ETERNUS SF Storage Cruiser User's Guide 13.2 - Solaris (TM) Operating System / Linux / Microsoft(R) Windows(R) -

Chapter 9 Maintenance of the Administrative Environment

This chapter explains how to maintain databases, and it describes the appropriate action to take for troubleshooting.

9.1 Maintenance of Manage r

9.1.1.1 Collecting data for troubleshootin g

If any problem occurs in a system that uses this software product, collect troubleshooting information according to "E.1 Troubleshooting information".

9.1.1.2 Backup of Manager

9.1.1.2.1 Backup of Solaris Manager

Perform the backup according to the procedure shown below, in which the backup destination directory is described as "backup location". For details about backup in a cluster environment, refer to "9.1.1.2.2 Backup of Solaris Manager (in a cluster environment)".

Stop the Manager.

# /opt/FJSVssmgr/sbin/managerctl stop
Perform backup of essential files.

Create the backup directories.

# mkdir -p backup location/etc/opt/FJSVtrccb/2.0

# mkdir -p backup location/var/opt/FJSVtrccb/2.0/ac

# mkdir -p backup location/var/opt/FJSVtrccb/2.0/db

# mkdir -p backup location/var/opt/FJSVtrccb/2.0/sr

# mkdir -p backup location/var/opt/FJSVssmgr/current

# mkdir -p backup location/var/opt/FJSVssmgr/current/opencimom/logr

# mkdir -p backup location/var/opt/FJSVssmgr/current/vsccompose

Copy the subdirectories and files to the backup location.

# cp -Rp /etc/opt/FJSVtrccb/2.0/* backup location/etc/opt/FJSVtrccb/2.0

# cp -Rp /var/opt/FJSVtrccb/2.0/ac/* backup location/var/opt/FJSVtrccb/2.0/ac

# cp -Rp /var/opt/FJSVtrccb/2.0/db/* backup location/var/opt/FJSVtrccb/2.0/db

# cp -Rp /var/opt/FJSVtrccb/2.0/sr/* backup location/var/opt/FJSVtrccb/2.0/sr

# cp -Rp /var/opt/FJSVssmgr/current/opencimom/logr/* backup location/var/opt/FJSVssmgr/current/opencimom/logr

# cp -Rp /var/opt/FJSVssmgr/current/vsccompose/* backup location/var/opt/FJSVssmgr/current/vsccompose

# cp -p /var/opt/FJSVssmgr/current/systemevent.csv backup location/var/opt/FJSVssmgr/current

If there are no subdirectories or files, backup is unnecessary.
Perform backup of files that are related to the performance management function.

If you are not using the performance management function, this step is unnecessary.

Create the backup directories.

# mkdir -p backup location/etc/opt/FJSVssmgr/current/

# mkdir -p backup location/var/opt/FJSVssmgr/current/perf

Copy the subdirectories and files to the backup location.

# cp -p /etc/opt/FJSVssmgr/current/perf.conf backup location/etc/opt/FJSVssmgr/current/

# cp -Rp /var/opt/FJSVssmgr/current/perf/* backup location/var/opt/FJSVssmgr/current/perf
Perform backup of files for which the maximum memory that can be used is customized.

If the maximum memory that can be used is not customized, backup is unnecessary.

Create the backup directory.

# mkdir -p backup location/opt/FJSVssmgr/bin

Copy the files to the backup location.

# cp -p /opt/FJSVssmgr/bin/manager.ini backup location/opt/FJSVssmgr/bin
Perform backup of other files that can be customized.

If there are files that are not customized, backup is unnecessary. Backup is also unnecessary if these files do not exist.

Create the backup directory.

# mkdir -p backup location/etc/opt/FJSVssmgr/current

Copy the files to the backup location.

# cp -p /etc/opt/FJSVssmgr/current/sanma.conf backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/ssmgr.conf backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/trapop.sh backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/traprm.sh backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/VscCommitScript backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/FJSVssvsc.ini backup location/etc/opt/FJSVssmgr/current
Perform backup of the SNMP trap XML definition file.

If you have not customized the SNMP trap XML definition file, this step is unnecessary.

In the example shown below, SNMP trap XML definition file '1_3_6_1_1588_2_1_1_1.xml' has been customized. Back up each customized file separately, as shown in the example.

Create the backup directory.

# mkdir -p backup location/etc/opt/FJSVssmgr/current/snmpth

Copy the files to the backup location.

# cp -p /etc/opt/FJSVssmgr/current/snmpth/1_3_6_1_1588_2_1_1_1.xml backup location/etc/opt/FJSVssmgr/current/snmpth
Perform backup of the device pooling settings file.

If you have not customized the device pooling settings file, this step is unnecessary.

In the example shown below, the device pooling settings file 'DP_config.xml' has been customized. Back up each customized file separately, as shown in the example.

Create the backup directory.

# mkdir -p backup location/etc/opt/FJSVssmgr/current/devicepolling

Copy the files to the backup location.

# cp -p /etc/opt/FJSVssmgr/current/devicepolling/DP_config.xml backup location/etc/opt/FJSVssmgr/current/devicepolling
Restart the Manager.

# /opt/FJSVssmgr/sbin/managerctl start

9.1.1.2.2 Backup of Solaris Manager (in a cluster environment)

Stop all cluster services that the Manager belongs to.
Mount the shared disk on the primary node.
Perform steps 2, 3, 5, 6, and 7 of "9.1.1.2.1 Backup of Solaris Manager" on the primary node.
Perform step 4 of "9.1.1.2.1 Backup of Solaris Manager" on both nodes.
Unmount the shared disk that was mounted in step 2.
Start all cluster services that the Manager belongs to.

9.1.1.2.3 Linux manager backup

The backup destination directory is referred to as [backup location] in the procedure explained below. Refer to "9.1.1.2.4 Linux manager backup (cluster environment)"for details on backing up in a cluster environment.

