extended binary-coded decimal interchange code) and
ASCII (American National Standard Code for Information Interchange). The IBM i Access Family can
be configured to automatically perform this conversion, and can even specify that the conversion be
performed on files with a specific extension.
Depending on the type of object, you can use PC interfaces and PC applications to work with it. For
example, a stream file containing text can be edited using a PC editor.
If you are connected to an IBM iproduct using your PC, the integrated file system makes your system's
directories and objects available to the PC. PCs can work with files in the integrated file system by using
file sharing clients built into the Windows operating systems, an FTP client, or IBM Navigator for i. Your PC
uses Windows file sharing clients to access IBM i NetServer, which runs on your system.
Related concepts
Accessing using IBM Navigator for i
IBM Navigator for i is a web interface for managing and administering your systems. IBM Navigator for i
makes operation and administration of your system easier and more productive.
Related tasks
Accessing using IBM i NetServer
IBM i Support for Windows Network Neighborhood (IBM i NetServer) is a function that enables Windows
clients to access IBM i shared directory paths and shared output queues. IBM i NetServer allows PCs that
run Windows software to seamlessly access data and printers that are managed by your IBM i platform.
Related reference
Accessing using File Transfer Protocol

Integrated file system 91

 The File Transfer Protocol (FTP) client allows you to transfer files that are found on your IBM i platform.

Accessing using IBM Navigator for i

IBM Navigator for i is a web interface for managing and administering your systems. IBM Navigator for i
makes operation and administration of your system easier and more productive.
For instance, you can copy a file to another folder by right-clicking and selecting Copy. Wizards guide you
through setting up security and TCP/IP services and applications.
You can perform many tasks using IBM Navigator for i. Following are some common integrated file system
tasks to help you get started:

Working with files and folders

• “Creating a folder” on page 131
• “Removing a file or folder” on page 132
• “Setting permissions” on page 134
• “Setting up file text conversion” on page 135
• “Sending a file or folder to another system” on page 135
• “Changing options for sending a file or folder” on page 136
• “Setting whether objects should be scanned or not” on page 141
• “Checking in objects” on page 142
• “Checking out objects” on page 143
• “Collecting and analyzing folder attributes with IBM Navigator for i” on page 90

Working with file shares

• “Creating a file share” on page 136
• “Changing a file share” on page 137

Working with user-defined file systems

• “Creating a new user-defined file system” on page 138
• “Mounting a user-defined file system” on page 138
• “Unmounting a user-defined file system” on page 139
• “Working with dynamically mounted file systems” on page 139

Journaling objects
• “Starting journaling” on page 105
• “Ending journaling” on page 106

Accessing using IBM i NetServer

IBM i Support for Windows Network Neighborhood (IBM i NetServer) is a function that enables Windows
clients to access IBM i shared directory paths and shared output queues. IBM i NetServer allows PCs that
run Windows software to seamlessly access data and printers that are managed by your IBM i platform.

About this task

PC clients on a network use the file and print sharing functions that are included in their operating
systems. This means that you do not need to install any additional software on your PC to use IBM i
NetServer.

92 IBM i: Integrated file system

 Linux clients with the Samba client software installed can also seamlessly access data and printers
through IBM i NetServer. Shared IBM i NetServer directories can be mounted on Linux clients as Samba
file systems in a similar manner to mounting NFS file systems that have been exported from IBM i.
An IBM i NetServer file share is a directory path that IBM i NetServer shares with clients on the IBM
i network. A file share can consist of any integrated file system directory on the system. Before you
can work with file sharing using IBM i NetServer, you must create an IBM i NetServer file share, and, if
necessary, change an IBM i NetServer file share using IBM Navigator for i.
To access integrated file system file shares using IBM i NetServer:

Procedure
1. Right-click Start, and select Explore to open Windows Explorer on your Windows PC.
2. Open the Tools menu, and select Map network drive.
3. Select a letter of a free drive for the file share (such as the I:\ drive).
4. Enter the name of anIBM i NetServer file share.
For example, you can enter the following syntax: \\QSYSTEM1\Sharename
Note: QSYSTEM1 is the system name of IBM i NetServer, and Sharename is the name of the file share
you want to use.
5. Click OK.

Results
Note: When connecting using IBM i NetServer, the system name might be different from the name
used by the IBM i Access Family. For example the IBM i NetServer name might be QAS400X,
and the path to work with files might be \\QAS400X\QDLS\MYFOLDER.FLR\MYFILE.DOC. However,
the IBM i Access Family name might be AS400X, and the path to work with files might be \
\AS400X\QDLS\MYFOLDER.FLR\MYFILE.DOC.
You choose which directories to share with the network using IBM i NetServer. Those directories appear
as the first level under the system name. For example, if you share the /home/fred directory with the
name fredsdir, a user can access that directory from the PC with the name \\QAS400X\FREDSDIR, or
from a Linux client with the name //qas400x/fredsdir.
The "root" (/) file system provides much better performance for PC file serving than other IBM i file
systems. You might want to move files to the "root" (/) file system. See “Moving files or folders to another
file system” on page 133 for more information.
Related information
i5/OS NetServer
i5/OS NetServer file shares

Accessing using File Transfer Protocol

The File Transfer Protocol (FTP) client allows you to transfer files that are found on your IBM i platform.
You can also transfer folders and documents in the document library services (QDLS) file system. The FTP
client can be run interactively in an unattended batch mode where client subcommands are read from
a file and the responses to these subcommands are written to a file. It also includes other features for
manipulating files on your system.
You can use FTP support to transfer files to and from any of the following file systems:
• "root" (/) file system
• Open systems file system (QOpenSys)
• Library file system (QSYS.LIB)
• Independent ASP QSYS.LIB file system

Integrated file system 93

 • Document library services file system (QDLS)
• Optical file system (QOPT)
• Network File System (NFS)
• IBM i NetClient file system (QNTC)
• QFileSvr.400 file system
However, be aware of the following restrictions:
• The integrated file system limits FTP support to transferring file data only. You cannot use FTP to
transfer attribute data.
• QSYS.LIB and independent ASP QSYS.LIB file systems limit FTP support to physical file members,
source physical file members, and save files. You cannot use FTP to transfer other object types, such
as programs (*PGM). However, you can save other object types to a save file, transfer the save file, and
then restore the objects.
Related information
File Transfer Protocol
Transferring files with File Transfer Protocol

Integrated file system conversion

The i5/OS operating system runs automatic conversions for file systems in the integrated file system to
support new directory format or Unicode standard.

Converting directories from *TYPE1 to *TYPE2

The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format.
The *TYPE2 directory format is an enhancement of the original *TYPE1 directory format. *TYPE2
directories have a different internal structure from *TYPE1 directories and provide improved performance
and reliability.
Shortly after IBM i V5R3 or a later release is installed, the conversion to *TYPE2 directories automatically
begins for any of the file systems that have not yet been converted to support *TYPE2 directories. This
conversion should not significantly impact your system activity.

Overview of *TYPE1 to *TYPE2 conversion

The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format.
The *TYPE2 directory format is an enhancement of the original *TYPE1 directory format. *TYPE2
directories have a different internal structure from *TYPE1 directories and provide improved performance
and reliability. In addition to improved performance and reliability, new functions such as integrated file
system scanning support are only available for objects in *TYPE2 directories. See “Scanning support” on
page 17 for more information.
Shortly after IBM i V5R3M0 operating system or a later release is installed, the conversion to *TYPE2
directories will automatically begin for any of the file systems which have not yet been converted to
support *TYPE2 directories. This conversion should not significantly impact your system activity as it will
run in a low priority background job.
If the conversion function has not yet completed, and the system has a normal or abnormal IPL, the
conversion function will resume when the IPL is completed. The conversion will restart on every initial
program load (IPL) until all eligible file systems have been fully converted.
The file systems that are eligible for this automatic conversion are "root" (/), QOpenSys and the user-
defined file systems for ASPs 1 through 32.

94 IBM i: Integrated file system

Note: You can avoid the automatic conversion to *TYPE2 directories if you convert the file systems before
installing V5R3 operating system or a later release.
Related concepts
*TYPE2 directories
The "root" (/), QOpenSys, and user-defined file systems (UDFS) in the integrated file system support the
*TYPE2 directory format. The *TYPE2 directory format is an enhancement of the original *TYPE1 directory
format.
Related reference
Conversion status determination
Shortly after IBM i V5R3M0 operating system or a later release is installed, the conversion to *TYPE2
directories will automatically begin for any of the file systems which have not yet been converted to
support *TYPE2 directories. This conversion processing will take place in a secondary thread of the
QFILESYS1 system job.
Tips: Independent ASP
If the user-defined file systems in an independent ASP have not yet been converted to the *TYPE2
directory format, they will be converted the first time the independent ASP is varied on to a system
installed with V5R2 or later of the operating system.

Directory conversion considerations

Here are several things to consider during the directory conversion process.

Conversion status determination

Shortly after IBM i V5R3M0 operating system or a later release is installed, the conversion to *TYPE2
directories will automatically begin for any of the file systems which have not yet been converted to
support *TYPE2 directories. This conversion processing will take place in a secondary thread of the
QFILESYS1 system job.
To determine the status of the conversion processing, you can use the Convert Directory (CVTDIR)
command as follows:

CVTDIR OPTION(*CHECK)

This invocation of the CVTDIR command lists the current directory format for the "root" (/), QOpenSys
and UDFS file systems and if the file system is currently being converted. Additionally, it lists the current
priority of the conversion function, the file system currently being converted by the system, the number
of links that have been processed for that file system, and the percentage of directories which have been
processed for that file system. The system starts the conversion function with a very low priority (99) so
that the conversion function does not significantly impact system activity. However, you can change the
priority of the conversion function by using the *CHGPTY value for the OPTION parameter of the CVTDIR
command. See CVTDIR for additional information about this parameter specification.
Because the QFILESYS1 job is processing the conversion, you can display the QFILESYS1 job log for
messages indicating any problems with the conversion. Additionally, various progress messages are
sent about the file system conversions. These messages include information such as: the file system
being converted, the number of links which have been processed in the file system, the percentage of
directories that have been processed in the file system, and so on. All of the error and many of the
progress messages are also sent to the QSYSOPR message queue. Therefore, for future reference, it is
good practice to ensure that the QHST logs or QFILESYS1 job logs are preserved, which contain these
messages. After the file systems have been fully converted, and the integrated file system is working as
expected, you can delete this historical information.
Related information
Convert Directory (CVTDIR) command

Integrated file system 95

 User profiles creation
The conversion function creates a user profile that is used while the conversion function is running. This
user profile has the name QP0FCWA. It is used by the conversion function to own converted directories in
the file system if the original owner is unable to own their directories.
The user profile is deleted when the conversion has completed, if possible. Message CPIA08B is sent to
the QFILESYS1 job log and the QSYSOPR message queue if ownership of a directory is given to this user
profile.
Related reference
Changing the owner of a directory
If the user profile that owns the *TYPE1 directory is unable to own the *TYPE2 directory that is created,
the owner of the *TYPE2 directory is set to the alternate user profile.

Objects renamed
*TYPE2 directories require the link names to be valid UTF-16 names.
The naming rule of *TYPE2 directories differs from *TYPE1 directories, which have UCS2 Level 1 names.
For this reason, invalid or duplicate names might be found during a directory conversion. When a name
is found to be invalid or a duplicate, the name is changed to a unique, valid, UTF-16 name, and message
CPIA08A is sent to the QFILESYS1 job log and the QSYSOPR message queue listing the original name and
the new name. Combined characters or invalid surrogate character pairs contained in a name might cause
an object to be renamed.

For more information about UTF-16, refer to the Unicode home page (www.unicode.org ).

Combined characters
Some characters can be made up of more than one Unicode character.
For example, characters that have an accent (for example, é or à) or an umlaut (for example, ä or ö) need
to be changed, or normalized, to a common format before they are stored in the directory, so that all
objects have a unique name. Normalizing a combined character is a process by which the character is put
in a known and predictable format. The format chosen for *TYPE2 directories is the canonical composed
form. If there are two objects in a *TYPE1 directory that contain the same combined characters, they are
normalized to the same name. This causes a collision, even if one object contains composed combined
characters and the other object contains decomposed combined characters. Therefore, one of them has
its name changed before it is linked in the *TYPE2 directory.

Surrogate characters
Some characters do not have a valid representation in Unicode.
These characters have some special values; they are made up of two Unicode characters in two specific
ranges such that the first Unicode character is in one range (for example 0xD800-0xD8FF) and the second
Unicode character is in the second range (for example 0xDC00-0xDCFF). This is called a surrogate pair.
If one of the Unicode characters are missing or if they are out of order, (only a partial character), it is an
invalid name. Names of this type have been allowed in *TYPE1 directories, but are not allowed in *TYPE2
directories. In order for the conversion function to continue, the name is changed before the object is
linked into the *TYPE2 directory if a name containing one of these invalid names is found.

User profile considerations

While the conversion is running, every attempt is made to ensure that the same user profile that owns any
*TYPE1 directories continues to own the corresponding *TYPE2 directories.
Since the *TYPE1 and *TYPE2 directories momentarily exist at the same time, this impacts the amount of
storage owned by the user profile and the number of entries in the user profile.

96 IBM i: Integrated file system

Changing maximum storage for a user profile
During the directory conversion processing, a number of directories that momentarily exist in both
formats at the same time are owned by the same user profile.
If during conversion processing the maximum storage limit for the user profile is reached, the limit for the
user profile is increased. Message CPIA08C is sent to the QFILESYS1 joblog and the QSYSOPR message
queue.

Changing the owner of a directory

If the user profile that owns the *TYPE1 directory is unable to own the *TYPE2 directory that is created,
the owner of the *TYPE2 directory is set to the alternate user profile.
Message CPIA08B is sent to the QFILESYS1 job log and the QSYSOPR message queue, and conversion
continues.
If the user profile that owns the *TYPE1 directory is unable to own the *TYPE2 directory that is created,
the owner of the *TYPE2 directory is set to the alternate user profile. Message CPIA08B is sent to the
QFILESYS1 job log and the QSYSOPR message queue, and conversion continues.
Related reference
User profiles creation
The conversion function creates a user profile that is used while the conversion function is running. This
user profile has the name QP0FCWA. It is used by the conversion function to own converted directories in
the file system if the original owner is unable to own their directories.

Auxiliary storage requirements

Auxiliary storage requirements should be considered while the system is converting the directories in a
file system to the *TYPE2 format.
Here are several considerations regarding auxiliary storage requirements:
• The final size of the directories after they have been converted to the *TYPE2 format
• Additional storage required while the conversion function is running
In many cases, the final size of a *TYPE2 directory is smaller than a *TYPE1 directory. Typically, *TYPE2
directories that have less than 350 objects require less auxiliary storage than *TYPE1 directories with
the same number of objects. *TYPE2 directories with more than 350 objects are ten percent larger (on
average) than *TYPE1 directories.
While the conversion function is running, additional storage is required. The conversion function requires
that the directories have both a *TYPE1 version and a *TYPE2 version in existence simultaneously.
Note: Before installing the IBM i V5R3M0 operating system or a later release, you might want to consider
running the *ESTIMATE option on the OS/400 V5R2 (CVTDIR) command, because it can provide a
conservative estimate of the amount of auxiliary storage needed during conversion.
Related information
Convert Directory (CVTDIR) command

Tips: Symbolic link

Symbolic links are objects within the integrated file system that contain a path to another object.
There are some instances during conversion when the name of an object can be changed. If one of
the elements of the path within a symbolic link is renamed during conversion, then the contents of the
symbolic link no longer point to the object.
Related concepts
Link

Integrated file system 97

 A link is a named connection between a directory and an object. A user or a program can tell the system
where to find an object by specifying the name of a link to the object. A link can be used as a path name or
as part of a path name.
Related reference
Objects renamed
*TYPE2 directories require the link names to be valid UTF-16 names.
Related information
symlink()--Make Symbolic Link

Tips: Independent ASP

If the user-defined file systems in an independent ASP have not yet been converted to the *TYPE2
directory format, they will be converted the first time the independent ASP is varied on to a system
installed with V5R2 or later of the operating system.

Tips: Saving and restoring

Directories that exist as *TYPE1 can be saved and restored in a file system that has been converted to
*TYPE2.
Likewise, directories that exist as *TYPE2 can be saved and restored in a file system that is *TYPE1
format, provided none of the *TYPE1 limits have been exceeded when the directory existed as a *TYPE2
directory.

