This section provides information on the maintenance of a backup management list. A backup management list is maintained on the server that performs the backup.
A backup management list is backed up.
For details of this, refer to swstresback (Resource backup command).
Backup management list backup data size
Check the size of the file that the subdirectory of the following directory contains.
/etc/opt/FJSVswsts/data |
In the cluster environment, check the size of the file that the subdirectory of the following directory contains
/etc/opt/FJSVswsts/<logic node name>/data |
Point
When you reuse a directory used before to save the data, the size of this directory must be capable of supporting repositories which have been allocated a size to cater for high growth. Rule of thumb would suggest that the file system in which these directories exist should be at least three times greater than the size returned in query above.
A backup management list is restored.
For details of this, refer to swstresrst (Resource restore command).
This section describes the maintenance of a replication management list.
To back up a replication management list, back up the repository of the Storage Management Server. For information about backing up a repository, refer to "Saving a database" relevant to the OS of the Storage Management Server.
Point
Operational information for replication management is saved in a repository.
It is recommended that a replication management list be backed up when operation information for replication management is changed with swsrpsetvol (Replication volume information configuration command), or with swsrpdelvol (Replication volume information deletion command).
Perform the following steps to restore a replication management list:
Restore the repository to the Storage Management Server. For information about restoring a repository, refer to "Recovering a database" in the "ETERNUS SF AdvancedCopy Manager Operator's Guide" relevant to the OS of the Storage Management Server.
If the Storage Management Server is Windows
Select "RDB dictionary and DSI recovery (LOAD Mode)" as the recovery method.
If the Storage Management Server is Solaris or Linux
Specify 0 for stgdbrcv -m option (recovery mode) as the recovery method.
On the Storage Server, execute swsrprecoverres (Resource adjustment command) with the -r option specified. For details of the resource adjustment command, refer to "swsrprecoverres (Resource adjustment command)".
Point
When step 1 above is performed, operational information for replication management of a repository is copied to a Storage Server, where it is then available for use.
When this information for replication management in the repository does not need to be changed, and only the operation information for replication management on the Storage Server is incorrect (for some reason), perform only step 2.
This section describes the maintenance of a database used by a Storage Management Server.
Execute the database save command to save a database space.
Execute this command on the Storage Management Server.
Using the database save command
Execute this command on the Storage Management Server.
/opt/swstorage/bin/stgdbdmp -d Backup-Directory [-y]Specify the name of the directory in which to save a database space.
This directory must have been created in advance.
If any files remain in the specified directory, they will be overwritten when the database is saved.
If this option is omitted and if any other files remain in the specified directory, processing ends without saving the database.
=0:Completed successfully
>0:Terminated abnormally
# /opt/swstorage/bin/stgdbdmp -d /storage/dmpdir -y ////// AdvancedCopy Manager database save ////// ================================================================== DB data will be saved. ================================================================== Editing of the work file used for save processing is complete. . Update of the DB space is suppressed. Forced switching of archive log files is complete. ......................... Save processing of DSI data is complete. Archive log files have been completely discarded. Save processing of RDB dictionary data is complete. Suppression of DB space updates has been completely canceled. ================================================================== Save processing of DB data is complete. ================================================================== Database save processing ended normally. #
NOTES
To execute the database save command, log in as a root user.
Use the locale specified at Storage Management Server installation.
If no data is saved by this command, the respective database cannot be recovered with the 9.1.1.3.3 Recovering a database if an I/O error occurs in the database space.
When the Storage Management Server is in cluster operation, you should prepare directories into which data is saved on shared disks.
Processing may end abnormally if another application or command is executing while this command is being processed. In this event, retry processing after the other operation has finished.
Use this save method only when the configuration information is changed due to the addition of a Storage Server, or fetching of new device information. It is recommended to save a database regularly (for example, once a day).
To safeguard against an input-output error in an external file while saved data is being acquired, it is recommended to keep at least two generations of saved data.