Stop the manager

# /opt/FJSVssmgr/sbin/managerctl stop
Perform backup of essential files

Create a backup directory

# mkdir -p backup location/etc/opt/FJSVtrccb/2.0

# mkdir -p backup location/var/opt/FJSVtrccb/2.0/ac

# mkdir -p backup location/var/opt/FJSVtrccb/2.0/db

# mkdir -p backup location/var/opt/FJSVtrccb/2.0/sr

# mkdir -p backup location/var/opt/FJSVssmgr/current

# mkdir -p backup location/var/opt/FJSVssmgr/current/opencimom/logr

# mkdir -p backup location/var/opt/FJSVssmgr/current/vsccompose

Copy the subdirectory and the file to thebackup location

# cp -Rp /etc/opt/FJSVtrccb/2.0/* backup location/etc/opt/FJSVtrccb/2.0

# cp -Rp /var/opt/FJSVtrccb/2.0/ac/* backup location/var/opt/FJSVtrccb/2.0/ac

# cp -Rp /var/opt/FJSVtrccb/2.0/db/* backup location/var/opt/FJSVtrccb/2.0/db

# cp -Rp /var/opt/FJSVtrccb/2.0/sr/* backup location/var/opt/FJSVtrccb/2.0/sr

# cp -Rp /var/opt/FJSVssmgr/current/opencimom/logr/* backup location/var/opt/FJSVssmgr/current/opencimom/logr

# cp -Rp /var/opt/FJSVssmgr/current/vsccompose/* backup location/var/opt/FJSVssmgr/current/vsccompose

# cp -p /var/opt/FJSVssmgr/current/systemevent.csv backup location/var/opt/FJSVssmgr/current

If there are no subdirectories or files, backup is unnecessary.
Back up the files relating to performance management functionality. When performance management functionality is not used, this procedure is unnecessary.

Create the backup directory

# mkdir -p backup location/etc/opt/FJSVssmgr/current/

# mkdir -p backup location/var/opt/FJSVssmgr/current/perf

Copy the subdirectory and the file to the thebackup location

# cp -p /etc/opt/FJSVssmgr/current/perf.conf backup location/etc/opt/FJSVssmgr/current/

# cp -Rp /var/opt/FJSVssmgr/current/perf/* backup location/var/opt/FJSVssmgr/current/perf
Perform a backup of the file created when memory use is customized (that is, when the maximum amount of memory that can be used is defined). If memory usage has not been customized this backup is not necessary as the file will not exist.

Create the backup directory

# mkdir -p backup location/opt/FJSVssmgr/bin

Copy the file to thebackup location

# cp -p /opt/FJSVssmgr/bin/manager.ini backup location/opt/FJSVssmgr/bin
Perform the backup of other files created as a result of customization.The following procedure shows all customization files; however these files are only created when settings are made therefore some may not exist for the backup procedure..

Create the backup directory

# mkdir -p backup location/etc/opt/FJSVssmgr/current

Copy the file to thebackup location.

# cp -p /etc/opt/FJSVssmgr/current/sanma.conf backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/ssmgr.conf backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/trapop.sh backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/traprm.sh backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/VscCommitScript backup location/etc/opt/FJSVssmgr/current

# cp -p /etc/opt/FJSVssmgr/current/FJSVssvsc.ini backup location/etc/opt/FJSVssmgr/current
Perform backup of the SNMP trap XML definition file.

This step is only required if customizations have been performed.

In the example shown below, the existence of the SNMP trap XML definition file '1_3_6_1_1588_2_1_1_1.xml' indicates SNMP has been customized. Back up each file separately, as shown in the example.

Create the backup directory # mkdir -p backup location/etc/opt/FJSVssmgr/current/snmpth

Copy the file to thebackup location # cp -p /etc/opt/FJSVssmgr/current/snmpth/1_3_6_1_1588_2_1_1_1.xml backup location/etc/opt/FJSVssmgr/current/snmpth
Perform backup of the device pooling settings file.

This step is only required if customizations have been performed.

In the example shown below, the existence of the device pooling settings file 'DP_config.xml' indicates that device pooling has been customized. Back up each customized file separately, as shown in the example.

Create the backup directory.

# mkdir -p backup location/etc/opt/FJSVssmgr/current/devicepolling

Copy the files to the backup location.

# mkdir -p backup location/etc/opt/FJSVssmgr/current/devicepolling

Copy the files to the backup location.

# cp -p /etc/opt/FJSVssmgr/current/devicepolling/DP_config.xml backup location/etc/opt/FJSVssmgr/current/devicepolling
Restart the Manager.

# /opt/FJSVssmgr/sbin/managerctl start

9.1.1.2.4 Linux manager backup (cluster environment)

Stop all cluster services that relate to the Manager.
Mount a shared disk on the primary node.
Perform steps 2, 3, 5 and 6 of "9.1.1.2.3 Linux manager backup" on the primary node.
Perform step 4 of "9.1.1.2.3 Linux manager backup" on both nodes.
Un-mount the disk mounted in step 2.
Restart all cluster services that relate to the Manager.

9.1.1.2.5 Backup of Windows Manager

Perform the backup according to the procedure shown below, in which the backup destination directory is described as "backup location".

Stop the Manager.

Go to [Control Panel] > [Administrative Tools] > [Services], select ETERNUS SF Storage Cruiser Manager and stop the service.
Perform backup of essential files.

Using Explorer, create the following backup directories:

- backup location\etc\opt\FJSVtrccb\2.0

- backup location\var\opt\FJSVtrccb\2.0\ac

- backup location\var\opt\FJSVtrccb\2.0\db

- backup location\var\opt\FJSVtrccb\2.0\sr

- backup location\var\opt\FJSVssmgr\current

- backup location\var\opt\FJSVssmgr\current\opencimom\logr

- backup location\var\opt\FJSVssmgr\current\vsccompose

Using Explorer, copy the subdirectories and files in the following directories to the backup location.