Tips: Reclaiming integrated file system objects

While the system is converting the "root" (/), QOpenSys and user ASP UDFS file systems to support the
*TYPE2 directory format, the Reclaim Storage (RCLSTG) and Reclaim Object Links (RCLLNK)
commands cannot be run on any integrated file system directories including those in independent ASPs.
The OMIT(*DIR) parameter value can be used on the RCLSTG command to omit integrated file system
directories and allow the objects that are not related to integrated file system to be reclaimed.
Related concepts
Reclaim operation of the "root" (/), QOpenSys, and user-defined file systems
Reclaiming the "root" (/), QOpenSys, and user-defined file systems can be accomplished using the
Reclaim Object Links (RCLLNK) and Reclaim Storage (RCLSTG) commands.
Related information
Reclaim Storage (RCLSTG) command
Reclaim Object Links (RCLLNK) command

Integrated file system scanning

Objects in the "root" (/), QOpenSys and user ASP UDFS file systems will not be scanned using the
integrated file system scan-related exit points until the file systems have fully converted to the *TYPE2
directory format.
The scan-related attributes can be set for objects in *TYPE1 and *TYPE2 directories to specify whether
the objects are to be scanned or not, even if the file system is not fully converted.
While the system converts objects from the *TYPE1 directory format to the *TYPE2 directory format, the
Scan control system value Scan on next access after object has been restored is taken
into consideration as if the converted object was being restored. For example, if the Scan on next
access after object has been restored value is specified while the conversion is in progress,
then an object that was in a *TYPE1 directory, and had the the object will not be scanned
attribute specified, will be scanned at least once after the file system has been fully converted.
Related concepts
Scanning support
With the IBM i operating system, you can scan integrated file system objects.
Related system values

98 IBM i: Integrated file system

 You can use the QSCANFS and QSCANFSCTL system values to establish the scanning environment you
want for your system.

Converting names to support additional characters

File systems store names in Unicode. Those file systems that are not case sensitive are affected by
changes in characters and the casing rules for a specific Unicode standard. File systems that are not case
sensitive, such as the "root" (/) and user-defined file systems (UDFS) that are created with CASE(*MONO),
support Unicode Standard 4.0 as of i5/OS V6R1.
Shortly after i5/OS is installed, the conversion of directories to the Unicode Standard 4.0 automatically
begins for any of the file systems that have not yet been converted to support this Unicode standard. This
conversion does not significantly affect your system activity as it runs in a low priority background job.
Related information
Unicode

Overview of automatic name conversion

Some file systems that are not case sensitive, such as “root” (/) and the UDFSs that are created with
CASE(*MONO), support names stored in Unicode Standard 4.0. The system runs an automatic name
conversion to support the additional characters in names.
In the integrated file system, searches for names occur when a name is provided on an integrated file
system CL command or API to identify the object being operated on. When additional characters are
defined or the casing rules are updated to support the new characters, names containing characters that
are affected by the change might no longer be found.
The integrated file system stores all names in Unicode. Prior to IBM i V6R1, the integrated file system
supports Unicode Standard 2.0. As of V6R1, the integrated file system supports Unicode Standard 4.0.
The name conversion utility runs automatically to update the directories in file systems that are not case
sensitive to support Unicode Standard 4.0. The file systems that are not case sensitive and that are
included in this conversion are "root" (/) and user-defined file systems (in any auxiliary storage pool)
created with CASE(*MONO).
The conversion for names in “root” (/) and UDFSs in basic user auxiliary storage pools (1-32) starts
automatically shortly after the i5/OS operating system is installed and runs in a low priority thread of the
QFILESYS1 system job. If the conversion function has not yet completed and the system performs an
initial program load (IPL), the conversion function resumes when the IPL is completed. The conversion
restarts on every IPL until all eligible file systems have been fully converted.
During the vary-on operation of an independent ASP, the job QFSYS##### for independent disk pool
group ##### starts automatically and the conversion for names in UDFSs in the independent ASP
##### runs in a low priority thread of this system job. The name conversion for the UDFSs in the
independent ASPs do not start converting directories until the name conversion for “root” (/) and UDFSs
in basic user ASPs has completed. If the conversion function has not yet completed and the independent
ASP is varied off, or the system performs an IPL, the conversion function resumes when the IPL is
completed and the independent ASP is varied on.
Note: If more than one independent ASP is varied on at the same time, only one independent ASP name
conversion is allowed to be active at any given time. When one conversion is completed, another is
allowed to be active.
You can use the Analyze Object Conversion (ANZOBJCVN) command in a previous release to
plan for this name conversion. The ANZOBJCVN command collects or reports conversion information
for objects on the system including which integrated file system object's names contains at least 1
character that might be affected by the new Unicode characters and casing rules in the new release. The
affected objects could potentially be renamed for the new Unicode characters and casing rules when the
automatic conversion is performed after the upgrade to the new release. As changed object names might
affect applications, you might want to rename any affected objects before upgrading to the new release.

Integrated file system 99

 Related concepts
Name continuity
When you use the "root" (/), QOpenSys, and user-defined file systems, you can take advantage of system
support that ensures characters in object names remain the same.
Related information
Unicode
Planning integrated file system security

Name conversion considerations

Here are several things to consider during the automatic name conversion process.

Conversion status determination

General messages indicating which file systems have been converted and various progress messages are
sent to the job log of the job performing the conversion.
For “root” (/) and UDFSs in ASPs 1-32, the messages are sent to the job log of the QFILESYS1 job.
For UDFSs in independent ASPs, the messages are sent to the job log of the job QFSYS##### for
independent disk pool group #####. These messages include information such as the file system being
converted and the number of links that have been processed in the file system. Many of the general
messages are also sent to the QSYSOPR message queue. Error messages are sent to the QHST history log.
Therefore, for future reference, it is good practice to keep the QHST logs, which contain these messages.
When the file systems have been fully converted and the integrated file system is working as expected,
you can delete this historical information.

Objects renamed
In a file system that is not case sensitive, the system does not distinguish between uppercase and
lowercase characters during searches for names. Any change in the casing rules will affect the characters
that are considered the same when not distinguishing between uppercase and lowercase.
During the automatic name conversion, names that were not considered to be the same in Unicode
Standard 2.0 might now be considered the same in Unicode Standard 4.0. Thus, one of the objects has to
be renamed. When a name is changed to be unique, message CPDA0BC is sent to QHST listing the original
name and the new name.
For more information about different versions of the Unicode standard, refer to the Unicode home page
(www.unicode.org ).

User profile considerations

While the conversion is running, the system needs additional storage, which impacts the amount of
storage owned by the user profile.
If during the name conversion process the maximum storage limit for the user profile is reached, the
system increases the limit for the user profile and sends message CPIA08C to the QSYSOPR message
queue.

Tips: Symbolic link

Symbolic links are objects within the integrated file system that contain path names pointing to other
objects. If one of the elements of the path name within a symbolic link is renamed during the conversion,
the contents of the symbolic link no longer point to the other object.

Tips: Independent ASP

If the user-defined file systems in an independent ASP have not yet been converted to use the Unicode
Standard version for the current release, a job is started to do that conversion when the independent ASP
is varied on.
The system starts the QFSYS##### job for independent disk pool group #####. The automatic name
conversion runs in a low priority thread of this job and one thread is started for each independent auxiliary

100 IBM i: Integrated file system

 storage pool in the group. The system starts this job each time the independent disk pool group is varied
on if the automatic name conversion has not completed.

Tips: Saving and restoring

Directories that are using Unicode Standard 2.0 can be saved and restored to a file system that is using
the Unicode Standard 4.0, and vice versa.
As objects are restored, they are linked into a directory. The Unicode Standard of the directory that the
object is being linked into applies to the name of the object being restored. Therefore, objects cannot be
restored if the name is not unique for the Unicode Standard being used by the directory. In such cases,
the system sends message CPD37B9 with reason code 1 to the job log.

Tips: Reclaiming integrated file system objects

If there are any errors during the automatic name conversion, the Reclaim Storage (RCLSTG) and Reclaim
Object Links (RCLLNK) commands can complete the conversion of the directory and its subdirectories.
This only happens after the automatic name conversion processing has completed for the file system.
Additionally, the RCLSTG command converts any lost directories to use Unicode Standard 4.0.
Related concepts
Reclaim operation of the "root" (/), QOpenSys, and user-defined file systems
Reclaiming the "root" (/), QOpenSys, and user-defined file systems can be accomplished using the
Reclaim Object Links (RCLLNK) and Reclaim Storage (RCLSTG) commands.
Related information
Reclaim Storage (RCLSTG) command
Reclaim Object Links (RCLLNK) command

Journaling objects
The primary purpose of journaling is to enable you to recover the changes to an object that have occurred
since the object was last saved. Additionally, a key use of journaling is to assist in the replication of object
changes to another system either for high availability or workload balancing.
This information will provide a brief overview of journal management, as well as provide considerations
for journaling integrated file system objects and a description of journaling support for integrated file
system objects.
Related information
Journal management

Journaling overview
These topics introduce journaling support for integrated file system objects.
Related information
Journal management

Journal management
The main purpose for journal management is to enable you to recover the changes to an object that have
occurred since the object was last saved.
You can also use journal management for:
• An audit trail of activity that occurs for objects on the system
• Recording activity that has occurred for objects other than those you can journal
• Quicker recovery when restoring from save-while-active media
• Assist in the replication of object changes to another system either for high availability or workload
balancing
• Assistance in testing application programs

Integrated file system 101

 You can use a journal to define what objects you want to protect with journal management. In the
integrated file system , you can journal stream files, directories, and symbolic links. Only objects in "root"
(/), QOpenSys, and UDFS file systems are supported.
Related concepts
Objects you should journal
There are certain questions to consider when deciding whether to journal an integrated file system object.

Objects you should journal

There are certain questions to consider when deciding whether to journal an integrated file system object.
You need to consider the following questions to determine which objects you need to journal:
• How much does the object change? An object with a high volume of changes between save operations is
a good candidate for journaling.
• How difficult is it to reconstruct the changes made to an object? Are many changes made to the object
without written records? For example, an object used for telephone order entries is more difficult to
reconstruct than an object used for orders that arrive in the mail on order forms.
• How critical is the information in the object? If the object had to be restored back to the last save
operation, what effect does the delay in reconstructing changes have on the business?
• How does the object relate to other objects on the system? Although the data in a particular object
might not change often, that object's data might be critical to other more dynamic objects on the
system. For example, many objects depend on a customer master file. If you are reconstructing orders,
the customer master file must include new customers or changes in credit limits that have taken place
since the previous save operation.

Journaled integrated file system objects

Some integrated file system object types can be journaled using IBM i journaling support.
The object types supported are stream files, directories, and symbolic links. The "root" (/), QOpenSys,
and UDFS are the only file systems that support journaling of these object types. Integrated file system
objects can be journaled using either the traditional system interface (CL commands or APIs) or by using
IBM Navigator for i. You can start journaling and end journaling through IBM Navigator for i, as well as
display journaling status information.
Note: Memory-mapped stream files, virtual volume files, and stream files that are used as Integrated
xSeries Server (IXS) network storage spaces cannot be journaled. Directories that can contain block
special file (*BLKSF) objects cannot be journaled. Examples of these are: /dev/QASP01, /dev/QASP22,
and /dev/IASPNAME.
The following list summarizes journaling support in the integrated file system:
• You can use both generic commands and APIs to perform journal operations on the supported object
types. These interfaces generally accept object identification in the form of a path name, file ID, or both.
• Some journal operation commands, including Start Journal, End Journal, Change Journaled Object, and
Apply Journaled Changes, can be performed on entire subtrees of integrated file system objects. You
can optionally use the include and exclude lists that can use wildcard patterns for object names. For
example, you can use the Start Journal command to specify to start journaling on all objects in the tree
"/MyCompany" that match the pattern "*.data" but that excludes any objects matching the patterns
"A*.data" and "B*.data".
• Journaling support on directories includes directory operations such as adding links, removing links,
creating objects, renaming objects, and moving objects within the directory.
Journaled directories support an attribute that can be set to cause new objects in the subtree to inherit
the current journaling state of the directory. When this attribute is turned on for a journaled directory,
all stream files, directories, and symbolic links that are created or linked into the directory (by adding a
hard link or by renaming or moving the object) will automatically have journaling started by the system.
Note: Inherit journaling attribute considerations:

102 IBM i: Integrated file system

 – If you rename an object in the same directory it currently resides in, journaling is not started for the
object, even if the directory has the inherit current journaling state attribute on.
– When a directory is moved to a directory which has the inherit journaling attribute on, only that
moved directory has journaling started for it if appropriate. The objects within that moved directory
are not affected.
– If an object is restored to a directory which has the inherit journaling attribute on, journaling is not
started for that object if the object has ever been journaled.
– When using the Apply Journaled Changes (APYJRNCHG) command, the current value of the
inherit journaling attribute for any directories is not used. Instead, any objects which are created as
part of the apply have journaling started or not based on what happened during the runtime activity
which is being applied.
• Object names and complete path names are contained within several journal entries of integrated file
system objects. Object names and path names are National Language Support (NLS)-enabled.
• If the system ends abnormally, system initial program load (IPL) recovery is provided for journaled
integrated file system objects.
• The maximum write limit supported by the various write interfaces is 2 GB - 1. The maximum journal
entry size if RCVSIZOPT (*MAXOPT2 or *MAXOPT3) is specified is 4 000 000 000 bytes. Otherwise, the
maximum journal entry size is 15 761 440 bytes. If you journal your stream file and have any writes that
exceed 15 761 440 bytes, you need to use the *MAXOPT2 or *MAXOPT3 support to prevent any errors
from occurring.
For more information about the layout of various journal entries, there is a C language include file,
qp0ljrnl.h, shipped in member QSYSINC/H (QP0LJRNL), that contains details of the integrated file system
journal entry specific data content and formats.
Related concepts
Stream file
A stream file is a randomly accessible sequence of bytes, with no further structure imposed by the system.
Directory
A directory is a special object that is used to locate objects by names that you specify. Each directory
contains a list of objects that are attached to it. That list can include other directories.
Symbolic link
A symbolic link, which is also called a soft link, is a path name contained in a file.
Related tasks
Starting journaling
To start journaling, do these steps on an object through the IBM Navigator for i.
Ending journaling
After journaling has started on an object and, for whatever reasons, you want to end journaling on this
object, you can use the steps described in this topic.
Changing journaling
After journaling has started on an object and, for whatever reasons, you want to change the journal
attributes on the object without having to end and restart journaling, you can use the Change
Journaled Object (CHGJRNOBJ) command to change journaled objects.
Related information
Journal management
Journal entry information finder

Journaled operations
These operations are only journaled when the type of the object or link that the operation is using is a type
that can also be journaled.
• Create an object.
• Add a link to an existing object.

