How can I be sure that my periodic database backups are valid in case they are needed for recovery?
Various factors can cause a database backup that completes successfully within DB2, to be reported as invalid when trying to restore the TPCDB database from it. For example the following message may unexpectedly be displayed when trying to restore the database: SQL2530N
As referenced in several procedures within the TPC infocenter site, periodic database backups are a critical step to protect the data that TPC has collected in the event of unexpected problems. It has however been observed that database backup files reported by DB2 to have been created successfully, are later found to be invalid or corrupt for one reason or another. This is not always noticed until the backup file is needed for recovery purposes. Backup files can become corrupt if the media they are stored on has problems, or in certain cases the backup will complete but the restore operation will expose problems present in the database that existed prior to taking the backup, resulting in a failed database restore operation.
A quick validation step within DB2 upon completing a backup database command is to issue the following DB2 command against the newly created backup file: db2ckbkp -h, supplying the name of the backup file created by the backup db command.
An example of this command and it's output:
C:\bkp>db2ckbkp -h TPCDB.0.DB2.NODE0000.CATN0000.20130418080828.001
MEDIA HEADER REACHED:
Server Database Name -- TPCDB
Server Database Alias -- TPCDB
Client Database Alias -- TPCDB
Timestamp -- 20130418080828
Database Partition Number -- 0
Instance -- DB2
Sequence Number -- 1
Release ID -- D00
Database Seed -- 51BFFDF8
DB Comment's Codepage (Volume) -- 0
DB Comment (Volume) --
DB Comment's Codepage (System) -- 0
DB Comment (System) --
Authentication Value -- 255
Backup Mode -- 0
Includes Logs -- 0
Compression -- 0
Backup Type -- 0
Backup Gran. -- 0
Merged Backup Image -- 0
Status Flags -- 1
System Cats inc -- 1
Catalog Partition Number -- 0
DB Codeset -- UTF-8
DB Territory --
LogID -- 1366297445
LogPath -- C:\DB2\NODE0000\SQL00002\SQLOGDIR\
Backup Buffer Size -- 15601664
Number of Sessions -- 1
Platform -- 5
The proper image file name would be:
 Buffers processed: ########
Image Verification Complete - successful.
If there are any errors detected, the DB2 support team should be engaged via the PMR process for analysis of why the backup file reports errors, and to perform root cause analysis of why the backup job completed successfully yet created a bad backup file.
This quick, extra step can save future frustration and add a level of assurance that your TPC server is protected and that you can more reliably recover from unexpected circumstances.
|Storage Management||Tivoli Storage Productivity Center Standard Edition||AIX, Linux, Windows||3.3.2, 4.1, 4.1.1, 4.2, 4.2.1, 4.2.2|
|Storage Virtualization Software||TotalStorage Productivity Center||Windows Server 2003, Windows Server 2008|
|Storage Management||Tivoli Storage Productivity Center Basic Edition||Windows, AIX, Linux||4.1, 4.1.1, 4.2, 4.2.1, 4.2.2|