- Copy the environment settings directory\etc\opt\FJSVtrccb\2.0 contents to backup location\etc\opt\FJSVtrccb\2.0

- Copy the work directory\var\opt\FJSVtrccb\2.0\ac contents to backup location\var\opt\FJSVtrccb\2.0\ac

- Copy the work directory\var\opt\FJSVtrccb\2.0\db contents to backup location\var\opt\FJSVtrccb\2.0\db

- Copy the work directory\var\opt\FJSVtrccb\2.0\sr contents to backup location\var\opt\FJSVtrccb\2.0\sr

- Copy the work directory\var\opt\FJSVssmgr\current\opencimom\logr contents to backup location\var\opt\FJSVssmgr\current\opencimom\logr

- Copy the work directory\var\opt\FJSVssmgr\current\vsccompose contents to backup location\var\opt\FJSVssmgr\current\vsccompose

Using Explorer, copy the following file to the backup location.

- Copy the work directory\var\opt\FJSVssmgr\current\systemevent.csv contents to backup location\var\opt\FJSVssmgr\current

If there are no subdirectories or files, backup is unnecessary.
Perform backup of files that are related to the performance management function.

If you are not using the performance management function, this step is unnecessary.

Using Explorer, create the following backup directories:

- backup location\etc\opt\FJSVssmgr\current

- backup location\var\opt\FJSVssmgr\current\perf

Using Explorer, copy the following files to the backup location.

- Copy environment settings directory\etc\opt\FJSVssmgr\current\perf.conf to backup location\etc\opt\FJSVssmgr\current

Using Explorer, copy the subdirectories and files in the following directories to the backup location.

- Copy the work directory\var\opt\FJSVssmgr\current\perf contents to backup location\var\opt\FJSVssmgr\current\perf
Perform backup of files that can be customized.

If there are files that are not customized, backup is unnecessary.Backup is also unnecessary if these files do not exist.

Using Explorer, create the following backup directories:

- backup location\opt\FJSVssmgr\bin

- backup location\etc\opt\FJSVssmgr\current

Using Explorer, copy the following files to the backup location.

- Copy program directory\opt\FJSVssmgr\bin\manager.ini to backup location\opt\FJSVssmgr\bin

- Copy environment settings directory\etc\opt\FJSVssmgr\current\sanma.conf to backup location\etc\opt\FJSVssmgr\current

- Copy environment settings directory\etc\opt\FJSVssmgr\current\ssmgr.conf to backup location\etc\opt\FJSVssmgr\current

- Copy environment settings directory\etc\opt\FJSVssmgr\current\trapop.bat to backup location\etc\opt\FJSVssmgr\current

- Copy environment settings directory\etc\opt\FJSVssmgr\current\traprm.bat to backup location\etc\opt\FJSVssmgr\current

- Copy environment settings directory\etc\opt\FJSVssmgr\current\VscCommitScript.bat to backup location\etc\opt\FJSVssmgr\current

- Copy environment settings directory\etc\opt\FJSVssmgr\current\FJSVssvsc.ini to backup location\etc\opt\FJSVssmgr\current
Perform backup of the SNMP trap XML definition file.

If you have not customized the SNMP trap XML definition file, this step is unnecessary.

In the example shown below, SNMP trap XML definition file '1_3_6_1_1588_2_1_1_1.xml'has been customized. Back up each customized file separately, as shown in the example.

Using Explorer, create the following backup directory:

- backup backup location\etc\opt\FJSVssmgr\current\snmpth

Using Explorer, copy the file to the backup location.

- Copy environment settings directory\etc\opt\FJSVssmgr\current\snmpth\1_3_6_1_1588_2_1_1_1.xml to backup location\etc\opt\FJSVssmgr\current\snmpth
Perform backup of the device pooling settings file.

If you have not customized the device pooling settings file, this step is unnecessary.

In the example shown below, the device pooling settings file 'DP_config.xml' has been customized. Back up each customized file separately, as shown in the example.

Using Explorer, create the following backup directory:

- backup location\etc\opt\FJSVssmgr\current\devicepolling

Using Explorer, copy the file to the backup location.

- Copy environment settings directory\etc\opt\FJSVssmgr\current\devicepolling DP_config.xml to backup location\etc\opt\FJSVssmgr\current\devicepolling.
Restart the Manager.

Go to [Control Panel] > [Administrative Tools] > [Services], select ETERNUS SF Storage Cruiser Manager and start the service.

9.1.1.3 Restoring Manager

9.1.1.3.1 Restoring Solaris Manager

Perform restore for the backup performed in "9.1.1.2.1 Backup of Solaris Manager" according to the procedure shown below.

To restore after reinstalling the Manager, first complete the entire installation and setup according to the description in "4.1 [Solaris OS] Manager" in the "ETERNUS SF Storage Cruiser Installation Guide".

To upgrade from a Manager in a previous version, proceed according to the Notes for each procedure.

For details about restoring in a cluster environment, refer to "9.1.1.3.2 Restore of Solaris Manager (in a cluster environment)".

Stop the Manager.

# /opt/FJSVssmgr/sbin/managerctl stop
Restore essential files.

Delete the subdirectories and files in the restore destination directories shown below.

Do not delete the restore destination directories themselves. If these directories do not exist, create them.

# rm -rf /etc/opt/FJSVtrccb/2.0/*

# rm -rf /var/opt/FJSVtrccb/2.0/ac/*

# rm -rf /var/opt/FJSVtrccb/2.0/db/*

# rm -rf /var/opt/FJSVtrccb/2.0/sr/*

# rm -rf /var/opt/FJSVssmgr/current/opencimom/logr/*

# rm -rf /var/opt/FJSVssmgr/current/vsccompose/*

Please delete systemevent.csv the restoration ahead.