Integrated file system 103

 • Unlink a link.
• Rename a link.
• Rename a file identifier.
• Move a link into or out of the directory.
The following journaled operations are specific to a stream file:
• Data write or clear
• File truncate/extend
• File data forced
• Save with storage freed
The following journaled operations apply to all journaled object types:
• Attribute changes (including security changes such as authorities and ownership)
• Open
• Close
• Start journaling
• Change Journaled Object (CHGJRNOBJ) command
• End journaling
• Start the Apply Journaled Changes (APYJRNCHG) command
• End the Apply Journaled Changes (APYJRNCHG) command
• Save
• Restore
Related information
Journal Management
Journal entry information finder

Special considerations for journal entries

Many journaled integrated file system operations internally use commitment control to form a single
transaction from the multiple functions performed during the operations.
These journaled operations should not be considered complete unless the commitment control cycle has
a Commit journal entry (Journal Code C, Type CM). Journaled operations that contain a Rollback journal
entry (Journal Code C, Type RB) in the commitment control cycle are failed operations, and the journal
entries within them should not be replayed or replicated.
Journaled integrated file system entries (Journal Code B) that use commitment control in this manner
include:
• AA — Change Audit Value
• B0 — Begin Create
• B1 — Create Summary
• B2 — Add link
• B3 — Rename/Move
• B4 — Unlink (Parent Directory)
• B5 — Unlink (Link)
• B7 — Created object authority information
• FA — Attribute Change
• JT — Start Journal (only when journaling is started because of an operation in a directory with the inherit
journaling attribute of Yes)

104 IBM i: Integrated file system

 • OA — Authority Change
• OG — Object Primary Group Change
• OO — Object Owner Change
Several integrated file system journal entries have a specific data field indicating whether the entry is a
summary entry. Operations that send summary entry types will send two of the same entry types to the
journal. The first entry contains a subset of the entry specific data. The second entry contains complete
entry specific data and will indicate that it is a summary entry. Programs that are replicating the object or
replaying the operation will generally only be interested in the summary entries.
For a create operation in a journaled directory, the B1 journal entry (Create Summary) is considered the
summary entry.
Some journaled operations need to send a journal entry that is conversely related to the operation. For
example, a commitment control cycle containing a B4 journal entry (Unlink) may also contain a B2 journal
entry (Add Link). This type of scenario will only occur in operations that result in a Rollback journal entry
(C — RB) .
This scenario may occur for two reasons:
1. The operation was about to fail, and the entry was needed internally for error path cleanup.
2. The operation was interrupted by a system outage, and during the subsequent IPL, recovery that
needed to send the entry was performed to rollback the interrupted operation.
Related information
Journal entry information finder

Considerations for multiple hard links and journaling

If you have multiple hard links to a journaled integrated file system object, all the links should be saved
and restored together so that the linkage is preserved as well as the associated journal information.
If specifying names in some of the journal related commands and if the names are really multiple hard
links, then the object will only be operated on 'once'. The other hard links are essentially ignored.
Because multiple hard links point to the same object, and the journal entry has only the file identifier
(File ID), which is the same for the object, then any journal interfaces that show the path name, for
example, Display Journal (DSPJRN), show only one link name for the object. However, this should not
cause problems because one can operate on an object by any name and get the same result.
Related concepts
Hard link
A hard link, which is sometimes just called a link, cannot exist unless it is linked to an actual object.

Starting journaling
To start journaling, do these steps on an object through the IBM Navigator for i.

Procedure
1. Start IBM Navigator for i.
2. Expand File Systems.
3. Right-click the object that you want to journal in the right work space, and select Journaling.
4. After selecting the appropriate journaling options, click Start.

Results
To start journaling on an object through the character-based interface, you can use either the Start
Journal (STRJRN) command or the QjoStartJournal API.

Integrated file system 105

 Related information
Start Journal (STRJRN) command
Start Journal (QjoStartJournal) API
Journal management

Changing journaling
After journaling has started on an object and, for whatever reasons, you want to change the journal
attributes on the object without having to end and restart journaling, you can use the Change
Journaled Object (CHGJRNOBJ) command to change journaled objects.
Related tasks
Starting journaling
To start journaling, do these steps on an object through the IBM Navigator for i.
Ending journaling
After journaling has started on an object and, for whatever reasons, you want to end journaling on this
object, you can use the steps described in this topic.
Related information
Change Journaled Object (CHGJRNOBJ) command

Ending journaling
After journaling has started on an object and, for whatever reasons, you want to end journaling on this
object, you can use the steps described in this topic.

About this task

To end journaling on an object through IBM Navigator for i, follow these steps:

Procedure
1. Start IBM Navigator for i.
2. Expand File Systems.
3. Right-click the object that you want to stop journaling in the right work space, and select Journaling.
4. Click End.

Results
To end journaling on an object through the character-based interface, you can use either the End
Journal (ENDJRN) command or the QjoEndJournal API.
Related tasks
Starting journaling
To start journaling, do these steps on an object through the IBM Navigator for i.
Related information
End Journal (ENDJRN) command
End Journal (QjoEndJournal) API
Journal management

Reclaim operation of the "root" (/), QOpenSys, and user-defined

file systems
Reclaiming the "root" (/), QOpenSys, and user-defined file systems can be accomplished using the
Reclaim Object Links (RCLLNK) and Reclaim Storage (RCLSTG) commands.
By using the RCLLNK and RCLSTG commands, you can perform the following tasks:

106 IBM i: Integrated file system

 • Correct object user profile problems
• Correct user-defined file system problems
• Correct internal object problems
• Remove invalid object links
• Handle damaged objects
• Create missing system objects
• Correct internal file system problems (RCLSTG only)
• Find lost objects (RCLSTG only)
• Complete the conversion of directories to use Unicode Standard 4.0 when errors occurred during the
automatic name conversion

Reclaim Object Links (RCLLNK) and Reclaim Storage (RCLSTG)

commands comparison
You can use both the Reclaim Object Links (RCLLNK) and Reclaim Storage (RCLSTG)
commands to correct problems in the "root" (/), QOpenSys, and user-defined file systems.
The RCLLNK command identifies and, if possible, corrects problems in mounted file systems that are
in use. The RCLSTG command does not have this function. However, the RCLSTG command can correct
problems that the RCLLNK command is unable to identify or correct. The following table provides a more
detailed comparison between the two commands.

Table 11. RCLLNK and RCLSTG command comparison

RCLLNK OBJ('/MyDir/ RCLSTG RCLSTG

MyObj') ASPDEV(*SYSBAS) ASPDEV(<IASPNAME>)

Is the system required

to be in a restricted No Yes No
state?
Are all file systems The file systems in the
usable during the Yes No independent ASP being
reclaim operation? reclaimed are unusable.
Reclaims objects in
In what ASPs can Reclaims objects in Reclaims objects in
system, user, and
objects be reclaimed? system and user ASPs. independent ASPs.
independent ASPs.
Objects are reclaimed on
Objects are reclaimed Objects are reclaimed
How are objects an individual or subtree
on a system-wide on an independent ASP
reclaimed? basis as specified on the
basis. basis.
command.
Most (See “Reclaim
What known and
operation of the "root"
applicable file system
(/), QOpenSys, and user-
problems are identified All All
defined file systems”
and corrected, if
on page 106 for more
possible?
information.)
Are lost objects found? No Yes Yes
Are objects in
unmounted file No Yes Yes
systems reclaimed?

Integrated file system 107

 Table 11. RCLLNK and RCLSTG command comparison (continued)

RCLLNK OBJ('/MyDir/ RCLSTG RCLSTG

MyObj') ASPDEV(*SYSBAS) ASPDEV(<IASPNAME>)

Is the command
Yes No No
threadsafe?
How many instances of
the command can be
Multiple instances Single instance Single instance
performed at the same
time?
What applicable Most (See “Re-creation
integrated file system of integrated file
provided objects All system provided None
are re-created if objects” on page 108
necessary? for more information.)
Can damaged objects
be identified without Yes No No
being reclaimed?

Related concepts
Examples: Reclaim Object Links (RCLLNK) command
These examples describe situations in which the Reclaim Object Links (RCLLNK) command can be
used to reclaim objects in the "root" (/), QOpenSys, and mounted user-defined file systems.
Related reference
Re-creation of integrated file system provided objects
This table shows the objects provided by the integrated file system that the Reclaim Object Links
(RCLLNK) command re-creates if they do not exist. These objects are normally created during the initial
program load (IPL). You can also re-create some of these objects, if necessary, using the Reclaim
Storage (RCLSTG) command.
Related information
Reclaim Storage (RCLSTG) command
Reclaim Object Links (RCLLNK) command

Reclaim Object Links (RCLLNK) command

The Reclaim Object Links (RCLLNK) command identifies and repairs damaged objects in the "root"
(/), QOpenSys, and mounted user-defined file systems without requiring the system to be in a restricted
state. You can correct problems in these file systems without sacrificing productivity.
The RCLLNK command can be used as an alternative to the Reclaim Storage (RCLSTG) command in
many situations. For example, RCLLNK is ideal for identifying and correcting problems in the following
situations:
• Problems are isolated to a single object.
• Problems are isolated to a group of objects.
• Damaged objects need to be identified or deleted.
• The system cannot be in a restricted state during the reclaim operation.
• Independent ASPs must be available during the reclaim operation.

Re-creation of integrated file system provided objects

This table shows the objects provided by the integrated file system that the Reclaim Object Links
(RCLLNK) command re-creates if they do not exist. These objects are normally created during the initial

108 IBM i: Integrated file system

 program load (IPL). You can also re-create some of these objects, if necessary, using the Reclaim
Storage (RCLSTG) command.

Table 12. Objects provided by the integrated file system and re-created by the RCLLNK and RCLSTG
commands
Recreated by RCLSTG
Path name Type Recreated by RCLLNK ASPDEV(*SYSBASE)
/dev/zero *CHRSF Yes Yes
/dev/null *CHRSF Yes Yes
/dev/xti/tcp *CHRSF Yes No
/dev/xti/udp *CHRSF Yes No
/etc/vfs *STMF Yes No

In order for the RCLLNK command to re-create an object provided by the integrated file system that
does not exist, it must be run with the SUBTREE parameter set to *DIR or *ALL while specifying the
parent directory. The command must successfully reclaim the parent directory of the system object. For
example,

RCLLNK OBJ('/dev') SUBTREE(*DIR)

re-creates the /dev/zero and /dev/null *CHRSF objects if they do not exist.
In order for the RCLSTG command to re-create an integrated file system provided object that does not
exist, it must be run with the ASPDEV parameter set to *SYSBASE and the directory recovery portion of
reclaim must not be omitted.
Related concepts
Provided directories
When the system is restarted, the integrated file system creates the directories listed here if they do not
already exist. These directories should not be moved or renamed after being created by the system.
Related information
Reclaim Object Links (RCLLNK) command

Examples: Reclaim Object Links (RCLLNK) command

These examples describe situations in which the Reclaim Object Links (RCLLNK) command can be
used to reclaim objects in the "root" (/), QOpenSys, and mounted user-defined file systems.

Example: Correcting problems for an object

In this situation, the known problems are isolated to one object. The object is damaged and unusable, and
you cannot restore a backup version of the object from media. You need to correct the problem quickly
without disrupting normal file system operations.
To reclaim the object, use this command:

RCLLNK OBJ('/MyDir/MyBadObject') SUBTREE(*NONE)

where /MyDir/MyBadObject is the name of the damaged and unusable object.

Integrated file system 109

 Example: Correcting problems that exist in a directory subtree
In this situation, the known problems are isolated to a group of objects within a directory subtree. An
application is failing due to the problems within the directory subtree. You need to correct the problems
quickly without disrupting normal file system operations.
To reclaim the objects within the directory subtree, use this command:

RCLLNK OBJ('/MyApplicationInstallDirectory') SUBTREE(*ALL)

where MyApplicationInstallDirectory is the name of the directory containing the problem objects.

Example: Finding all damaged objects in the "root" (/), QOpenSys, and
mounted user-defined file systems
In this situation, a disk failure has caused damage to a number of objects. You must identify the damaged
objects before determining how to properly recover them.
You need a solution to identify the damaged objects, but not take action against them. You must not
disrupt normal file system operations.
To identify the damaged objects, use this command:

RCLLNK OBJ('/') SUBTREE(*ALL) DMGOBJOPT(*KEEP *KEEP)

In addition, this command will also correct problems other than damaged objects as it identifies damaged
objects.

Example: Deleting all damaged objects in the "root" (/), QOpenSys, and
mounted user-defined file systems
In this situation, a disk failure caused a number of objects to become damaged. You must delete the
damaged objects so that a backup copy of the objects can be restored from media.
To delete the damaged objects, use this command:

RCLLNK OBJ('/') SUBTREE(*ALL) DMGOBJOPT(*DELETE *DELETE)

The damaged objects are deleted without disruption to normal file system operations. In addition,
problems other than damage are corrected as the damaged objects are being deleted.

Example: Running multiple RCLLNK commands to quickly reclaim all objects

in the "root" (/), QOpenSys, and mounted user-defined file systems
In this situation, as part of routine system maintenance, all objects in the "root" (/), QOpenSys, and
mounted user-defined file systems are reclaimed. You want to finish the reclaim operation as quickly as
possible to allow for additional system maintenance to be completed.
By breaking the reclaim operation into separate groups, multiple RCLLNK commands can be performed
concurrently to allow the reclaim operation to finish sooner.
To run multiple reclaim operations against key system directories and other top-level directories, use the
following commands (each in a separate job or thread).

RCLLNK OBJ('/') SUBTREE(*DIR)

RCLLNK OBJ('/tmp') SUBTREE(*ALL)
RCLLNK OBJ('/home') SUBTREE(*ALL)
RCLLNK OBJ('/etc') SUBTREE(*ALL)
RCLLNK OBJ('/usr') SUBTREE(*ALL)
RCLLNK OBJ('/QIBM') SUBTREE(*ALL)
RCLLNK OBJ('/QOpenSys') SUBTREE(*ALL)
RCLLNK OBJ('/IaspName') SUBTREE(*ALL)
RCLLNK OBJ('/dev') SUBTREE(*ALL)
RCLLNK OBJ('/OtherTopLevelDirectories') SUBTREE(*ALL)

110 IBM i: Integrated file system

 where OtherTopLevelDirectories are other directories that you want to reclaim.

Programming support
To take advantage of the stream files, directories, and other support of the integrated file system, you
need to use a set of application programming interfaces (APIs) provided for accessing integrated file
system functions.
Additionally, the addition of the integrated file system allows you to copy data between physical database
files and stream files. You can perform this copy using CL commands, the data transfer function of IBM i
Access Family, or APIs.

Copying data between stream files and database files

If you are familiar with operating on database files using record-oriented facilities such as data
description specifications (DDS), you might find some fundamental differences in the way you operate
on stream files.
The differences result from the different structure (or perhaps lack of structure) of stream files in
comparison with database files. To access data in a stream file, you indicate a byte offset and a length. To
access data in a database file, you typically define the fields to be used and the number of records to be
processed.
Because you define the format and characteristics of a record-oriented file ahead of time, the operating
system has knowledge of the file and can help you avoid performing operations that are not appropriate
for the file format and characteristics. With stream files, the operating system has little or no knowledge
of the format of the file. The application must know what the file looks like and how to operate on it
properly. Stream files allow an extremely flexible programming environment, but at the cost of having
little or no help from the operating system. Stream files are better suited for some programming
situations; record-oriented files are better suited for other programming situations.
Related concepts
Stream file
A stream file is a randomly accessible sequence of bytes, with no further structure imposed by the system.

Copying data using CL commands

There are two sets of CL commands that allow you to copy data between stream files and database file
members.

CPYTOSTMF and CPYFRMSTMF commands

You can use the Copy from Stream File (CPYFRMSTMF) and Copy to Stream File
(CPYTOSTMF) commands to copy data between stream files and database file members. You can create
a stream file from a database file member by using the CPYTOSTMF command. You can also create a
database file member from a stream file by using the CPYFRMSTMF command. If the file or member that is
the target of the copy does not exist, it is created.
There are some limitations, however. The database file must be either a program-described physical file
containing only one field or a source physical file containing only one text field. The commands give you a
variety of options for converting and reformatting the data that is being copied.
The CPYTOSTMF and CPYFRMSTMF commands can also be used to copy data between a stream file and a
save file.

CPYTOIMPF and CPYFRMIMPF commands

You can also use the Copy to Import File (CPYTOIMPF) and Copy from Import File
(CPYFRMIMPF) commands to copy data between stream files and database members. The CPYTOSTMF
and CPYFRMSTMF commands do not allow you to move data from complex, externally-described (DDS-

Integrated file system 111

 described) database files. The word import file refers to the stream type file; the term typically refers to a
file created for purposes of copying data between heterogeneous databases.
When copying from a stream (or import) file, the CPYFRMIMPF command allows you to specify a field
definition file (FDF), which describes the data in the stream file. Or, you can specify that the stream
files is delimited, and what characters are used to mark string, field, and record boundaries. Options for
converting special data types such as time and date are also provided.
Data conversion is provided on these commands if the target stream file or database member already
exists. If the file does not exist, you can use the following two-step method to get the data converted:
1. Use the CPYTOIMPF and CPYFRMIMPF commands to copy the data between the externally-described
file and a source physical file.
2. Use the CPYTOSTMF and CPYFRMSTMF commands (which provide full data conversion regardless of
whether the destination file exists) to copy between the source physical file and the stream file.
Here is an example:

CPYTOIMPF FROMFILE(DB2FILE) TOFILE(EXPFILE) DTAFMT(*DLM)

FLDDLM(';') RCDDLM(X'07') STRDLM(*DBLQUOTE) DATFMT(*USA) TIMFMT(*USA)

The DTAFMT parameter specifies that the input stream (import) file is delimited; the other choice is
DTAFMT(*FIXED), which requires an field definition file to be specified. The FLDDLM, RCDDLM and
STRDLM parameters identify the characters that act as the delimiters, or separators for fields, records,
and strings.
The DATFMT and TIMFMT parameters indicate the format for any date and time information that is copied
to the import file.
The commands are useful because they can be placed into a program, and they run entirely on your
system. However, the interfaces are complex.
Related information
Copy to Stream File (CPYTOSTMF) command
Copy from Stream File (CPYFRMSTMF) command
Copy to Import File (CPYTOIMPF) command
Copy from Import File (CPYFRMIMPF) command
Control language (CL)

Copying data using APIs

If you want to copy database file members to a stream file in an application, you can use the integrated
file system open(), read(), and write() functions to open a member, read data from it, and write data
to it or another file.
Related information
open()--Open File API
read()--Read from Descriptor API
write()--Write to Descriptor API
Integrated file system APIs

Copying data using data-transfer functions

The data-transfer applications in the IBM i Access Family licensed program have the advantage of an
easy-to-follow graphical interface, and automatic numeric and character data conversion.
However, data transfer requires the installation of the IBM i Access Family product and requires the use of
both PC and IBM i resources and communications between the two.
If you have IBM i Access Family installed on the PC and your system, you can use the data transfer
applications to transfer data between stream files and database files. You can also transfer data into a

112 IBM i: Integrated file system

new database file that is based on an existing database file, into an externally described database file, or
into a new database file definition and file.

