Scribd
Upload
140 views5 pages

X Plore Re Feed

The document discusses signs of corruption in an XPLORE installation and provides steps to re-feed or re-index the data to recover from corruption when a backup is not available. There are two types of corruption: hardware faults caused by damage or interruptions, and software bugs. Steps to re-feed include stopping XPLORE, deleting the data and log directories, and restarting the services to re-index all documents. Re-feeding preserves existing configurations while re-installing requires reconfiguring options.

Uploaded by

sorinel01
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
140 views5 pages

X Plore Re Feed

The document discusses signs of corruption in an XPLORE installation and provides steps to re-feed or re-index the data to recover from corruption when a backup is not available. There are two types of corruption: hardware faults caused by damage or interruptions, and software bugs. Steps to re-feed include stopping XPLORE, deleting the data and log directories, and restarting the services to re-index all documents. Re-feeding preserves existing configurations while re-installing requires reconfiguring options.

Uploaded by

sorinel01
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

XPLORE RE-FEED (RE-INDEX) PROCESS

Esmail Fahimi
July 1, 2011
Introduction ________________________________________________________________ 2
Signs of corruptions __________________________________________________________ 2
Types of corruptions ___________________________________________________________ 3
Steps to re-feed (re-index) ____________________________________________________ 3
Difference between re-feed and re-install _________________________________________ 5

Introduction
On rare occasions you may be asked to re-feed (or re-index). The main reason for this is the lack
of proper backup. When there is a backup, the restoration will be the fastest way to recover a
corrupted installation. The re-feed becomes required when there is no backup and there is severe
corruption in the file system in a way that xPlore tools cannot even re-build the index.
Though the shown figures are from Windows installation, this can be equally applied to Linux.

Signs of corruptions
Here is a sign of corruption from the xdb log file. The log files are located under
XPLORE_ROOT\jboss4.3.0\server\DctmServer_PrimaryDsearch\logs.
<event timestamp="2011-05-21 02:20:58,516" level="ERROR" thread="MetricsCollectorService"
logger="com.emc.documentum.core.fulltext.common" timeInMilliSecs="1305958858516">
<message ><![CDATA[Failed to persist metrics information ]]></message>
<throwable><![CDATA[com.emc.documentum.core.fulltext.utils.ThrowableStack: com.xhive.error.XhiveDataCorruptionException:
DATA_CORRUPTION: Library /SystemData/MetricsDB/PrimaryDsearch is possibly corrupted. Error reading page in file 16 offset c000., Original
message: The specified network name is no longer available
at com.emc.documentum.core.fulltext.indexserver.core.metrics.SystemMetricsPurge.deleteFromXdb(SystemMetricsPurge.java:70)
at com.emc.documentum.core.fulltext.indexserver.core.metrics.SystemMetricsPurge.run(SystemMetricsPurge.java:42)
at java.util.TimerThread.mainLoop(Timer.java:512)
at java.util.TimerThread.run(Timer.java:462)
Caused by: com.xhive.error.XhiveDataCorruptionException: DATA_CORRUPTION: Library /SystemData/MetricsDB/PrimaryDsearch is possibly
corrupted. Error reading page in file 16 offset c000., Original message: The specified network name is no longer available

Types of corruptions
There are two fundamental types of corruptions
Corruptions due to hardware faults
o Severe hardware damage
Hardware corruptions can happen by any malfunctioning hardware, not just disk
In this case the hardware may have to be replaced/fixed
This can cause irrecoverable corruptions (both application and data)
System tools like chkdsk or fsck may see these but may not fix
o Interruption (i.e., power failure)
System tools like chkdsk or fsck may be able to fix these
Corruptions due to software bugs
o In some cases such corruptions can be fixed by the supporting tools
o In other cases, recovery is only possible by restoring from backup or a patch
o System tools like chkdsk or fsck will not see these corruptions

Example of a severely corrupted file system on Linux (from /var/logs/messages):


Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096208
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096200
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096216
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096216
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096160
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096160
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096168
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096176
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096168
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096184
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096192
Jun 30 11:18:22 linserv102 kernel: end_request: I/O error, dev sdd, sector 2096184

Steps to re-feed (re-index)


1) Stop xPlore
2) Delete the data directory

3) Go to config directory

4)
5)
6)
7)
8)

Delete the log directory


Delete all files except indexserverconfig.xml
Start xPlore
Start Index Agent
Go to Index Agent Admin URL

9) Check Start new reindexing operation


10) Click Submit

This will start indexing of all indexable documents in your repository.

Difference between re-feed and re-install


It may be easier to re-install than re-feed depending on your configuration.
When you re-feed, the configuration that you have made (assuming any) will remain intact. For
instance, if you have tuned the JVM options, that would still remain the same.
When you re-install, you need to re-tune your JVM option.
For complex installations (i.e.; multiple nodes, multiple instances), it will be easier to re-feed.

You might also like