# rm -rf /var/opt/FJSVssmgr/current/systemevent.csv

Copy the subdirectories and files from the backup location.

# cp -Rp backup location/etc/opt/FJSVtrccb/2.0/* /etc/opt/FJSVtrccb/2.0

# cp -Rp backup location/var/opt/FJSVtrccb/2.0/ac/* /var/opt/FJSVtrccb/2.0/ac

# cp -Rp backup location/var/opt/FJSVtrccb/2.0/db/* /var/opt/FJSVtrccb/2.0/db

# cp -Rp backup location/var/opt/FJSVtrccb/2.0/sr/* /var/opt/FJSVtrccb/2.0/sr

# cp -Rp backup location/var/opt/FJSVssmgr/current/opencimom/logr/* /var/opt/FJSVssmgr/current/opencimom/logr

# cp -Rp backup location/var/opt/FJSVssmgr/current/vsccompose/* /var/opt/FJSVssmgr/current/vsccompose

# cp -p backup location/var/opt/FJSVssmgr/current/systemevent.csv /var/opt/FJSVssmgr/current

If there are no subdirectories or files, restore is unnecessary.
Restore files that are related to the performance management function.

If you are not using the performance management function, this step is unnecessary.

Delete the subdirectories and files in the restore destination directories shown below.

Do not delete the restore destination directories themselves. If these directories do not exist, create them.

# rm -rf /var/opt/FJSVssmgr/current/perf/*

Copy the subdirectories and files from the backup location.

# cp -p backup location/etc/opt/FJSVssmgr/current/perf.conf /etc/opt/FJSVssmgr/current/

# cp -Rp backup location/var/opt/FJSVssmgr/current/perf/* /var/opt/FJSVssmgr/current/perf
Restore files for which the maximum memory that can be used is customized.

If the maximum memory that can be used is not customized, restore is unnecessary.

Copy the files from the backup location.

# cp -p backup location/opt/FJSVssmgr/bin/manager.ini /opt/FJSVssmgr/bin
Restore the other files that can be customized.

If there are files that are not customized, restore is unnecessary. Restore is also unnecessary if the files do not exist.

Copy the files from the backup location.

# cp -p backup location/opt/FJSVssmgr/bin/manager.ini /opt/FJSVssmgr/bin

# cp -p backup location/etc/opt/FJSVssmgr/current/sanma.conf /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/ssmgr.conf /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/trapop.sh /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/traprm.sh /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/VscCommitScript /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/FJSVssvsc.ini /etc/opt/FJSVssmgr/current
Restore the SNMP trap XML definition file.

If you have not customized the SNMP trap XML definition file, this step is unnecessary.

Copy the SNMP trap XML definition file from the backup location.

# cp -p backup location/etc/opt/FJSVssmgr/current/snmpth/*.xml /etc/opt/FJSVssmgr/current/snmpth

Do not copy the file if you are upgrading from a previous version. Instead, reflect the contents of the customization in the previous version, according to the description in "D.6 SNMP Trap XML Definition File".
Restore the device pooling settings file.

If you have not customized the device pooling settings file, this step is unnecessary.

Copy the device pooling settings file from the backup location.

# cp -p backup location/etc/opt/FJSVssmgr/current/devicepolling/*.xml /etc/opt/FJSVssmgr/current/devicepolling

Do not copy the file if you are upgrading from a previous version. Instead, reflect the contents of the customization in the previous version, according to the description in "D.10 Device Polling Setting File".
Restart the Manager.

# /opt/FJSVssmgr/sbin/managerctl start

9.1.1.3.2 Restore of Solaris Manager (in a cluster environment)

Perform restore for the backup performed in "9.1.1.2.2 Backup of Solaris Manager (in a cluster environment)" according to the procedure shown below.

To restore after reinstalling the Manager, first complete the entire installation and setup according to the description in "Appendix A High-availability admin servers" in the "ETERNUS SF Storage Cruiser Installation Guide".

Stop all cluster services that the Manager belongs to.
Mount the shared disk on the primary node.
Perform steps 2, 3, 5, 6, and 7 of "9.1.1.3.1 Restoring Solaris Manager" on the primary node.
Perform step 4 of "9.1.1.3.1 Restoring Solaris Manager" on both nodes.
Unmount the shared disk that was mounted in step 2.
Start all cluster services that the Manager belongs to.

9.1.1.3.3 Linux manager restore

The following procedure explains the process to restore files that were backed up using the procedure described above in "9.1.1.2.3 Linux manager backup".

If the manager is to be restored by a reinstallation process all of the steps outlined in "4.2 [Linux version] manager" of the "ETERNUS SF Storage Cruiser Installation Guide" must be followed.

If the manager is to be restored by an upgrade process the procedures applicable to the upgrade must be followed.

For details about restoring in a cluster environment, refer to "9.1.1.3.4 Linux manager restore (Cluster environment)".

Execute the following command to stop the Manager.

# /opt/FJSVssmgr/sbin/managerctl stop
Perform the following steps to restore essential files.

Delete the subdirectories and files in the restore destination directories shown below.

Do not delete the restore destination directories themselves. If these directories do not exist, create them.

# rm -rf /etc/opt/FJSVtrccb/2.0/*

# rm -rf /var/opt/FJSVtrccb/2.0/ac/*

# rm -rf /var/opt/FJSVtrccb/2.0/db/*

# rm -rf /var/opt/FJSVtrccb/2.0/sr/*

# rm -rf /var/opt/FJSVssmgr/current/opencimom/logr/*

# rm -rf /var/opt/FJSVssmgr/current/vsccompose/*

Delete the restore destination files.

# rm -rf /var/opt/FJSVssmgr/current/systemevent.csv

Copy the subdirectories and files from the backup location.