Transferring data from a database file to a stream file

To transfer a file from a database file to a stream file on your system, follow these steps.
1. Establish a connection to the system.
2. Map a network drive to the appropriate path in the IBM i file system.
3. From the IBM i Access for Windows window, click Data Transfer From IBM i .
4. Select the system you want to transfer from.
5. Select the file names, using the IBM i database library and file name to copy from, and the network
drive for the location of the resulting stream file. You can also click PC File Details to select the PC
file format for the stream file. Data transfer supports common PC file types, such as ASCII text, BIFF3,
CSV, DIF, Tab-delimited Text, or WK4.
6. Click Transfer data from IBM i to run the file transfer.
You can also perform this data movement in a batch job with the data transfer applications. Proceed as
above, but select the File menu option to save the transfer request. The Data Transfer To IBM i application
creates a .DTT or a .TFR file. The Data Transfer From IBM i application creates a .DTF or a .TTO file. In the
IBM i Access Family directory, two programs can be run in batch from a command line:
• RTOPCB takes either a .DTF or a .TTO file as a parameter
• RFROMPCB takes either a .DTT or a .TFR file as a parameter
You can set either of these commands to run on a scheduled basis by using a scheduler application. For
example, you can use the System Agent Tool (a part of the Microsoft Plus Pack) to specify the program to
run (for instance, RTOPCB MYFILE.TTO) and the time at which you want to run the program.

Transferring data from a stream file to a database file

To transfer data from a stream file to a database file on your system, follow these steps.
1. Establish a connection to the system.
2. Map a network drive to the appropriate path in the IBM i file system.
3. From the IBM i Access for Windows window, click Data Transfer To IBM i.
4. Select the PC file name that you want to transfer. For the PC file name, you can choose Browse for the
network drive you assigned, and choose a stream file. You can also use a stream file located on the PC
itself.
5. Select the system on which you want the externally described database file to be located.
6. Click Transfer data to IBM i to run the file transfer.
Note: If you are moving data to an existing database file definition on the system, the Data Transfer to
IBM i application requires you to use an associated format description file (FDF). A .FDF file describes
the format of a stream file, and is created by the Data Transfer from IBM i application when data is
transferred from a database file to a stream file. To complete the transfer of data from a stream file
to a database file, click Transfer data to IBM i. If an existing .FDF file is not available, you can quickly
create a .FDF file.
You can also perform this data movement in a batch job with the data transfer applications. Proceed as
above, but select the File menu option to save the transfer request. The Data Transfer To IBM i application
creates a .DTT or a .TFR file. The Data Transfer From IBM i application creates a .DTF or a .TTO file. In the
IBM i Access Family directory, two programs can be run in batch from a command line:
• RTOPCB takes either a .DTF or a .TTO file as a parameter
• RFROMPCB takes either a .DTT or a .TFR file as a parameter
You can set either of these commands to run on a scheduled basis by using a scheduler application. For
example, you can use the System Agent Tool (a part of the Microsoft Plus Pack) to specify the program to
run (for instance, RTOPCB MYFILE.TTO) and the time at which you want to run the program.

Integrated file system 113

 Related reference
Creating a format description file
If you are moving data to an existing database file definition on the system, the Data Transfer to IBM i
application requires you to use an associated format description file (FDF).

Transferring data into a newly created database file definition and file
You can follow these directions to transfer data into a newly created database file definition and file.
1. Establish a connection to the system.
2. Map a network drive to the appropriate path in the IBM i file system.
3. From the IBM i Access for Windows window, click Data Transfer to IBM i.
4. Open Tools of the Data Transfer to IBM i application.
5. Click Create IBM i database file.
A wizard appears that allows you to create a new IBM i database file from an existing PC file. You need
to specify the name of the PC file from which the IBM i file will be based, the name of the IBM i file
to create, and several other necessary details. This tool parses a given stream file to determine the
number, type, and size of the fields that are required in the resulting database file. The tool can then
create the database file definition on your system.

Creating a format description file

If you are moving data to an existing database file definition on the system, the Data Transfer to IBM i
application requires you to use an associated format description file (FDF).
An FDF file describes the format of a stream file, and is created by the Data Transfer from IBM i
application when data is transferred from a database file to a stream file.
To create a .FDF file:
1. Create an externally described database file with a format that matches your source stream file
(number of fields, types of data).
2. Create one temporary data record within the database file.
3. Use the Data Transfer from IBM i function to create a stream file and its associated .FDF file from this
database file.
Now, you can use the Data Transfer to IBM i function. Specify this .FDF file with the source stream file
that you want to transfer.
Related reference
Transferring data from a database file to a stream file
To transfer a file from a database file to a stream file on your system, follow these steps.
Transferring data from a stream file to a database file
To transfer data from a stream file to a database file on your system, follow these steps.

Copying data between stream files and save files

A save file is used with save and restore commands to retain data that would otherwise be written on tape
or diskette.
The file can also be used like a database file to read or write records that contain save/restore
information. A save file can also be used to send objects to another user on the SNADS network.
You can use the Copy Object (CPY) command to copy a save file both to and from a stream file.
However, when copying a stream file back into a save file object, the data must be valid save file data (it
must have originated from a save file and been copied into a stream file).
By using a PC client, you can also access the save file and copy the data to your PC storage or LAN. Keep
in mind, though, that data in save files cannot be accessed through the Network File System (NFS).

114 IBM i: Integrated file system

 Related information
Copy Object (CPY) command

Performing operations using APIs

Many of the application programming interfaces (APIs) that perform operations on integrated file system
objects are in the form of C language functions.
You have a choice of two sets of functions, either of which you can use in programs that are created using
Integrated Language Environment (ILE) C:
• Integrated file system C functions that are included in the IBM i operating system.
• C functions provided by the ILE C licensed program.
For information about the exit programs that the integrated file system supports, see Table 14 on page
121.
The integrated file system functions operate only through the integrated file system stream I/O support.
The following APIs are supported:

Table 13. Integrated file system APIs

Function Description
access() Determine file accessibility
accessx() Determine file accessibility for a class of users
chdir() Change current directory
chmod() Change file authorizations
chown() Change owner and group of file
close() Close file descriptor
closedir() Close directory
creat() Create new file or rewrite existing file
Create new file or rewrite existing file (large file
creat64()
enabled)
DosSetFileLocks() Lock and unlock byte range of a file
DosSetFileLocks64() Lock and unlock byte range of a file (large file enabled)
DosSetRelMaxFH() Change the maximum number of file descriptors
dup() Duplicate open file descriptor
dup2() Duplicate open file descriptor to another descriptor
Determine file accessibility for a class of users by
faccessx()
descriptor
fchdir() Change current directory by descriptor
fchmod() Change file authorizations by descriptor
fchown() Change owner and group of file by descriptor
fclear() Clear a file
fclear64() Clear a file (large file enabled)
fcntl() Perform file control action
fpathconf() Get configurable path name variables by descriptor

Integrated file system 115

 Table 13. Integrated file system APIs (continued)
Function Description
fstat() Get file information by descriptor
fstat64() Get file information by descriptor (large file enabled)
Get file information by descriptor (large file and
fstat64_time64()
time64_t enabled)
fstatvfs() Get information by descriptor
fstatvfs64() Get information by descriptor (64-bit enabled)
fsync() Synchronize changes to file
ftruncate() Truncate file
ftruncate64() Truncate file (large file enabled)
getcwd() Get path name of current directory
getegid() Get effective group ID
geteuid() Get effective user ID
getgid() Get real group ID
getgrgid() Get group information using group ID
getgrgid_r() Get group information using group ID (threadsafe)
Get group information using group ID (threadsafe with
getgrgid_r_ts64()
8-byte pointers)
getgrnam() Get group information using group name
getgrnam_r() Get group information using group name (threadsafe)
Get group information using group name (threadsafe
getgrnam_r_ts64()
with 8-byte pointers)
getgroups() Get group IDs
getpwnam() Get user information for user name
getpwnam_r() Get user information for user name (threadsafe)
Get user information for user name (threadsafe with
getpwnam_r_ts64()
8-byte pointers)
getpwuid() Get user information for user ID
getpwuid_r() Get user information for user ID (threadsafe)
Get user information for user ID (threadsafe with 8-
getpwuid_r_ts64()
byte pointers)
getuid() Get real user ID
givedescriptor() Give file access to another job
ioctl() Perform file I/O control action
lchown Change owner and group of file or symbolic link
link() Create link to file
lseek() Set file read/write offset

116 IBM i: Integrated file system

Table 13. Integrated file system APIs (continued)
Function Description
lseek64() Set file read/write offset (large file enabled)
lstat() Get file or link information
lstat64() Get file or link information (large file enabled)
Get file or link information (large file and time64_t
lstat64_time64()
enabled)
mkdir() Make directory
mkfifo() Make FIFO special file
mmap() Create a memory map
mmap64() Create a memory map (large file enabled)
mprotect() Change a memory map protection
msync() Synchronize a memory map
munmap() Remove a memory map
open() Open file
open64() Open file (large file enabled)
opendir() Open directory
pathconf() Get configurable path name variables
pipe() Create an interprocess channel
pread() Read from descriptor with offset
pread64() Read from descriptor with offset (large file enabled)
pwrite() Write to descriptor with offset
pwrite64() Write to descriptor with offset (large file enabled)
QjoEndJournal() End journaling
QjoRetrieveJournal Information() Retrieve journal information
QjoRetrieveJournalEntries() Retrieve Journal Entries
QJORJIDI() Retrieve journal identifier information
QJOSJRNE() Send journal entry
QjoStartJournal() Start journaling
Determine file accessibility (using NLS-enabled path
QlgAccess()
name)
Determine file accessibility for a class of users (using
QlgAccessx()
NLS-enabled path name)
QlgChangeAuthority() Change authority (using NLS-enabled path name)
Change current directory (using NLS-enabled path
QlgChdir()
name)
Change file authorizations (using NLS-enabled path
QlgChmod()
name)

Integrated file system 117

 Table 13. Integrated file system APIs (continued)
Function Description
Change owner and group of file (using NLS-enabled
QlgChown()
path name)
Create new file or rewrite existing file (using NLS-
QlgCreat()
enabled path name)
Create new file or rewrite existing file (large file
QlgCreat64()
enabled and using NLS-enabled path name)
Resolve Integrated File System path name into QSYS
QlgCvtPathToQSYSObjName()
Object Name (using NLS-enabled path name)
Get system attributes for an object (using NLS-enabled
QlgGetAttr()
path name)
Get path name of current directory (using NLS-enabled
QlgGetcwd()
path name)
Get path name of object from its file ID (using NLS-
QlgGetPathFromFileID()
enabled path name)
Get user information for user name (using NLS-enabled
QlgGetpwnam()
path name)
Get user information for user name (using NLS-enabled
QlgGetpwnam_r()
path name)
Get user information for user ID (using NLS-enabled
QlgGetpwuid()
path name)
Get user information for user ID (using NLS-enabled
QlgGetpwuid_r()
path name)
Change owner and group of symbolic link (using NLS-
QlgLchown()
enabled path name)
QlgLink() Create link to file (using NLS-enabled path name)
Get file or link information (using NLS-enabled path
QlgLstat()
name)
Get file or link information (large file enabled and using
QlgLstat64()
NLS-enabled path name)
Get file or link information (large file and time64_t
QlgLstat64_time64()
enabled and using NLS-enabled path name)
QlgMkdir() Make directory (using NLS-enabled path name)
QlgMkfifo() Make FIFO special file (using NLS-enabled path name)
QlgOpen() Open file (using NLS-enabled path name)
Open file (large file enabled and using NLS-enabled
QlgOpen64()
path name)
QlgOpendir() Open directory (using NLS-enabled path name)
Get configurable path name variables (using NLS-
QlgPathconf()
enabled path name)
Process directories or objects within a directory tree
QlgProcessSubtree()
(using NLS-enabled path name)

118 IBM i: Integrated file system

Table 13. Integrated file system APIs (continued)
Function Description
QlgReaddir() Read directory entry (using NLS-enabled path name)
Read directory entry (threadsafe and using NLS-
QlgReaddir_r()
enabled path name)
Read value of symbolic link (using NLS-enabled path
QlgReadlink()
name)
Rename file or directory, keep new if it exists (using
QlgRenameKeep()
NLS-enabled path name)
Rename file or directory, unlink new if it exists (using
QlgRenameUnlink()
NLS-enabled path name)
QlgRmdir() Remove directory (using NLS-enabled path name)
Save objects data and free its storage (using NLS-
QlgSaveStgFree()
enabled path name)
Set system attributes for an object (using NLS-enabled
QlgSetAttr()
path name)
QlgStat() Get file information (using NLS-enabled path name)
Get file information (large file enabled and using NLS-
QlgStat64()
enabled path name)
Get file information (large file and time64_t enabled
QlgStat64_time64()
and using NLS-enabled path name)
Get file system information (using NLS-enabled path
QlgStatvfs()
name)
Get file system information (large file enabled and
QlgStatvfs64()
using NLS-enabled path name)
QlgSymlink() Make symbolic link (using NLS-enabled path name)
QlgUnlink() Unlink file (using NLS-enabled path name)
Set file access and modification times (using NLS-
QlgUtime()
enabled path name)
Set file access and modification times (time64_t
QlgUtime_time64() enabled and using NLS-enabled path name)
Qp0lReaddir() Read directory entry with extensions
QP0FPTOS() Perform miscellaneous file system functions
QP0LCHSG() Change scan signature
Resolve integrated file system path name into QSYS
Qp0lCvtPathToSYSObjName()
Object Name
QP0LFLOP() Perform miscellaneous operations on objects
Qp0lGetAttr() Get system attributes for an object
Qp0lGetPathFromFileID() Get path name of object from its file ID
Qp0lOpen() Open file with NLS-enabled path name
Qp0lProcessSubtree() Process directories or objects within a directory tree

Integrated file system 119

 Table 13. Integrated file system APIs (continued)
Function Description
Qp0lReaddir() Read directory entry with extensions
Qp0lRenameKeep() Rename file or directory, keep new if it exists
Qp0lRenameUnlink() Rename file or directory, unlink new if it exists
QP0LROR() Retrieve object references
QP0LRRO() Retrieve referenced objects
QP0LRTSG() Retrieve scan signature
Qp0lSaveStgFree() Save objects data and free its storage
Qp0lSetAttr() Set system attributes for an object
Qp0lUnlink() Unlink file with NLS-enabled path name
Qp0zPipe() Create interprocess channel with sockets
qsysetegid() Set effective group ID
qsysetgroups() Set supplemental group IDs
qsyseteuid() Set effective user ID
qsysetgid() Set group ID
qsysetregid() Set real and effective group IDs
qsysetreuid() Set real and effective user IDs
qsysetuid() Set user ID
QZipUnzip() Decompress files and directories
QzipZip() Compress files and directories
QZNFNFSO() Perform NFS options
QZNFRTVE() Retrieve NFS export information
read() Read from file
readdir() Read directory entry
readdir_r() Read directory entry (threadsafe)
readdir_r_ts64() Read directory entry (threadsafe with 8-byte pointers)
readlink() Read value of symbolic link
readv() Read from file (vector)
Rename file or directory. Can be defined to
rename() have the semantics of Qp0lRenameKeep() or
Qp0lRenameUnlink().
rewinddir() Reset directory stream
rmdir() Remove directory
select() Check I/O status of multiple file descriptors
stat() Get file information
stat64() Get file information (large file enabled)

120 IBM i: Integrated file system

Table 13. Integrated file system APIs (continued)
Function Description
stat64_time64() Get file information (large file and time64_t enabled)
statvfs() Get file system information
statvfs64() Get file system information (large file enabled)
symlink() Make symbolic link
sysconf() Get system configuration variables
takedescriptor() Take file access from another job
umask() Set authorization mask for job
unlink() Remove link to file
utime() Set file access and modification times
Set file access and modification times (time64_t
utime_time64()
enabled)
write() Write to file
writev() Write to file (vector)

Note: Some of these functions are also used for IBM i sockets.