Database backup data size
Calculate the size using the following formula:
60 + (number of devices to be managed x 0.001) megabytes |
Refer to the following file, and beginning from the last line, search for any message with "rdb: ERROR: qdgXXXXX-" in its contents (also check the messages displayed at the same time). Then, obtain the value of "qdgXXXXX" to determine the recovery mode.
/var/opt/FJSVswstf/log/RDBSWSTF.log
/var/opt/FJSVswstf/<logical-node-name>/log/RDBSWSTF.log
Note
If multiple errors occur simultaneously, recovery using recovery modes 1 to 5 may not be possible.
To restore a database to its state when it was last saved, re-create the database and execute 9.1.1.3.1 Saving a database in recovery mode 0.
Data updated or inserted after the last time the database was saved cannot be recovered.
swstsrsemtch (Resource adjustment command) with option -x will need to be executed on all Storage Servers on which the backup is performed, after which swsrprecoverres (Resource adjustment command) with option -r will need to be executed on the Storage Servers on which the replication is performed.
Execute the database recovery command for database recovery if an I/O error occurs in a database space.
Execute this command on the Storage Management Server.
/opt/swstorage/bin/stgdbrcv -m mode [-d Backup-Directory]
Specify a recovery mode based on the type of error that occurred in a database space.
For the relationship between types of errors and recovery modes, see "Table 9.4 Relationship between qdg messages and recovery modes.
Specify the name of the directory storing the saved data.
This option is required if 0, 1, 2, or 4 is specified in the -m option.
=0:Completed successfully
>0:Terminated abnormally
# /opt/swstorage/bin/stgdbrcv -m 1 -d /storage/dmpdir ////// AdvancedCopy Manager database recovery ////// ================================================================== DSI recovery (FW mode) will start. ================================================================== Editing of the work file used for recovery is complete. DSI access inhibition is complete. ......................... DSI data recovery is complete. ================================================================== DSI recovery (FW mode) is complete. ================================================================== Database recovery ended normally. #
NOTES
To execute the database recovery command, log in as a root user.
Use the location specified at Storage Management Server installation.
After the database recovery command ends normally, acquire the latest save data using the database save method. Fujitsu recommends using a different save data storage directory than the save data storage directory used by the recovery command.
After the database recovery command ends normally, restart (start after stopping) the AdvancedCopy Manager daemons. For an explanation on restarting the daemons, see "Chapter 2 Starting and Stopping Daemons"
This section describes the maintenance of the authentication feature. Execute the command on the server (ie, on the Storage Server) where the authentication feature is implemented.
On a Storage Management Server, execute the /opt/FJSVswssc/bin/smmkbat command to extract authentication management information. The root user must execute this command.
/opt/FJSVswssc/bin/Smmkbat -f definition-information-output-shell-script-name |
/opt/FJSVswssc/bin/smmkbat -f acldata.sh
The definition information output shell script is extracted as a shell script in which the command example for reflecting definition information is described.
Authentication management list backup data size
Calculate the size with the following formula:
400 + (length of the account name with an access permission assigned + 29 (*1)...) bytes *1 Add the lengths of all accounts with an access permission assigned. |
Execute the definition information output script.
This script must be executed by a root user.
This section describes the maintenance of a registry. Perform this procedure on a Storage Management Server or a Storage Server.
In a command prompt window, back up the following files using the copy command, etc.
/etc/opt/swstorage/swnode.ini
/etc/opt/swstorage/Logical-node-name/swnode.ini
Save it in the /backup directory.
# cp /etc/opt/swstorage/swnode.ini /backup |
Registry backup data size
Check the size of the file to be backed up.
In a command prompt window, use the copy command or another method to restore the following files.
/etc/opt/swstorage/swnode.ini
/etc/opt/swstorage/Logical-node-name/swnode.ini
Restore the following file saved in the /backup directory:
# cp /backup/swnode.ini /etc/opt/swstorage |