# cp -Rp backup location/etc/opt/FJSVtrccb/2.0/* /etc/opt/FJSVtrccb/2.0

# cp -Rp backup location/var/opt/FJSVtrccb/2.0/ac/* /var/opt/FJSVtrccb/2.0/ac

# cp -Rp backup location/var/opt/FJSVtrccb/2.0/db/* /var/opt/FJSVtrccb/2.0/db

# cp -Rp backup location/var/opt/FJSVtrccb/2.0/sr/* /var/opt/FJSVtrccb/2.0/sr

# cp -Rp backup location/var/opt/FJSVssmgr/current/opencimom/logr/* /var/opt/FJSVssmgr/current/opencimom/logr

# cp -Rp backup location/var/opt/FJSVssmgr/current/vsccompose/* /var/opt/FJSVssmgr/current/vsccompose

# cp -p backup location/var/opt/FJSVssmgr/current/systemevent.csv /var/opt/FJSVssmgr/current

Only files and subdirectories created during the backup procedure will be available to restore.
Restore files that are related to the performance management function.

This step is only required if the performance management function file was backed up.

Delete the subdirectories and files in the restore destination directories shown below.

Do not delete the restore destination directories themselves. If these directories do not exist, create them.

Copy the subdirectories and files from the backup location.

# cp -p backup location/etc/opt/FJSVssmgr/current/perf.conf /etc/opt/FJSVssmgr/current/

# cp -Rp backup location/var/opt/FJSVssmgr/current/perf/* /var/opt/FJSVssmgr/current/perf
Restore files for which the maximum memory that can be used is customized.

This step is only required if the file was backed up.

Copy the files from the backup location.

# cp -p backup location/opt/FJSVssmgr/bin/manager.ini /opt/FJSVssmgr/bin
Restore the other files that can be customized.

This step is only required if customization had taken place and the relative file was backed up.

Copy the files from the backup location.

# cp -p backup location/opt/FJSVssmgr/bin/manager.ini /opt/FJSVssmgr/bin

# cp -p backup location/etc/opt/FJSVssmgr/current/sanma.conf /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/ssmgr.conf /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/trapop.sh /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/traprm.sh /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/VscCommitScript /etc/opt/FJSVssmgr/current

# cp -p backup location/etc/opt/FJSVssmgr/current/FJSVssvsc.ini /etc/opt/FJSVssmgr/current
Restore the SNMP trap XML definition file.

This step is only required if the definition file was backed up.

Copy the SNMP trap XML definition file from the backup location.

# cp -p backup location/etc/opt/FJSVssmgr/current/snmpth/*.xml /etc/opt/FJSVssmgr/current/snmpth

Do not copy the file if you are upgrading from a previous version, refer to "D.6 SNMP Trap XML Definition File" in the manual for that version.
Restore the device pooling settings file.

This step is only necessary if the file was backed up.

Copy the device pooling settings file from the backup location.

# cp -p backup location/etc/opt/FJSVssmgr/current/devicepolling/*.xml /etc/opt/FJSVssmgr/current/devicepolling

Do not copy the file if you are upgrading from a previous version, refer to "D.6 SNMP Trap XML Definition File" in the manual for that version.
Restart the Manager.

# /opt/FJSVssmgr/sbin/managerctl start

9.1.1.3.4 Linux manager restore (Cluster environment)

The following procedure explains the process to restore files that were backed up using the procedure described above in "9.1.1.2.4 Linux manager backup (Cluster environment)".

If the manager is to be restored by a reinstallation process all of the steps outlined in "4.2 [Linux version] manager" of the ETERNUS SF Storage Cruiser Installation Guide" must be followed.

Stop all cluster services that relate to the Manager.
Mount a shared disk on the primary node.
Perform steps 2, 3, 5, 6, 7, and 8 of "9.1.1.3.3 Linux Manager Restore" on the primary node.
Perform step 4 of "9.1.1.3.3 Linux Manager Restore" on both nodes.
Un-mount the shared disk that was mounted in step 2.
Start all cluster services that relate to the Manager.

9.1.1.3.5 Restoring Windows Manager

Perform restore for the backup performed in "9.1.1.2. 5 Backup of Windows Manager" according to the procedure shown below.

To restore after reinstalling the Manager, first complete the entire installation and setup according to the description in "4.2 [Windows] Manager" in the "ETERNUS SF Storage Cruiser Installation Guide".

To upgrade from a Manager in a previous version, proceed according to the Notes for each procedure.

Stop the Manager.

Go to [Control Panel] > [Administrative Tools] > [Services], select ETERNUS SF Storage Cruiser Manager and stop the service.
Restore essential files.

Using Explorer, delete the subdirectories and files in the restore destination directories shown below.

Do not delete the restore destination directories themselves. If these directories do not exist, create them.

- environment settings directory\etc\opt\FJSVtrccb\2.0

- work directory\var\opt\FJSVtrccb\2.0\ac

- work directory\var\opt\FJSVtrccb\2.0\db

- work directory\var\opt\FJSVtrccb\2.0\sr

- work directory\var\opt\FJSVssmgr\current

- work directory\var\opt\FJSVssmgr\current\opencimom\logr

- work directory\var\opt\FJSVssmgr\current\vsccompose

Using Explorer, delete the following file.

- work directory\var\opt\FJSVssmgr\current\systemevent.csv

Using Explorer, copy the subdirectories and files from the backup location.

- Copy the backup location\etc\opt\FJSVtrccb\2.0 contents to environment settings directory\etc\opt\FJSVtrccb\2.0

- Copy the backup location\var\opt\FJSVtrccb\2.0\ac contents to work directory\var\opt\FJSVtrccb\2.0\ac

- Copy the backup location\var\opt\FJSVtrccb\2.0\db contents to work directory\var\opt\FJSVtrccb\2.0\db

- Copy the backup location\var\opt\FJSVtrccb\2.0\sr contents to work directory\var\opt\FJSVtrccb\2.0\sr

- Copy the backup location\var\opt\FJSVssmgr\current\opencimom\logr contents to work directory\var\opt\FJSVssmgr\current\opencimom\logr

- Copy the backup location\var\opt\FJSVssmgr\current\vsccompose contents to work directory\var\opt\FJSVssmgr\current\vsccompose

- Copy the backup location\var\opt\FJSVssmgr\current\systemevent.csv to work directory\var\opt\FJSVssmgr\current

If there are no subdirectories or files, restore is unnecessary.
Restore files that are related to the performance management function.