Table 14. Integrated file system exit programs

Function Description
Integrated File System Scan on Called during close processing such as with the
Close Exit Program close() API. This exit program must be provided by
the user.
Integrated File System Scan on Open Called during open processing such as with the
Exit Program open() API. This exit program must be provided by
the user.
Process a Path Name Called by the Qp0lProcessSubtree() API for each
object in the API's search that meets the caller's
selection criteria. This exit program must be provided
by the user.
Save Storage Free Called by the Qp0lSaveStgFree() API to save an
*STMF object type. This exit program must be provided
by the user.

Related concepts
File systems
A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.
Related reference
Example: Integrated file system C functions
This simple C language program illustrates the use of several integrated file system functions.
Copying data using APIs

Integrated file system 121

 If you want to copy database file members to a stream file in an application, you can use the integrated
file system open(), read(), and write() functions to open a member, read data from it, and write data
to it or another file.
Related information
Application programming interfaces (APIs)

ILE C functions
ILE C provides the standard C functions defined by the American National Standards Institute (ANSI).
These functions can operate through either the data management I/O support or the integrated file
system stream I/O support, depending on what you specify when you create the C program. The compiler
uses the data management I/O unless you tell it differently.
To tell the compiler to use the integrated file system stream I/O, you must specify *IFSIO for the System
interface option (SYSIFCOPT) parameter in the Create ILE C Module (CRTCMOD) or Create Bound C
Program (CRTBNDC) command. When you specify *IFSIO, the integrated file system I/O functions are
bound instead of the data management I/O functions. In effect, the C functions of ILE C use the integrated
file system functions to perform I/O.

Figure 10. ILE C functions use the integrated file system stream I/O functions

For more information about using ILE C functions with integrated file system stream I/O, see the

publication WebSphere® Development Studio: ILE C/C++ Programmer's Guide . For details on each C

function of ILE C, see the publication WebSphere Development Studio: C/C++ Language Reference .

Large file support

The integrated file system APIs are enhanced to allow your applications to store and manipulate very
large files. The integrated file system allows stream file sizes up to approximately 1 TB (1 TB equals
approximately 1 099 511 627 776 bytes) in the "root" (/), QOpenSys, and user-defined file systems.
The integrated file system provides a set of 64-bit UNIX-type APIs and allows an easy mapping of existing
32-bit APIs to 64-bit APIs that are capable of accessing large file sizes and offsets by using eight byte
integer arguments.
The following situations are provided to allow applications to use large file support:
• If the macro label _LARGE_FILE_API is defined at compile time, applications have access to APIs and
data structures that are 64-bit enabled. For example, an application intending to use stat64() API
and stat64 structure will need to define _LARGE_FILE_API at compile time.
• If the macro label _LARGE_FILES is defined by the applications at compile time, existing APIs and data
structures are mapped to their 64-bit versions. For example, if an application defines _LARGE_FILES

122 IBM i: Integrated file system

 at compile time, a call to stat() API is mapped to stat64() API and stat()structure is mapped to
stat64() structure.
The applications that intend to use the large file support can either define _LARGE_FILE_API at compile
time and code directly to the 64-bit APIs, or they can define _LARGE_FILES at compile time. All the
appropriate APIs and data structures are then mapped to the 64-bit version automatically.
Applications that do not intend to use the large file support are not impacted and can continue to use
integrated file system APIs without any changes.
Related information
Integrated file system APIs
stat64()--Get File Information (Large File Enabled) API
stat()--Get File Information API

Path name rules for APIs

When using an API to operate on an integrated file system object, you identify the object by supplying its
path name.
The following is a summary of rules to keep in mind when specifying these path names. The term object in
these rules refers to any directory, file, link, or other object.
• All characters in the path name must be valid characters in the operative coded character set identifier
(CCSID). For the APIs that are not NLS-enabled, the operative CCSID is the CCSID of the job or, if the job
CCSID is 65535, the default CCSID of the job. When using the NLS-enabled APIs, you must specify the
operative CCSID.
• Path names are specified in hierarchical order, beginning with the highest-level directory and ending
with the name of the object to be operated on by the API. A delimiter character separates each
component of the path name. For the APIs that are not NLS-enabled, the delimiter must be the slash (/)
character; for example:
Dir1/Dir2/Dir3/UsrFile
The NLS-enabled APIs require you to specify the delimiter to be used. You may specify any character in
the operative CCSID as a delimiter on the NLS-enabled APIs, including the null character. It is important
to choose a delimiter character that is not used in any of the components of the path name.
• Unlike the integrated file system commands, the components of a path name specified to the APIs
may contain any character that is valid for the file system and that can be represented in the operative
CCSID, with the exception of the null character and the delimiter character.
• An absolute path name begins with a delimiter. Path name resolution starts at the "root" (/) directory;
for example:
/Dir1/Dir2/Dir3/UsrFile
where Dir1 is a subdirectory of the “root” (/) directory.
• A relative path name does not begin with a delimiter. Path name resolution starts at the current
directory; for example:
MyDir/MyFile
where MyDir is a subdirectory of the current directory. Applications that use relative path names must
be very careful to ensure that the current directory is set correctly.
• Object names must be unique within a directory.
• The maximum length of the path name and the maximum length, allowable characters, and format of
each component of the path name can vary for different file systems. Also, the path name will not be
converted to uppercase by the APIs, but the file system may do so before creating or searching for
the specified object. See the individual file system descriptions in the "File systems" topic of the IBM i
Information Center for more information about these considerations.

Integrated file system 123

 • Even though file systems may allow certain special characters in component names, you should avoid
using those characters because they may prevent you from being able to work with objects using
interfaces other than the APIs. The slash (/), backslash (\), asterisk (*), question mark (?), single
quotation mark ('), double quotation mark ("), tilde (~), and colon (:) all have special significance
to certain interfaces. For example, most CL commands treat both the slash (/) and backslash (\) as
delimiters and can not properly interpret a path where component names contain either of those
characters. Similarly, the presence of backslash (\) or colon (:) characters may make some objects
unusable from PC client interfaces.
Related reference
Path name rules for CL commands and displays
When using a CL command or display to operate on an object, you identify the object by supplying its path
name.
Related information
End Journal (QjoEndJournal) API
Start Journal (QjoStartJournal) API

File descriptor
When using ILE C stream I/O functions as defined by the American National Standards Institute (ANSI) to
perform operations on a file, you identify the file through the use of pointers. When using the integrated
file system C functions, you identify the file by specifying a file descriptor. A file descriptor is a positive
integer that must be unique in each job.
The job uses a file descriptor to identify an open file when performing operations on the file. The file
descriptor is represented by the variable fildes in C functions that operate on the integrated file system
and by the variable descriptor in C functions that operate on sockets.
Each file descriptor refers to an open file description, which contains information such as a file offset,
status of the file, and access modes for the file. The same open file description can be referred to by more
than one file descriptor, but a file descriptor can refer to only one open file description.

Figure 11. File descriptor and open file description

If an ILE C stream I/O function is used with the integrated file system, the ILE C runtime support converts
the file pointer to a file descriptor.
When using the "root" (/), QOpenSys, or user-defined file systems, you can pass access to an open
file description from one job to another, thus allowing the job to access the file. You do this by using
the givedescriptor(), takedescriptor(), sendmsg(), or recvmsg() function to pass the file
descriptor between jobs.

124 IBM i: Integrated file system

 Related information
givedescriptor()--Pass Descriptor Access to Another Job API
takedescriptor()--Receive Socket Access from Another Job API
sendmsg()--Send a Message Over a Socket
recvmsg()--Receive a Message Over a Socket
Sockets programming
Sockets APIs

Security
When using the integrated file system APIs, you can restrict access to objects as you can when using data
management interfaces. Be aware, however, that adopting authorities is not supported. An integrated file
system API uses the authority of the user profile under which the job is running.
Each file system may have its own special authority requirements. NFS server and file server jobs have
special considerations. These jobs are generally performing functions on behalf of users who do not
necessarily own the user profile for the job. NFS server requests run under the profile of the user whose
user identification (UID) number was received by the NFS server at the time of the request. Other file
server jobs perform requests for the user that are connected to the server.
Authorities on your system are the equivalent of permissions on UNIX systems. The types of permissions
are read and write (for a file or a directory) and execute (for a file) or search (for a directory). The
permissions are indicated by a set of permission bits, which make up the mode of access of the file
or directory. You can change the permission bits by using the change mode functions chmod() or
fchmod(). You can also use the umask() function to control which file permission bits are set each
time a job creates a file.
Related information
chmod()--Change File Authorizations API
fchmod()--Change File Authorizations by Descriptor API
umask()--Set Authorization Mask for Job API
Integrated file system APIs
Security reference

Socket support
If your application is using the "root" (/), QOpenSys, or user-defined file systems, you can take advantage
of the integrated file system local socket support. A local socket object (object type *SOCKET) allows two
jobs running on the same system to establish a communications connection with each other.
One of the jobs establishes a connection point by using the bind() C language function to create a local
socket object. The other job specifies the name of the local socket object on the connect(), sendto(),
or sendmsg() function.
After the connection is established, the two jobs can send data to and receive data from each other using
the integrated file system functions such as write() and read(). None of the data that is transferred
actually goes through the socket object. The socket object is just a meeting point where the two jobs can
find each other.
When the two jobs are finished communicating, each job uses the close() function to close the socket
connection. The local socket object remains in the system until it is removed using the unlink() function
or the Remove Link (RMVLNK) command.
A local socket object cannot be saved.
Related information
Sockets programming
write()--Write to Descriptor API
read()--Read from Descriptor API

Integrated file system 125

 close()--Close File or Socket Descriptor API
unlink()--Remove Link to File API
Remove link (RMVLNK) command

Naming and international support

The support for the "root" (/) and QOpenSys file systems ensures that the characters in object names
remain constant across encoding schemes used for different national languages and devices.
When an object name is passed to the system, each character of the name is converted to a 16-bit form
in which all characters have a standard coded representation. When the name is used on input by an
integrated file system interface, it is converted to the appropriate coded-form from the code page being
used by the invoker. If the code page to which the name is being converted on output does not contain a
character used in the original name, the results could be an error or information returned with unprintable
characters.
Because characters remain constant across code pages, you should not do an operation on the
assumption that a particular character will change to another particular character when a specific code
page is used. For example, you should not assume the number sign character will change to the pound
sterling character even though they may have the same coded representation in different code pages.
Note that the names of the extended attributes of an object are converted in the same way as the name of
the object, so the same considerations apply.
Related concepts
Name continuity
When you use the "root" (/), QOpenSys, and user-defined file systems, you can take advantage of system
support that ensures characters in object names remain the same.
Overview of automatic name conversion
Some file systems that are not case sensitive, such as “root” (/) and the UDFSs that are created with
CASE(*MONO), support names stored in Unicode Standard 4.0. The system runs an automatic name
conversion to support the additional characters in names.

Data conversion
When you access files through the integrated file system, data in the files may or may not be converted,
depending on the open mode requested when the file is opened.
An open file can be in one of two open modes:
Binary
The data is read from the file and written to the file without conversion. The application is responsible
for handling the data.
Text
The data is read from the file and written to the file, assuming it is in textual form. When the data
is read from the file, it is converted from the coded character set identifier (CCSID) of the file to
the CCSID of the application, job, or system receiving the data. When data is written to the file, it is
converted from the CCSID of the application, job, or system to the CCSID of the file. For true stream
files, any line-formatting characters (such as carriage return, tab, and end-of-file) are just converted
from one CCSID to another.
When reading from record files that are being used as stream files, end-of-line characters (carriage
return and line feed) are appended to the end of the data in each record. When writing to record files:
• End-of-line characters are removed.
• Tab characters are replaced by the appropriate number of blanks to the next tab position.
• Lines are padded with either blanks (for a source physical file member) or nulls (for a data physical
file member) to the end of the record.
On an open request, one of the following can be specified:

126 IBM i: Integrated file system

 Binary, Forced
The data is processed as binary regardless of the actual content of the file. The application is
responsible for knowing how to handle the data.
Text, Forced
The data is assumed to be text. The data is converted from the CCSID of the file to the CCSID of the
application.
A default of Binary, Forced is used for the integrated file system open() function.
Related information
open()--Open File API

Example: Integrated file system C functions

This simple C language program illustrates the use of several integrated file system functions.
The program performs the following operations:
1
Uses the getuid() function to determine the real user ID (uid).
2
Uses the getcwd() function to determine the current directory.
3
Uses the open() function to create a file. It establishes read, write, and execute authority to the file
for the owner (the person who created the file).
4
Uses the write() function to write a byte string to the file. The file descriptor that was provided in
the open operation (3), identifies the file.
5
Uses the close() function to close the file.
6
Uses the mkdir() function to create a new subdirectory in the current directory. The owner is given
read, write, and execute access to the subdirectory.
7
Uses the chdir() function to change the new subdirectory to the current directory.
8
Uses the link() function to create a link to the file that was previously created (3).
9
Uses the open() function to open the file for read only. The link that was created in (8) allows access
to the file.
10
Uses the read() function to read a byte string from the file. The file descriptor that was provided in
the open operation (9) identifies the file.
11
Uses the close() function to close the file.
12
Uses the unlink() function to remove the link to the file.
13
Uses the chdir() function to change the current directory back to the parent directory in which the
new subdirectory was created.
14
Uses the rmdir() function to remove the subdirectory that was previously created (6).
15
Uses the unlink() function to remove the file that was previously created (3).

Integrated file system 127

 Note: This sample program will run correctly on systems where the CCSID of the job in which it is run is
37. The integrated file system APIs must have the object and path names encoded in the job's CCSID;
however, the C compiler stores character constants in CCSID 37. For complete compatibility, translate
character constants, such as object and path names, before passing APIs to the job's CCSID.
Note: By using the code examples, you agree to the terms of the “Code license and disclaimer
information” on page 151.