If you are not using the performance management function, this step is unnecessary.

Using Explorer, delete the subdirectories and files in the restore destination directories shown below.

Do not delete the restore destination directories themselves. If these directories do not exist, create them.

- work directory\var\opt\FJSVssmgr\current\perf

Using Explorer, copy the following files from the backup location.

- Copy backup location\etc\opt\FJSVssmgr\current\perf.conf to environment settings directory\etc\opt\FJSVssmgr\current

Using Explorer, copy the subdirectories and files in the following directories from the backup location.

- Copy the backup location\var\opt\FJSVssmgr\current\perf contents to work directory\var\opt\FJSVssmgr\current\perf
Restore files that can be customized.

If there are files that are not customized, restore is unnecessary. Restore is also unnecessary if the files do not exist.

Using Explorer, copy the following files from the backup location.

- Copy backup location\opt\FJSVssmgr\bin\manager.ini to program directory\opt\FJSVssmgr\bin

- Copy backup location\etc\opt\FJSVssmgr\current\sanma.conf to environment settings directory\etc\opt\FJSVssmgr\current

- Copy backup location\etc\opt\FJSVssmgr\current\ssmgr.conf to environment settings directory\etc\opt\FJSVssmgr\current

- Copy backup location\etc\opt\FJSVssmgr\current\trapop.bat to environment settings directory\etc\opt\FJSVssmgr\current

- Copy backup location\etc\opt\FJSVssmgr\current\traprm.bat to environment settings directory\etc\opt\FJSVssmgr\current

- Copy backup location\etc\opt\FJSVssmgr\current\VscCommitScript.bat to environment settings directory\etc\opt\FJSVssmgr\current

- Copy backup location\etc\opt\FJSVssmgr\current\FJSVssvsc.ini to environment settings directory\etc\opt\FJSVssmgr\current
Restore the SNMP trap XML definition file.

If you have not customized the SNMP trap XML definition file, this step is unnecessary.

Using Explorer, copy the SNMP trap XML definition file from the backup location.

- Copy backup location\etc\opt\FJSVssmgr\current\snmpth\*.xml to environment settings directory\etc\opt\FJSVssmgr\current\snmpth

Do not copy the file if you are upgrading from a previous version. Instead, reflect the contents of the customization in the previous version, according to the description in "D.6 SNMP Trap XML Definition File".
Restore the device pooling settings file.

If you have not customized the device pooling settings file, this step is unnecessary.

Using Explorer, copy the device pooling settings file from the backup location.

- Copy environment settings directory\etc\opt\FJSVssmgr\current\devicepolling DP_config.xml to backup location\etc\opt\FJSVssmgr\current\devicepolling

Do not copy the file if you are upgrading from a previous version. Instead, reflect the contents of the customization in the previous version, according to the description in "D.10 Device Polling Setting File".
Restart the Manager.

Go to [Control Panel] > [Administrative Tools] > [Services], select ETERNUS SF Storage Cruiser Manager and start the service.

9.1.2 Changing the operating environmen t

9.1.2.1 Changing the IP address of the administrative serve r

Follow the procedure given below to change the IP address of the administrative server.

Change the system IP address on the administrative server.
Reboot the administrative server.
Start the ESC Client, connect to the new IP address of the administrative server. When it connected to the administrative server successfully, delete the old IP address of administrative server when start the ESC client next time by click <delete> button in the login dialog.

If the administrative server is cluster environment, it is necessary to reconstruct the cluster.

Please refer to "Appendix A High-availability admin servers" in ETERNUS SF Storage Cruiser Install Guide for detail information.

For the device which is under fault management automatically by SNMP trap, it is necessary to change the SNMP Trap transmission place IP address setting. To change SNMP trap transmission setting, refer to "SNMP Trap transmission place IP address change command" in "Appendix C Command References", and execute the command.

9.1.2.2 Changing the IP address of a server node (host)

Follow the procedure given below to change the IP address of a server node.

On the GUI screen, delete the device icon of the server node (without changing the access path).
Change the system IP address on the server node.
Stop the server node Agent on the server node.
Execute the agent information change command on the server node to change the start IP of server node Agent.

For the command that change to Agent start IP, refer to "Agent information change command" in "Appendix C Command References" for the detail information.
Restart the server node Agent on the server node.
On the GUI screen, perform "Server registration" with the new IP address.

9.1.2.3 Changing the server name of the administrative serve r

No special work is required.

9.1.2.4 Changing the name of a server node (host )

To change the name of a server node, use "Delete" (without changing the access path) to delete the device icon registered with the old name. Next, change the name of the server node, and then use "Register server" to register the server node with the new name.

For ESC Agent and SSC Agent, follow the procedure given below to change a server node name.

On the GUI screen, delete the device icon of the server node (without changing the access path). On the GUI screen, delete (no access path change) a device icon of the server node that changes the IP address.
Change the system IP address on the server node.
Stop the server node Agent on the server node.
Execute the agent information change command on the server node to change the start IP of server node Agent.

For the command that change to Agent start IP, refer to "Agent information change command" in "Appendix C Command References" for the detail information.
Restart the server node Agent on the server node.
On the GUI screen, perform "Server registration" on the server node.

Information of an associated application element created in the Correlation window is not inherited when the server node name is changed. Register the application element again.