#include <stdlib.h>
#include <stdio.h>
#include <fcntl.h>
#include <unistd.h>
#include <sys/types.h>

#define BUFFER_SIZE 2048

#define NEW_DIRECTORY "testdir"
#define TEST_FILE "test.file"
#define TEST_DATA "Hello World!"
#define USER_ID "user_id_"
#define PARENT_DIRECTORY ".."

char InitialFile[BUFFER_SIZE];
char LinkName[BUFFER_SIZE];
char InitialDirectory[BUFFER_SIZE] = ".";
char Buffer[32];
int FilDes = -1;
int BytesRead;
int BytesWritten;
uid_t UserID;

void CleanUpOnError(int level)

{
printf("Error encountered, cleaning up.\n");
switch ( level )
{
case 1:
printf("Could not get current working directory.\n");
break;
case 2:
printf("Could not create file %s.\n",TEST_FILE);
break;
case 3:
printf("Could not write to file %s.\n",TEST_FILE);
close(FilDes);
unlink(TEST_FILE);
break;
case 4:
printf("Could not close file %s.\n",TEST_FILE);
close(FilDes);
unlink(TEST_FILE);
break;
case 5:
printf("Could not make directory %s.\n",NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 6:
printf("Could not change to directory %s.\n",NEW_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 7:
printf("Could not create link %s to %s.\n",LinkName,InitialFile);
chdir(PARENT_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 8:
printf("Could not open link %s.\n",LinkName);
unlink(LinkName);
chdir(PARENT_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 9:
printf("Could not read link %s.\n",LinkName);

128 IBM i: Integrated file system

 close(FilDes);
unlink(LinkName);
chdir(PARENT_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 10:
printf("Could not close link %s.\n",LinkName);
close(FilDes);
unlink(LinkName);
chdir(PARENT_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 11:
printf("Could not unlink link %s.\n",LinkName);
unlink(LinkName);
chdir(PARENT_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 12:
printf("Could not change to directory %s.\n",PARENT_DIRECTORY);
chdir(PARENT_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 13:
printf("Could not remove directory %s.\n",NEW_DIRECTORY);
rmdir(NEW_DIRECTORY);
unlink(TEST_FILE);
break;
case 14:
printf("Could not unlink file %s.\n",TEST_FILE);
unlink(TEST_FILE);
break;
default:
break;
}
printf("Program ended with Error.\n"\
"All test files and directories may not have been removed.\n");
}

int main ()
{
1
/* Get and print the real user id with the getuid() function. */
UserID = getuid();
printf("The real user id is %u. \n",UserID);

2
/* Get the current working directory and store it in InitialDirectory. */
if ( NULL == getcwd(InitialDirectory,BUFFER_SIZE) )
{
perror("getcwd Error");
CleanUpOnError(1);
return 0;
}
printf("The current working directory is %s. \n",InitialDirectory);

3
/* Create the file TEST_FILE for writing, if it does not exist.
Give the owner authority to read, write, and execute. */
FilDes = open(TEST_FILE, O_WRONLY | O_CREAT | O_EXCL, S_IRWXU);
if ( -1 == FilDes )
{
perror("open Error");
CleanUpOnError(2);
return 0;
}
printf("Created %s in directory %s.\n",TEST_FILE,InitialDirectory);

4
/* Write TEST_DATA to TEST_FILE via FilDes */
BytesWritten = write(FilDes,TEST_DATA,strlen(TEST_DATA));
if ( -1 == BytesWritten )
{
perror("write Error");
CleanUpOnError(3);
return 0;
}
printf("Wrote %s to file %s.\n",TEST_DATA,TEST_FILE);

Integrated file system 129

 5
/* Close TEST_FILE via FilDes */
if ( -1 == close(FilDes) )
{
perror("close Error");
CleanUpOnError(4);
return 0;
}
FilDes = -1;
printf("File %s closed.\n",TEST_FILE);

6
/* Make a new directory in the current working directory and
grant the owner read, write and execute authority */
if ( -1 == mkdir(NEW_DIRECTORY, S_IRWXU) )
{
perror("mkdir Error");
CleanUpOnError(5);
return 0;
}
printf("Created directory %s in directory %s.\n",NEW_DIRECTORY,InitialDirectory);

7
/* Change the current working directory to the
directory NEW_DIRECTORY just created. */
if ( -1 == chdir(NEW_DIRECTORY) )
{
perror("chdir Error");
CleanUpOnError(6);
return 0;
}
printf("Changed to directory %s/%s.\n",InitialDirectory,NEW_DIRECTORY);

/* Copy PARENT_DIRECTORY to InitialFile and

append "/" and TEST_FILE to InitialFile. */
strcpy(InitialFile,PARENT_DIRECTORY);
strcat(InitialFile,"/");
strcat(InitialFile,TEST_FILE);

/* Copy USER_ID to LinkName then append the

UserID as a string to LinkName. */
strcpy(LinkName, USER_ID);
sprintf(Buffer, "%d\0", (int)UserID);
strcat(LinkName, Buffer);

8
/* Create a link to the InitialFile name with the LinkName. */
if ( -1 == link(InitialFile,LinkName) )
{
perror("link Error");
CleanUpOnError(7);
return 0;
}
printf("Created a link %s to %s.\n",LinkName,InitialFile);

9
/* Open the LinkName file for reading only. */
if ( -1 == (FilDes = open(LinkName,O_RDONLY)) )
{
perror("open Error");
CleanUpOnError(8);
return 0;
}
printf("Opened %s for reading.\n",LinkName);

10
/* Read from the LinkName file, via FilDes, into Buffer. */
BytesRead = read(FilDes,Buffer,sizeof(Buffer));
if ( -1 == BytesRead )
{
perror("read Error");
CleanUpOnError(9);
return 0;
}
printf("Read %s from %s.\n",Buffer,LinkName);
if ( BytesRead != BytesWritten )
{
printf("WARNING: the number of bytes read is "\
"not equal to the number of bytes written.\n");
}

130 IBM i: Integrated file system

 11
/* Close the LinkName file via FilDes. */
if ( -1 == close(FilDes) )
{
perror("close Error");
CleanUpOnError(10);
return 0;
}
FilDes = -1;
printf("Closed %s.\n",LinkName);

12
/* Unlink the LinkName link to InitialFile. */
if ( -1 == unlink(LinkName) )
{
perror("unlink Error");
CleanUpOnError(11);
return 0;
}
printf("%s is unlinked.\n",LinkName);

13
/* Change the current working directory
back to the starting directory. */
if ( -1 == chdir(PARENT_DIRECTORY) )
{
perror("chdir Error");
CleanUpOnError(12);
return 0;
}
printf("changing directory to %s.\n",InitialDirectory);

14
/* Remove the directory NEW_DIRECTORY */
if ( -1 == rmdir(NEW_DIRECTORY) )
{
perror("rmdir Error");
CleanUpOnError(13);
return 0;
}
printf("Removing directory %s.\n",NEW_DIRECTORY);

15
/* Unlink the file TEST_FILE */
if ( -1 == unlink(TEST_FILE) )
{
perror("unlink Error");
CleanUpOnError(14);
return 0;
}
printf("Unlinking file %s.\n",TEST_FILE);

printf("Program completed successfully.\n");

return 0;
}

Working with files and folders using IBM Navigator for i

You can perform these tasks with files and folders.

Creating a folder
To create a folder, follow these steps.

About this task

In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the object that contains the file system or
folder to which you want to add a new folder.

Integrated file system 131

 3. Click on this object to display its contents in the console work space.
4. Right-click on the folder within the file system to which you want to add a new folder in the console
work space and select New Folder.
5. Provide a name for the folder and specify folder attributes in the New Folder dialog.
• For folders created in the "root" (/), QOpenSys, or user-defined file systems, you can specify values
for these fields: Restrict rename and unlink, Audit objects created in folder, and Scan objects
created in folder.
• For folders created in other file systems, you can specify a value for the Audit objects created in
folder field.
6. Click OK.
In System i Navigator
1. In System i Navigator, expand MyConnections > your system > File Systems > Integrated File
System.
2. Right-click the file system or the folder within the file system to which you want to add the new folder
and select New Folder.
3. Provide a name for the folder and specify folder attributes in the New Folder dialog.
• For folders created in the "root" (/), QOpenSys, or user-defined file systems, you can specify values
for these fields: Restrict rename and unlink, Audit objects created in folder, and Scan objects
created in folder.
• For folders created in other file systems, you can specify a value for the Audit objects created in
folder field.
Click OK.
When you create a folder on the IBM i platform, you need to consider whether you want to protect the
new folder (or object) with journal management. You also need to consider whether you want objects
created in this folder to be scanned or not.
Related tasks
Setting whether objects should be scanned or not
You can specify whether objects should be scanned or not in the "root" (/), QOpenSys and user-defined
file systems. Follow these steps to set the scanning options.
Related information
Journal management

Removing a file or folder

To remove a file or folder, follow these steps:

About this task

In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
wish to delete.
3. Click on this folder to display its contents in the console work space.
4. Right-click the file or folder in the console work space and select Delete. A confirmation panel showing
the list of items you selected for deletion is shown. You can clear items you no longer want to delete.
Then click Delete on the confirmation panel.
Note: When a folder is deleted, all the contents of the folder are deleted as well.
In System i Navigator

132 IBM i: Integrated file system

 1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System. Continue to expand until the object that you want to remove is visible.
2. Right-click the file or folder and select Delete. A confirmation panel showing the list of items you
selected for deletion is shown. You can clear items you no longer want to delete. Then click Delete on
the confirmation panel.
Note: When a folder is deleted, all the contents of the folder are deleted as well.

Moving files or folders to another file system

Each file system has its own unique characteristics. However, moving objects to a different file system
might mean losing the advantages of the file system in which the objects are currently stored. You might
want to move objects from one file system to another to take advantage of those characteristics.

About this task

Before moving objects to another file system, you should be familiar with the file systems on the
integrated file system and their characteristics.
You should also consider the following situation:
• Are you using applications that use advantages of the file system that the objects are currently in?
Some file systems support interfaces that are not part of the integrated file system support.
Applications that use these interfaces may no longer be able to access objects that are moved to
another file system. For example, the QDLS and QOPT file systems support the hierarchical file system
(HFS) APIs and commands to work with document and folder objects. You cannot use these interfaces
on objects that are in other file systems.
• What characteristics of the objects are important to you?
Not all characteristics are supported by all file systems. For example, the QSYS.LIB or independent
ASP QSYS.LIB file systems support storing and retrieving only a few extended attributes, whereas the
"root" (/) and QOpenSys file systems support storing and retrieving all extended attributes. Therefore
QSYS.LIB and independent ASP QSYS.LIB are not good candidates for storing objects that have
extended attributes.
Good candidates for moving are the PC files that are stored in QDLS. Most PC applications should be
able to continue working with PC files that are moved from QDLS to other file systems. The "root" (/),
QOpenSys, and QNTC file systems are good choices for storing these PC files.
In System i Navigator, you can move a file or folder to a different file system by dragging the object to the
new location. You can also accomplish this by using the Copy and Paste, or Cut and Paste functions.
To move objects to another file system using CL commands, perform the following steps:

Procedure
1. Save a copy of all objects that you are planning to move.
Having a backup copy allows you to restore the objects to the original file system if you find that
applications cannot access the objects in the file system to which you have moved them.
Note: You cannot save objects from one file system and restore them to another.
2. Create the directories in the file system that you want to move the objects to using the Create
Directory (CRTDIR) command.
You should carefully examine the attributes of the directory the objects are currently in to determine if
you want to duplicate those attributes on the directories you create. For example, the user who creates
the directory is its owner, rather than the user who owned the old directory. You may want to transfer
ownership of the directory after you have created it, if the file system supports setting the owner of a
directory.
3. Move the files to the file system that you have chosen using the Move Object (MOV) command.

Integrated file system 133

 MOV is recommended because it preserves the ownership of the objects, if the file system supports
setting the ownership of objects. You can, however, use the Copy Object (CPY) command to
preserve the ownership of the objects by using the OWNER(*KEEP) parameter. Keep in mind that this
only works for file systems that support setting the owner of an object. Note that when using MOV or
CPY:
• Attributes may not match and may be discarded.
• Extended attributes may be discarded.
• Authorities may not be equivalent and may be discarded.
This means that if you decide to return the object to its original file system, you may not want to just
move or copy it back because of the attributes and authorities that have been discarded. The safest
way to return an object is to restore a saved version of it.
Related concepts
File systems
A file system provides you with the support to access specific segments of storage that are organized as
logical units. These logical units on your system are files, directories, libraries, and objects.
Related reference
File system comparison
These tables summarize the features and limitations of each file system.
Related information
Create Directory (CRTDIR) command
Move Object (MOV) command
Copy Object (COPY) command

Setting permissions
Adding permissions to an object allows you to control the ability of others to manipulate that object. With
permissions, you can allow some users to only view objects, while allowing others to actually edit the
objects.

About this task

To set permissions to a file or folder, follow these steps:
In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
wish to add permissions.
3. Click on this folder to display its contents in the console work space.
4. Right-click the object for which you want to add permissions in the console work space and select
Permissions.
5. Click Add on the Permissions dialog.
6. Select one or more users and groups or enter the name of a user or group in the user or groups name
field in the Add dialog.
7. Click OK. This will add the users or groups to the top of the list.
8. Click the Remove button to remove selected entries from the list.
9. Apply the permissions you want for the user by checking the box by the appropriate check box.
10. Click OK.
In System i Navigator

134 IBM i: Integrated file system

 1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System. Continue to expand until the object for which you want to add permissions is visible.
2. Right-click the object for which for you want to add permissions and select Permissions.
3. Click Add on the Permissions dialog.
4. Select one or more users and groups or enter the name of a user or group in the user or groups name
field in the Add dialog.
5. Click OK. This will add the users or groups to the top of the list.
6. Click the Details button to implement detailed permissions.
7. Apply the permissions you want for the user by checking the box by the appropriate check box.
8. Click OK.

Setting up file text conversion

You can set up automatic text file conversion in IBM Navigator for i. Automatic text file conversion allows
you to use file extensions for file data conversion.

About this task

The integrated file system can convert a data file when it is transferred between a IBM i platform and a
PC. When you access the data file from a PC, it is handled as if it were in ASCII.
To set up file text conversion, follow these steps:

Procedure
1. In IBM Navigator for i, expand File systems > All tasks.
2. Expand Integrated File System and select Properties.
3. Enter the file extension that you want to convert automatically in the File extensions for automatic
text file conversion text box.
4. Click OK.

Results

Sending a file or folder to another system

To send a file or folder to another system, follow these steps.

Procedure
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System. Continue to expand until the object that you want to send is visible.
2. Right-click the file or folder and select Send. The file or folder appears in the Selected Files and
Folders list of the Send Files from dialog.
3. Expand the list of available systems and groups.
4. Select a system and click Add to add the system to the Target systems and groups list. Repeat this
step for all the systems you want to send this file or folder.
5. Click OK to send the file or folder.

Integrated file system 135

Changing options for sending a file or folder
You can define whether to include subfolders and whether to replace an existing file when the system
sends a file or folder to another system. You can also schedule a time for the system to send the file or
folder. To change the options for sending a file, follow these steps.

Procedure
1. Complete the steps for “Sending a file or folder to another system” on page 135.
2. Click the Options tab. The default options are to include subfolders when packaging and sending files
and to replace an existing file with the file being sent.
3. Change these options as required.
4. Click Advanced to set advanced save and restore options.
5. Click OK to save the advanced options.
6. Click OK to send the file or folder, or click Schedule to set a time for sending the file or folder.
7. Select the options for when you want to send the file or folder.
The schedule function gives you the flexibility to do your work at a convenient time.

Creating a file share

A file share is a directory path that IBM i NetServer shares with PC clients on the IBM i network. A file
share can consist of any integrated file system directory on the IBM i platform.

About this task

To create a file share, follow these steps.
In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the folder for which
you want to create a share.
3. Click on this folder to display its contents in the console work space.
4. Right-click the folder containing the file share that you want to create a share in the console work
space and select Sharing.
5. Select Create Share.
6. On the File Share dialog that is shown, specify the attributes for the new file share and click OK.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Expand the folder that has the share defined for it that you want to create a share.
3. Right-click the folder containing the file share that you want to create a share and select Sharing.
4. Select Create Share.
5. On the File Share dialog that is shown, specify the attributes for the new file share and click OK.

136 IBM i: Integrated file system

Changing a file share
A file share is a directory path that IBM i NetServer shares with PC clients on the IBM i network. A file
share can consist of any integrated file system directory on the IBM i platform.

About this task

To change a file share, follow these steps.
In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the folder for which
you want to change a share.
3. Click on this folder to display its contents in the console work space.
4. Right-click the folder containing the file share that you want to change in the console work space and
select Sharing.
5. Select the name of the file share you want to change.
6. On the File Share dialog that is shown, make changes to the attributes of the file share and click OK to
confirm the changes.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Expand the folder that has the share defined for it that you want to change.
3. Right-click the folder containing the file share that you want to change and select Sharing.
4. Select the name of the file share you want to change.
5. On the File Share dialog that is shown, make changes to the attributes of the file share and click OK to
confirm the changes.

Removing a file share

A file share is a directory path that IBM i NetServer shares with PC clients on the System i network. A file
share can consist of any integrated file system directory on the IBM i platform. You can stop sharing an
existing file share by using the System i Navigator.

About this task

To remove a file share, follow these steps:
In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file for which you
want to remove a share.
3. Click on this folder to display its contents in the console work space.
4. Right-click the shared directory that you want to stop sharing in the console work space and select
Sharing > Stop Sharing.
5. On the Stop Sharing window that displays next, click OK.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Expand the file system that contains the file share that you want to stop sharing.

Integrated file system 137

 3. Right-click the shared directory that you want to stop sharing and select Sharing > Stop Sharing.
4. On the Stop sharing window that displays next, click OK.

Creating a new user-defined file system

A user-defined file system (UDFS) is a file system that you create and define the attributes for. UDFSs
reside in auxiliary storage pools (ASPs) and independent ASPs on the system.

About this task

To create a new user-defined file system (UDFS), follow these steps:
In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System > Root > dev.
2. Click the auxiliary storage pool (ASP) that you want to contain the new UDFS.
3. Select New UDFS from Actions list in the console work space.
4. On the New User-Defined File System dialog, specify the UDFS name, description (optional), auditing
values, default file format, default scanning attribute, default disk space allocation, default memory
allocation, and whether the files in the new UDFS will have case-sensitive file names.
Note: Default disk space allocation and default memory allocation are only available on V6R1 or later.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System > Root > dev.
2. Click the auxiliary storage pool (ASP) that you want to contain the new UDFS.
3. Select New UDFS from the File menu.
4. On the New User-Defined File System dialog, specify the UDFS name, description (optional), auditing
values, default file format, default scanning attribute, default disk space allocation, default memory
allocation, and whether the files in the new UDFS will have case-sensitive file names.
Note: Default disk space allocation and default memory allocation are only available on V6R1 or later.

Mounting a user-defined file system

To access or view the data stored in a UDFS, you must mount the UDFS after every IPL.

About this task

When you mount a UDFS, it covers up any file systems, directories, or objects that exist beneath the
mount point in the folder hierarchy. This makes those file systems, directories, or objects inaccessible
until you unmount the UDFS. To ensure that access to all data in the integrated file system is maintained,
mount the UDFS over an empty folder. After the UDFS is mounted, the files within the UDFS will be
accessible from within that folder. Any changes made in the folder will be changes to the UDFS, rather
than to the covered up folder.
Note: A UDFS on an independent ASP cannot be mounted over.
To mount a user-defined file system (UDFS), follow these steps:
In IBM Navigator for i
1. In IBM Navigator for i, under IBM i Management node, expand File Systems > Integrated File
System > Root > dev.
2. Click the auxiliary storage pool (ASP) that contains the UDFS that you want to mount.
3. Right-click the UDFS that you want to mount in the console work space and select Mount.

138 IBM i: Integrated file system

 4. On the Mount UDFS dialog that is shown, specify the path of the directory to mount over, the type of
access (read only, read and write), and whether to allow user and group ID settings. Then click OK.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > Integrated File System > Root > dev.
2. Click the auxiliary storage pool (ASP) that contains the UDFS that you want to mount.
3. Right-click the UDFS that you want to mount in the UDFS Name column of Systems Director Navigator
for i’s right pane.
4. Select Mount.
5. On the Mount UDFS dialog that is shown, specify the path of the directory to mount over, the type of
access (read only, read and write), and whether to allow user and group ID settings. Then click OK.
If you like to drag, you can mount a UDFS by dragging it to a folder within the integrated file system on the
same system. You cannot drop the UDFS on /dev, /dev/QASPxx, /dev/asp_name, another system, or the
desktop.

Unmounting a user-defined file system

When you mount a UDFS, it covers up any file systems, directories, or objects that exist beneath the
mount point in the folder hierarchy. This makes those file systems, directories, or objects inaccessible
until you unmount the UDFS.

About this task

To unmount a user-defined file system (UDFS), follow these steps:
In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System > Root > dev.
2. Click the auxiliary storage pool (ASP) that contains the UDFS that you want to unmount.
3. Right-click the UDFS that you want to unmount in the console work space.
4. Select Unmount. An unmount confirmation panel is displayed showing the UDFSs you selected for
unmounting.
5. Clear any of the UDFSs you no longer want to unmount. Then click Unmount.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System > Root > dev.
2. Click the auxiliary storage pool (ASP) that contains the UDFS that you want to unmount.
3. Right-click the UDFS that you want to unmount in the Name column of System i Navigator's right pane.
4. Select Unmount. An unmount confirmation panel is displayed showing the UDFSs you selected for
unmounting.
5. Clear any of the UDFSs you no longer want to unmount. Then click Unmount.

Working with dynamically mounted file systems

You can use the Dynamic Mount Information function to see which dynamically mounted file systems are
currently mounted, display their properties, and unmount any of them.

About this task

To use this function, follow these steps:
In IBM Navigator for i

Integrated file system 139

 1. In IBM Navigator for i, under theIBM i Management node, expand File Systems > Integrated File
System.
2. Click Integrated File System.
3. Select Dynamic Mount Information from Actions list.
4. The Dynamic Mount Information window opens, listing all the file systems that are currently mounted.
The window displays the name of the mounted file system, the remote system that the file system
is mounted over, and the type of mount. The types of mount supported are user-defined file system
(UDFS), Network File System version 2 (NFSv2), Network File System version 3 (NFSv3), and Network
File System version 4 (NFSv4). You can choose to unmount any of the file systems listed and view the
properties of a particular file system.
Note: NFSv4 is only available starting in V6R1,
• To unmount a file system, select the file system from the list and click Unmount. A Confirm Unmount
window opens. Verify whether the file system displayed is the one you want to unmount. You can
clear items that you no longer want to unmount. Then click Unmount to confirm the operation.
• To view the properties of a file system, select the file system from the list and click Properties. A
Mount Properties window opens.
– For a user-defined file system (UDFS), the Mount Properties window contains a General tab. It
shows the following properties: name, the path where the UDFS is mounted over, type of mount,
time of mount, whether the file system is read only, and whether it allows user and group settings.
– For a Network File System (NFS), the Mount Properties window contains a General tab and an
Advanced tab. The General tab shows the following properties: name, remote server name, the
path where the NFS is mounted over, type of mount, time of mount, whether the file system is
read only, and whether it allows user and group settings. The Advanced tab shows the following
properties: mount type, time out value, read buffer size, write buffer size, retry attempts number,
retransmission attempts number, regular object attribute minimum and maximum time, folder
attribute minimum and maximum time, whether to force refresh of attributes on open, and
whether to allow attribute and name caching.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Right-click Integrated File System.
3. On the pop-up menu, select Dynamic Mount Information.
4. The Dynamic Mount Information window opens, listing all the file systems that are currently mounted.
The window displays the name of the mounted file system, the remote system that the file system
is mounted over, and the type of mount. The types of mount supported are user-defined file system
(UDFS), Network File System version 2 (NFSv2), Network File System version 3 (NFSv3), and Network
File System version 4 (NFSv4). You can choose to unmount any of the file systems listed and view the
properties of a particular file system.
Note: NFSv4 is only available starting in V6R1,
• To unmount a file system, select the file system from the list and click Unmount. A Confirm Unmount
window opens. Verify whether the file system displayed is the one you want to unmount. You can
clear items that you no longer want to unmount. Then click Unmount to confirm the operation.
• To view the properties of a file system, select the file system from the list and click Properties. A
Mount Properties window opens.
– For a user-defined file system (UDFS), the Mount Properties window contains a General tab. It
shows the following properties: name, the path where the UDFS is mounted over, type of mount,
time of mount, whether the file system is read only, and whether it allows user and group settings.
– For a Network File System (NFS), the Mount Properties window contains a General tab and an
Advanced tab. The General tab shows the following properties: name, remote server name, the
path where the NFS is mounted over, type of mount, time of mount, whether the file system is

140 IBM i: Integrated file system

 read only, and whether it allows user and group settings. The Advanced tab shows the following
properties: mount type, time out value, read buffer size, write buffer size, retry attempts number,
retransmission attempts number, regular object attribute minimum and maximum time, folder
attribute minimum and maximum time, whether to force refresh of attributes on open, and
whether to allow attribute and name caching.

Setting whether objects should be scanned or not

You can specify whether objects should be scanned or not in the "root" (/), QOpenSys and user-defined
file systems. Follow these steps to set the scanning options.

About this task

In IBM Navigator for i
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
wish to scan.
3. Click on this folder to display its contents in the console work space.
4. Right-click on the desired file or folder in the console work space and select Properties.
5. Select the Security tab.
6. Select Scan objects with the option that you want.
In System i Navigator
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System. Continue to expand until the object of interest is visible.
2. Right-click the folder or file and select Properties.
3. Select the Security tab.
4. Select Scan objects with the option that you want.

Results
For more information about the options, see the following sections. The descriptions for these options are
for files. Only files may be scanned. With folders and user-defined file systems, you can specify what scan
attribute should be given to files created in that folder or user-defined file system.
• Yes
The object will be scanned according to the rules described in the scan-related exit programs if the
object has been modified or if the scanning software has been updated since the last time the object
was scanned.
• No
The object will not be scanned by the scan-related exit programs.
Note: If the Scan on next access after object has been restored option is selected in the
system values when an object with this attribute is restored, the object will be scanned at least once
after the restore.
• Only when the object has changed
The object will be scanned according to the rules described in the scan-related exit programs only if
the object has been modified since the last time the object was scanned. It will not be scanned if the
scanning software has been updated.
If the Use only when objects have changed attribute to control scan option is not
specified in the system values, this object change only attribute will not be used, and the object
will be scanned after it is modified and when scan software indicates an update.

Integrated file system 141

 Notes:
1. On this tab for files, you can also determine the scan status of an object.
2. If the Scan on next access after object has been restored option is selected in the
system values when an object with this attribute is restored, the object will be scanned at least once
after the restore.

Checking in objects
You can check in a file or all eligible objects within a folder by using the Check In option on the pop-up
menu or the Properties page.

About this task

You can check in objects that meet the following requirements:
• The object type is supported by the Check In Object (CHKIN) command.
• The object is currently checked out.
To check in objects with the pop-up menu, follow these steps.
In IBM Navigator for i
Note: This method is available only on IBM Navigator for i V6R1 or later. For earlier versions, use the
Properties page method.
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
wish to check in.
3. Click on this folder to display its contents in the console work space.
4. Right-click on the file you want to check in or the folder that you want to have all its contents checked
in in the console work space and select Check In.
To check in objects with the Properties page, follow these steps.
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
wish to check in.
3. Click on this folder to display its contents in the console work space.
4. Select the file you want to check in or the folder that you want to have all its contents checked in in the
console workspace.
Note: The function of checking in all the contents of a folder is only available on V6R1 or later.
5. Right-click the selected object and select Properties.
6. In the Properties window that opens, click the Use tab.
7. Check in the file or all objects within the folder.
• If you want to check in a file, click Check In.
• If you want to check in all the objects within a folder, click Check In. A confirmation window opens.
Click Continue to proceed with the check in operation. This operation might take a long time to
complete, depending on the number of objects being checked in.
In System i Navigator
Note: This method is available only on System i Navigator V6R1 or later. For earlier versions, use the
Properties page method.

142 IBM i: Integrated file system

 1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Right-click the file you want to check in or the folder that you want to have all its contents checked in.
3. On the pop-up menu, select Check In.
To check in objects with the Properties page, follow these steps.
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Right-click the file you want to check in or the folder that you want to have all its contents checked in.
Note: The function of checking in all the contents of a folder is only available on V6R1 or later.
3. On the pop-up menu, select Properties.
4. In the Properties window that opens, click the Use tab.
5. Check in the file or all objects within the folder.
• If you want to check in a file, click Check In.
• If you want to check in all the objects within a folder, click Check In. A confirmation window opens.
Click Continue to proceed with the check in operation. This operation might take a long time to
complete, depending on the number of objects being checked in.

Checking out objects

You can check out a file or all eligible objects within a folder by using the Check Out option on the pop-up
menu or the Properties page.

About this task

You can check out objects that meet the following requirements:
• The object type is supported by the Check Out Object (CHKOUT) command.
• The object is currently checked in.
To check out objects with the pop-up menu, follow these steps.
In IBM Navigator for i
Note: This method is available only on IBM Navigator for i V6R1 or later. For earlier versions, use the
Properties page method.
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
wish to check out.
3. Click on this folder to display its contents in the console work space.
4. Right-click on the file you want to check out or the folder that you want to have all its contents checked
out in the console work space and select Check Out.
To check out objects with the Properties page, follow these steps.
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
wish to check out.
3. Click on this folder to display its contents in the console work space.
4. Select the file you want to check out or the folder that you want to have all its contents checked out in
the console workspace.
Note: The function of checking in all the contents of a folder is only available on V6R1 or later.

Integrated file system 143

 5. Right-click the selected object and select Properties.
6. In the Properties window that opens, click the Use tab.
7. Check out the file or all objects within the folder.
• If you want to check out a file, click Check Out.
• If you want to check out all the objects within a folder, click Check Out. A confirmation window
opens. Click Continue to proceed with the check out operation. This operation might take a long time
to complete, depending on the number of objects being checked out.
In System i Navigator
Note: This method is available only on System i Navigator V6R1 or later. For earlier versions, use the
Properties page method.
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Right-click the file you want to check out or the folder that you want to have all its contents checked
out.
3. On the pop-up menu, select Check Out.
To check out objects with the Properties page, follow these steps.
1. In System i Navigator, expand My Connections > your system > File Systems > Integrated File
System.
2. Right-click the file you want to check out or the folder that you want to have all its contents checked
out.
Note: The function of checking out all the contents of a folder is only available on V6R1 or later.
3. On the pop-up menu, select Properties.
4. In the Properties window that opens, click the Use tab.
5. Check out the file or all objects within the folder.
• If you want to check out a file, click Check Out.
• If you want to check out all the objects within a folder, click Check Out. A confirmation window
opens. Click Continue to proceed with the check out operation. This operation might take a long time
to complete, depending on the number of objects being checked out.

Downloading a file or folder

You can download a file or folder from and IBM i system to a local PC.

About this task

To download a file or folder, follow these steps:

Procedure
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the file or folder you
want to download.
3. Click on this folder to display its contents in the console work space.
4. Right-click the file or folder in the console work space and select Download.
5. A confirmation panel showing the list of items you selected for downloading is displayed. Select
Compress selected files if you want selected items get compressed before downloading.
6. Click Download on the confirmation panel.

144 IBM i: Integrated file system

Uploading a file
You can upload a file from a local PC to an IBM i system.

About this task

To upload a file, follow these steps:

Procedure
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate through the file system folders until you locate the folder that contains the folder to which
you wish to upload a file.
3. Click on this folder to display its contents in the console work space.
4. Right-click the folder to which you want to upload a file in the console work space and select Upload.
5. Select a file from your local file system by clicking Browse button. Input CCSID for the file to be
uploaded or leave it to System Default (1252).
6. Click Upload.

Navigate directly to an integrated file system folder

Using IBM Navigator for i, you can easily go directly to any integrated file system folder.

About this task

To go directly to a file, follow these steps:

Procedure
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Click on the Go to Integrated File System folder node.
3. A panel is shown in the work space, where you can enter the path of the folder you wish to work with,
or, you can select from previously accessed folders by choosing the path from the drop-down choices.
4. Click on the Go button once you have entered or selected the folder you wish to work with.
5. Any folder path for which you enter into the path combo box and press Go, will be automatically saved
in the list of previously accessed folders.
6. Click Close to exit the panel.

Save integrated file system folder as a favorite

Using IBM Navigator for i, you can save any integrated file system folder as a favorite, allowing quick
access back to that folder.

About this task

To save an integrated file system folder as a favorite, follow these steps:

Procedure
1. In IBM Navigator for i, under the IBM i Management node, expand File Systems > Integrated File
System.
2. Navigate to any integrated file system folder.

Integrated file system 145

 3. While viewing the contents of the folder in the work space, click on the Actions drop-down in the table
header, and select the Save as Favorite action.
4. A panel is shown in the work space, where you can enter your name for this favorite, and a category, if
desired. Press OK.
5. You can then access this folder directly by expanding the Favorites node in the navigation area, and
clicking on the favorite name.
6. Click directly on the Favorites node in the navigation area to work with favorites, including removing a
favorite, changing the name, or adding it to a specific category.

Transport-independent remote procedure call

Developed by Sun Microsystems, remote procedure call (RPC) easily separates and distributes client
applications from a server mechanism.
RPC includes a standard for data representation, called eXternal Data Representation (XDR), to enable
more than one type of machine to access transmitted data. Transport-independent RPC (TI-RPC) is the
latest version of RPC. It provides a method of separating the underlying protocol that is used at the
network layer, providing a more seamless transition from one protocol to another. The only protocols that
are currently available on the IBM i platform are TCP and UDP.
Developing distributed applications across a network is a seamless task when using RPC. The primary
targets are applications that gravitate more toward distributing the user interface or data retrieval.