9.1.2.5 Changing the IP address or the SNMP community name of a supported devic e

To change the IP address of a device with support level A or B registered with this software product or to change the SNMP community name, follow the procedure below:
When the IP address of a device with support level A or B registered with this software product is changed, this software product can automatically change the IP address part in the properties of the same device.

+To change the IP address

Referring to "5.2.1 Device search" use "2. Detecting a device by specifying an IP address " to perform device detection by specifying an IP address. Based on this operation, the product detects that the IP address of a registered device has been renewed and automatically changes the IP address information of the target device on this software product.
For a device with support level A or B, excluding the server node, IP address information for multiple devices in the subnet can be automatically changed by performing "1. Detecting for devices in the same subnet as the administrative server automatically" in "5.2.1 Device search".

Management Software call information is not changed automatically. From "Properties" of each device icon, rewrite information manually.

When the SNMP community name of the device has been changed to something other than "public" and its IP address is changed, set a new IP address for the SNMP community name before detecting any device using a new IP address. For the setting, see "6. SNMP_COMMUNITY_NAME_FOR_IP" and "7. SNMP_COMMUNITY_NAME_FOR_AUTODISCOVER" in "D.2 sanma.conf Parameter", and reflect the contents of the setting file on this software product.

+To change the SNMP community name

Set the community name of each target device.

For information on the community name of each target device, see "Chapter 4 Setup of the Operating Environment".
Set up the admin server.

Set the "SNMP_COMMUNITY_NAME_FOR_IP" to the setting file by referring to "D.2 sanma.conf Parameter", and reflect the contents of the setting file on this software product.

9.1.2.6 Changing the operation management of the cluster system (node name takeover )

In the environment of a Cluster System Operation, Node name inheritance may be used for Inherit network setting. Node name takeover affects the Name of a Server node.

In changing the Name of a Server node by setting Node name inheritance, refer to "9.1.2.4 Changing the name of a server node (host)".

9.1.2.7 Changing the port number

ESC uses the following four ports:.

Port	Description
nfport 23456/tcp	Used for communication between administrative server and client
sscruisera 4917/tcp	Used for communication between administrative server and managed server node
ssvscme 7420/tcp ssvscme 7420/udp	Used for communication between the administrative server and virtual switches when using Virtual Storage Conductor

These port numbers are managed as internal information by the site domain server, domain admin servers, managed server node files (*1), and admin clients. When not setting the port numbers in the files (*1), the numbers above will be used by default.

*1: The files for managed server nodes are as follows:

[Solaris]
/etc/inet/services

[Windows]
Windows_system_directory\system32\drivers\etc\services

[Linux]
/etc/services

[HP-UX]
/etc/services

Change the port number as follows.

If you change the port number, set the same number for the other server nodes in ESC that used the previous number.

<Example>
If nfport of the administrative server is set as 23456, set 23456 for nfport of the client.

For further details about the ports, refer to "3.1.1 Network requirements".

++How to change the port number of the administrative server and managed server nodes

You can change the port numbers by editing each file as follows:

++File editing

[Solaris]
Edit the /etc/inet/services file using the vi(1) command.

[Windows]
Edit the Windows_system_directory\system32\drivers\etc\services file using an editor application.

[Linux]
Edit the /etc/services file using the vi(1) command.

[HP-UX]

Edit the /etc/services file using the vi(1) command.

++Procedure

Change the file

Administrative server

[Solaris, Linux]
Change the lines under /etc/inet/services.

# service names    port nos. /protocol names

  nfport           23456/tcp
  sscruisera       4917/tcp
  ssvscme           7420/tcp
  ssvscme           7420/udp

[Windows]
Change the lines under Windows_system_directory\system32\drivers\etc\services.

# service names    port nos. /protocol names

  nfport           23456/tcp
  sscruisera       4917/tcp
  ssvscme           7420/tcp
  ssvscme           7420/udp

For managed server nodes

[Solaris]
Change the lines under /etc/inet/services.

# service names    port nos. /protocol names

  sscruisera     4917/tcp

[Windows]
Change the lines under Windows_system_directory\system32\drivers\etc\services.

# Service names    port nos. /Protocol names

  sscruisera     4917/tcp

[Linux]
Change the lines under /etc/services.

# service names    port nos. /protocol names

  sscruisera     4917/tcp

[HP-UX]
Change the lines under /etc/services.

# service names    port nos. /protocol names

  sscruisera     4917/tcp

Restart all the server nodes where the port numbers have been changed.

When changing the port numbers used by Virtual Storage Conductor, refer to "Chapter 3 Building an Environment" of the "ETERNUS SF Storage Cruiser User's Guide for Virtual Storage Conductor".

++Change the port number of the admin client

Change the port number with which the admin client is connected to the site admin server.

Start ESC by selecting [start]-[All Programs]-[ETERNUS SF Storage Cruiser] of the admin client or double-clicking the icon on the desktop.
The login window will appear.
Click <Port>.
The "Port number" window will appear.
Set the port number changed, and click <OK>.
You are returned to the login window.
Click <Login> on the login window, then connect to the admin server.

The port number becomes enabled as setting information of the site admin server when the network connection to the administrative server succeeds.

9.1.2.8 Changing the configuration of supported devices

After hardware product is added to or removed from the supported devices or after information is added to or deleted from the configuration settings (for a disk array unit, change in a RAID Group, Logical Volume, etc.) for the supported devices or such information is modified, the information of the target device must manually be updated in the following windows:

Main view

Execute [View]-[Refresh] to read the latest conditions of SAN from all the devices. The target device information is updated and displayed.
Relationship management window

When the Relationship management window is being displayed, execute [View]-[Refresh F5] to read the latest condition of the target device. The target device information is updated and displayed. To display the Relationship management window, see "6.2.1 Correlation window".
Performance management window

Select the target device in the tree of the Performance management window, and then select [Device]-[Create Device Configuration File] from the menu to read the latest condition of the target device. The target device information is updated and displayed. For details, see "7.2.11 Updating configuration information".

9.1.2.9 Changing the password of supported devices

Follow the procedure given below to change the password of a device which is been registed to this product.

Objected Device	Conduction
SN200 (Brocade) fiber switch and PRIMEGY BX600 fiber channel switch blade	Refer to "2. User name and password for Fibre Channel switch control" in "4.2.1.1 Setting" to change the registered information.

Except for the devices described above, there is no necessary to do any deal with this product when change the device's password.

9.2 [Solaris OS Version] Maintenance of Agen t

9.2.1 Definition file s

When an access path is set from this software product, Agent automatically modifies the following files that contain device definitions on a server node and Fibre Channel environment definitions:

Path setting

/kernel/drv/sd.conf shared to different vendors' HBAs
Storage affinity setting

If the HBA is the PW008FC2A, PW008FC2-G, PW008FC3, GP7B8FC1A, GP7B8FC1-G, GP7B8FC1, SE0X7F11F, SE0X7F12F: /kernel/drv/fjpfca.conf

If the HBA is the Emulex LP9000/9002S/9002L/9802/10000: /kernel/drv/lpfc.conf or lpfs.conf

In the event that a server node operation becomes abnormal after the setting of an access path from this software product, a backup copy of the setting files that was created beforehand can be used for recovery.

The backup file is stored in /var/opt/FJSVssage/backup. The file name is the old file name suffixed with a serial number starting from 1.

9.2.2 Action to take for troubleshootin g

If an Agent error (message ID = 3400) is output to /var/adm/messages during the setting or deletion of an access path, a definition file may not have been updated successfully. In this event, stop Agent, and take the following action:

As a user having root authority, enter the following command to stop Agent:

# /opt/FJSVssage/bin/pstorageagt
Execute the ps command as follows to confirm that Agent is stopped:

# /bin/ps -ef|grep FJSVssage
As necessary, save the following files to any directory:
- /kernel/drv/sd.conf
- /kernel/drv/fjpfca.conf
- /kernel/drv/lpfc.conf
- /kernel/drv/lpfs.conf
Restore the following definition files:
Restore the original file from the backup file that is stored in the backup file storage area (/var/opt//FJSVssage/backup).
Restart the server node.

Solaris 8, 9, 10 OS

# touch /reconfigure

# /usr/sbin/shutdown -y -i6 -g0

After the server node is restarted, enter the following command:

# disks -C

Other examples of major errors and the appropriate action to be taken are explained below.

Agent detects an environment error when a new HBA is added (except in cases where the HBA is a replacement).
The server node may not have been started in re-configuration mode after the addition of the HBA. In this event, restart the server node in re-configuration mode.
Agent detects an environment error when a new HBA driver is installed.
The server node may not have been started in re-configuration mode after installation of the driver. In this event, restart the server node in re-configuration mode.

9.2.3 Collecting data for troubleshootin g

If any problem occurs in Agent, collect troubleshooting information according to "E.1 Troubleshooting information".

9.2.4 Changing the administrative environment

9.2.4.1 Changing the IP address of a server node (host)

For detail, please refer to "9.1.2.2 Changing the IP address of a server node (host)".

9.2.4.2 Changing the port number

For detail, please refer to "9.1.2.7 Changing the port number".

9.3 [Windows Version] Maintenance of Agen t

9.3.1 Action to take for troubleshootin g

Major errors related to Agent running on a Windows 2000/2003 server node and the corrective action to be taken are explained below. In the Event viewer or the GUI window, confirm the occurrence of the errors described below.

Manager cannot detect Agent.
1. If all access paths under multipath control have been deleted by mistake, follow the recovery procedure given below.
  - Restart the Windows server node.
    After an access path is deleted, multipath control no longer controls the path, and the related device is not recognized.
  - "Refresh" the GUI window. The monitoring status of [HOST Property] in the Server node view of the GUI window then becomes normal.
  - Set the access path that was deleted by mistake.
2. Communication with Agent is disabled when an access path is connected.
  Depending on the particular system environment, Agent may not be able to communicate when an access path is set in the Windows 2000 environment. This is because the simultaneous operation of the access path connection and the plug&play function of Windows 2000 places a temporary heavy load on the system. In such cases, select [Refresh] in the GUI window.
  If communication remains disabled, the problem described in Item 3 may have occurred. See Item 3 below.
3. Too many devices are connected to the Windows server node.
  If too many devices are recognized by the Windows server node, Agent cannot perform processing within the response time, leading to a communication failure. In this event, check the connection configuration, and reduce the number of devices recognized by the Windows server node.
  Access paths can be set for approximately 100 devices per Windows NT/2000 server node, but the number of devices varies depending on the server node environment.
4. Anti-virus software "Real Time Scan" is operated
  If an anti-virus program performs a real-time scan on a Windows server node, it is possible that the server cannot be detected or that "communication disabled" is displayed. It is because file access processing by agent takes more time than usual and it leads to timeout. Change the setting value of SNMP_HCM_TIMEOUT of sanma.conf to higher value on the side of manager to be able to detect and display the applicable server.
  See, "D.2 sanma.conf Parameter" for path of sanma.conf and the procedure for reflecting it.
The GUI window contents differ from the physical configuration.
1. An access path from an Emulex HBA cannot be drawn.
  The Topology value in a driver parameter in the registry may be incorrect.
  This caused by values in the registry being reset to default values, which occurs at either of the times: when a driver is re-installed; and when a driver is automatically re-installed because of the installation a new HBA having a different model number from that of a HBA already installed in a Windows 2000/2003 machine (e.g., LP9802 and LP1050).
  Using the registry editor, check the driver parameters of the HBA, correct any incorrect value that you find, and then select [Refresh].
  For information about driver parameters, see "4.1.2 Windows server node (host)".