Network selection APIs

These APIs provide the means to choose the transport on which an application should run.
These APIs require that *STMF /etc/netconfig file exist on the system. If the netconfig file does not exist in
the /etc directory, the user must copy it from /QIBM/ProdData/OS400/RPC directory. The netconfig file is
always in the /QIBM/ProdData/OS400/RPC directory.

API Description
endnetconfig() Releases the pointer to the records stored in the netconfig file
freenetconfigent() Frees the netconfig structure that is returned from the call to the
getnetconfigent() function
getnetconfig() Returns the pointer to the current record in the netconfig file and
increments its pointer to the next record
getnetconfigent() Returns the pointer to the netconfig structure that corresponds to
the input netid
setnetconfig() Initializes the record pointer to the first entry in the netconfig file.
The setnetconfig() function must be used before the first use of
getnetconfig() function. The setnetconfig() function returns
a unique handle (a pointer to the records stored in the netconfig file)
to be used by the getnetconfig() function.

Related information
API finder

146 IBM i: Integrated file system

Name-to-address translation APIs
These APIs allow an application to obtain the address of a service or a specified host in a transport-
independent manner.

API Description
netdir_free() Frees structures that are allocated by name-to-address translation
APIs
netdir_getbyaddr() Maps addresses into host names and service names
netdir_getbyname() Maps the host name and service name that are specified in the
service parameter to a set of addresses that are consistent with the
transport identified in the netconfig structure
netdir_options() Provides interfaces to transport-specific capabilities such as the
broadcast address and reserved port facilities of TCP and UDP
netdir_sperror() Issues an informational message that states why one of the name-
to-address translation APIs failed
taddr2uaddr() Translates a transport-specific (local) address to a transport-
independent (universal) address
uaddr2taddr() Translates a transport-independent (universal) address to a
transport-specific (local) address (netbuf structure)

Related information
API finder

eXternal Data Representation (XDR) APIs

These APIs allow RPC applications to handle arbitrary data structures, regardless of their different hosts'
byte orders or structure layout conventions.

API Description
xdr_array() A filter primitive that translates between variable-length arrays and
their corresponding external representations. This function is called
to encode or decode each element of the array
xdr_bool() A filter primitive that translates between Booleans (C integers)
and their external representations. When encoding data, this filter
produces values of either 1 or 0.
xdr_bytes() A filter primitive that translates between counted byte arrays and
their external representations. This function treats a subset of
generic arrays in which the size of array elements is known to
be 1 and the external description of each element is built-in. The
length of the byte sequence is explicitly located in an unsigned
integer. The byte sequence is not ended by a null character. The
external representation of the bytes is the same as their internal
representation.
xdr_char() A filter primitive that translates between C-language characters and
their external representation
xdr_double() A filter primitive that translates between C-language double-
precision numbers and their external representations
xdr_double_char() A filter primitive that translates between C-language 2-byte
characters and their external representation

Integrated file system 147

 API Description
xdr_enum() A filter primitive that translates between C-language enumeration
(enum) and its external representation
xdr_free() Recursively frees the object pointed to by the pointer passed in
xdr_float() A filter primitive that translates between C-language floating-point
numbers (normalized single floating-point numbers) and their
external representations
xdr_int() A filter primitive that translates between C-language integers and
their external representation
xdr_long() A filter primitive that translates between C-language long integers
and their external representations
xdr_netobj() A filter primitive that translates between variable-length opaque
data and its external representation
xdr_opaque() A filter primitive that translates between fixed-size opaque data and
its external representation
xdr_pointer() Provides pointer chasing within structures and serializes null
pointers. Can represent recursive data structures, such as binary
trees or linked lists.
xdr_reference() a filter primitive that provides pointer chasing within structures.
This primitive allows the serializing, deserializing, and freeing of
any pointers within one structure that are referenced by another
structure. The xdr_reference() function does not attach special
meaning to a null pointer during serialization, and passing the
address of a null pointer may cause a memory error. Therefore,
the programmer must describe data with a two-sided discriminated
union. One side is used when the pointer is valid; the other side,
when the pointer is null.
xdr_short() A filter primitive that translates between C-language short integers
and their external representation
xdr_string() A filter primitive that translates between C-language strings and their
corresponding external representations
xdr_u_char() A filter primitive that translates between unsigned C-language
characters and their external representations
xdr_u_int() A filter primitive that translates between C-language unsigned
integers and their external representations
xdr_u_long() A filter primitive that translates between C-language unsigned long
integers and their external representations
xdr_u_short() A filter primitive that translates between C-language unsigned short
integers and their external representations
xdr_union() A filter primitive that translates between discriminated C unions and
their corresponding external representations
xdr_vector() A filter primitive that translates between fixed-length arrays and
their corresponding external representations
xdr_void() Has no parameters. It is passed to other RPC functions that require a
parameter, but does not transmit data

148 IBM i: Integrated file system

 API Description
xdr_wrapstring() A primitive that calls the xdr_string(xdr, sp, maxuint)
API, where maxuint is the maximum value of an unsigned integer.
The xdr_wrapstring() is useful because the RPC package
passes a maximum of two XDR functions as parameters, and the
xdr_string() function requires three.

Related information
API finder

Authentication APIs
These APIs provide authentication to the TI-RPC applications.

API Description
auth_destroy() Destroys the authentication information structure that is pointed to
by the auth parameter
authnone_create() Creates and returns a default RPC authentication handle that passes
null authentication information with each remote procedure call.
authsys_create() Creates and returns an RPC authentication handle that contains
authentication information

Related information
API finder

Transport-independent RPC (TI-RPC) APIs

These APIs provide a distributed application development environment by isolating the application from
any specific transport feature. This adds ease-of-use to the transports.
Related information
API finder

TI-RPC simplified APIs

These simplified APIs specify the type of transport to use. Applications using this level do not have to
explicitly create handles.

API Description
rpc_call() Call a remote procedure on the specified system
rpc_reg() Register a procedure with RPC service package

Related information
API finder

TI-RPC top-level APIs

These APIs allow the application to specify the type of transport.

API Description
clnt_call() Call a remote procedure associated with the client
clnt_control() Change information about a client object
clnt_create() Create a generic client handle

Integrated file system 149

 API Description
clnt_destroy() Destroy the client's RPC handle
svc_create() Create a server handle
svc_destroy() Destroy an RPC service transport handle

Related information
API finder

TI-RPC intermediate-level APIs

These APIs are similar to the top-level APIs, but the user applications select the transport specific
information using network selection APIs.

API Description
clnt_tp_create() Create a client handle
svc_tp_create() Create a server handle

Related information
API finder

TI-RPC expert-level APIs

These APIs allow the application to select which transport to use. They also offer an increased level of
control over the details of the CLIENT and SVCXPRT handles. These APIs are similar to the intermediate-
level APIs with an additional control that is provided by using the name-to-address translation APIs.
An additional control that is provided by using the name-to-address translation APIs.

API Description
clnt_tli_create() Create a client handle
rpcb_getaddr() Find the universal address of a service
rpcb_set() Register the server address with the RPCbind
rpcb_unset() Used by the servers to unregister their addresses
svc_reg() Associate program and version with dispatch
svc_tli_create() Create a server handle
svc_unreg() Delete an association set by svc_reg()

Related information
API finder

Other TI-RPC APIs

These APIs allow the various applications to work in coordination with the simplified, top-level,
intermediate-level, and expert-level APIs.

API Description
clnt_freeres() Free data allocated by the RPC or XDR system
clnt_geterr() Get the error structure from the client handle
svc_freeargs() Free data allocated by the RPC or XDR system

150 IBM i: Integrated file system

 API Description
svc_getargs() Decodes the arguments of an RPC request
svc_getrpccaller() Get the network address of the caller
svc_run() Waits for RPC requests to arrive
svc_sendreply() Sends the results of a procedure call to a remote client.
svcerr_decode() Sends information to client for decode error
svcerr_noproc() Sends information to client for procedure number error
svcerr_systemerr() Sends information to client for system error

Related information
API finder

Related information for integrated file system

Product manuals and other information center topic collections contain information that relates to the
Integrated file system topic collection. You can view or print any of the PDF files.

Manuals

• IBM i Network File System Support . (2105 KB) This book describes the Network File System
through a series of real-life applications. Included is information about exporting, mounting, file locking,
and security considerations. From this book, you can learn how to use NFS to construct and develop a
secure network namespace.

• WebSphere Development Studio: ILE C/C++ Language Reference . (4490 KB) This book provides
information needed to design, edit, compile, run, and debug ILE C programs on the IBM i platform.

• APPC Programming . (1497 KB) This book describes the advanced program-to-program
communications (APPC) support for IBM i platforms. It guides in developing application programs that
use APPC and in defining the communications environment for APPC.

• Recovering your system . (8404 KB) This book provides general information about recovery and
availability options for IBM i platforms.

Other information
• Backing up the integrated file system
• Control language
• i5/OS globalization
• Application programming interfaces
• Journal management
• Commitment control
• Security reference

Code license and disclaimer information

IBM grants you a nonexclusive copyright license to use all programming code examples from which you
can generate similar function tailored to your own specific needs.

Integrated file system 151

 SUBJECT TO ANY STATUTORY WARRANTIES WHICH CANNOT BE EXCLUDED, IBM, ITS PROGRAM
DEVELOPERS AND SUPPLIERS MAKE NO WARRANTIES OR CONDITIONS EITHER EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OR CONDITIONS OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT, REGARDING THE PROGRAM OR
TECHNICAL SUPPORT, IF ANY.
UNDER NO CIRCUMSTANCES IS IBM, ITS PROGRAM DEVELOPERS OR SUPPLIERS LIABLE FOR ANY OF
THE FOLLOWING, EVEN IF INFORMED OF THEIR POSSIBILITY:
1. LOSS OF, OR DAMAGE TO, DATA;
2. DIRECT, SPECIAL, INCIDENTAL, OR INDIRECT DAMAGES, OR FOR ANY ECONOMIC CONSEQUENTIAL
DAMAGES; OR
3. LOST PROFITS, BUSINESS, REVENUE, GOODWILL, OR ANTICIPATED SAVINGS.
SOME JURISDICTIONS DO NOT ALLOW THE EXCLUSION OR LIMITATION OF DIRECT, INCIDENTAL, OR
CONSEQUENTIAL DAMAGES, SO SOME OR ALL OF THE ABOVE LIMITATIONS OR EXCLUSIONS MAY NOT
APPLY TO YOU.

152 IBM i: Integrated file system

Notices
This information was developed for products and services offered in the U.S.A.
IBM may not offer the products, services, or features discussed in this document in other countries.
Consult your local IBM representative for information on the products and services currently available in
your area. Any reference to an IBM product, program, or service is not intended to state or imply that
only that IBM product, program, or service may be used. Any functionally equivalent product, program, or
service that does not infringe any IBM intellectual property right may be used instead. However, it is the
user's responsibility to evaluate and verify the operation of any non-IBM product, program, or service.
IBM may have patents or pending patent applications covering subject matter described in this
document. The furnishing of this document does not grant you any license to these patents. You can
send license inquiries, in writing, to:

IBM Director of Licensing

IBM Corporation
North Castle Drive
Armonk, NY 10504-1785
U.S.A.

For license inquiries regarding double-byte (DBCS) information, contact the IBM Intellectual Property
Department in your country or send inquiries, in writing, to:

Intellectual Property Licensing

Legal and Intellectual Property Law
IBM Japan Ltd.
1623-14, Shimotsuruma, Yamato-shi
Kanagawa 242-8502 Japan

The following paragraph does not apply to the United Kingdom or any other country where such
provisions are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION
PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR
IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT,
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of
express or implied warranties in certain transactions, therefore, this statement may not apply to you.
This information could include technical inaccuracies or typographical errors. Changes are periodically
made to the information herein; these changes will be incorporated in new editions of the publication.
IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this
publication at any time without notice.
Any references in this information to non-IBM Web sites are provided for convenience only and do not in
any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of
the materials for this IBM product and use of those Web sites is at your own risk.
IBM may use or distribute any of the information you supply in any way it believes appropriate without
incurring any obligation to you.
Licensees of this program who wish to have information about it for the purpose of enabling: (i) the
exchange of information between independently created programs and other programs (including this
one) and (ii) the mutual use of the information which has been exchanged, should contact:

IBM Corporation
Software Interoperability Coordinator, Department YBWA
3605 Highway 52 N
Rochester, MN 55901
U.S.A.

© Copyright IBM Corp. 1999, 2019 153

 Such information may be available, subject to appropriate terms and conditions, including in some cases,
payment of a fee.
The licensed program described in this document and all licensed material available for it are provided by
IBM under terms of the IBM Customer Agreement, IBM International Program License Agreement or any
equivalent agreement between us.
Any performance data contained herein was determined in a controlled environment. Therefore, the
results obtained in other operating environments may vary significantly. Some measurements may have
been made on development-level systems and there is no guarantee that these measurements will be
the same on generally available systems. Furthermore, some measurements may have been estimated
through extrapolation. Actual results may vary. Users of this document should verify the applicable data
for their specific environment.
Information concerning non-IBM products was obtained from the suppliers of those products, their
published announcements or other publicly available sources. IBM has not tested those products and
cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM
products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of
those products.
All statements regarding IBM's future direction or intent are subject to change or withdrawal without
notice, and represent goals and objectives only.
All IBM prices shown are IBM's suggested retail prices, are current and are subject to change without
notice. Dealer prices may vary.
This information is for planning purposes only. The information herein is subject to change before the
products described become available.
This information contains examples of data and reports used in daily business operations. To illustrate
them as completely as possible, the examples include the names of individuals, companies, brands, and
products. All of these names are fictitious and any similarity to the names and addresses used by an
actual business enterprise is entirely coincidental.
COPYRIGHT LICENSE:
This information contains sample application programs in source language, which illustrate programming
techniques on various operating platforms. You may copy, modify, and distribute these sample programs
in any form without payment to IBM, for the purposes of developing, using, marketing or distributing
application programs conforming to the application programming interface for the operating platform
for which the sample programs are written. These examples have not been thoroughly tested under
all conditions. IBM, therefore, cannot guarantee or imply reliability, serviceability, or function of these
programs. The sample programs are provided "AS IS", without warranty of any kind. IBM shall not be
liable for any damages arising out of your use of the sample programs.
Each copy or any portion of these sample programs or any derivative work, must include a copyright
notice as follows:
© (your company name) (year). Portions of this code are derived from IBM Corp. Sample Programs.

© Copyright IBM Corp. _enter the year or years_.

Programming interface information

This Integrated file system publication documents intended Programming Interfaces that allow the
customer to write programs to obtain the services of IBM i.

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business
Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be
trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at
"Copyright and trademark information" at www.ibm.com/legal/copytrade.shtml.

154 Notices
 Adobe, the Adobe logo, PostScript, and the PostScript logo are either registered trademarks or
trademarks of Adobe Systems Incorporated in the United States, and/or other countries.
Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the
United States, other countries, or both.
UNIX is a registered trademark of The Open Group in the United States and other countries.
Other product and service names might be trademarks of IBM or other companies.

Terms and conditions

Permissions for the use of these publications is granted subject to the following terms and conditions.
Personal Use: You may reproduce these publications for your personal, noncommercial use provided that
all proprietary notices are preserved. You may not distribute, display or make derivative works of these
publications, or any portion thereof, without the express consent of IBM.
Commercial Use: You may reproduce, distribute and display these publications solely within your
enterprise provided that all proprietary notices are preserved. You may not make derivative works of
these publications, or reproduce, distribute or display these publications or any portion thereof outside
your enterprise, without the express consent of IBM.
Except as expressly granted in this permission, no other permissions, licenses or rights are granted, either
express or implied, to the publications or any information, data, software or other intellectual property
contained therein.
IBM reserves the right to withdraw the permissions granted herein whenever, in its discretion, the use
of the publications is detrimental to its interest or, as determined by IBM, the above instructions are not
being properly followed.
You may not download, export or re-export this information except in full compliance with all applicable
laws and regulations, including all United States export laws and regulations.
IBM MAKES NO GUARANTEE ABOUT THE CONTENT OF THESE PUBLICATIONS. THE PUBLICATIONS
ARE PROVIDED "AS-IS" AND WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED,
INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY, NON-INFRINGEMENT,
AND FITNESS FOR A PARTICULAR PURPOSE.

Notices 155
156 IBM i: Integrated file system
IBM®

Product Number: 5770-SS1