IBM Support

IBM TSM FastBack: Metadata Backup Chain Can Be Corrupted Resulting in Data Loss.

Flashes (Alerts)


Abstract

There is a potential for data corruption with IBM Tivoli Storage Manager (TSM) FastBack when a partial record of a snapshot is erroneously created in the FastBack Server database. If this occurs, metadata will be lost for a specific backup chain (specific backups of a single volume). The conditions that trigger this problem cannot be predicted. A fix is available.

Content

Data corruption can occur when a partial record of a snapshot is erroneously created in the FastBack Server database (called history.txt). As a result, the metadata of the backup is lost, and the customer is unable to restore from the affected backup chain because it is not considered a valid record by FastBack server. Other backup chains for other volumes remain intact. APAR IC81787 was opened to address this issue.

Affected Products:


Tivoli Storage Manager FastBack versions 5.5.x and 6.1.0 - 6.1.6.

How to Determine Whether Your Environment is Affected:

Step 1. If you experience a consistent snapshot failure on a backup chain with error message FBSS7017E, you are a likely candidate to have been affected by this problem. Check your FastBack Server logs for the following error messages:

W8.FSI : REP_FSI_S_OpenFileInRepository: File [\JOBXXXXX\MBLK0003.CTL] could not be accessed for reading.


E18.LIB : VolumeControl::get_lut_from_cache: LUTCache::get_lut failed for mega block 3 (job XXXXX)

E14.DM : DM_S_ReadVolumeMappingReply: Sorry, can't convert BITMAP to physical

I4.GENERAL : JOB_S_AddEvent: [YYYYY] 01/01/2012 01:02:03

FBSS7017E Failure to read volume structure. Procedure ended. If the problem persists try to manually run a Check Point snapshot.

Note: XXX and YYY stand for numbers seen in the logs.

If the log contains all the above error messages, you have been affected by this issue.
If the log does not contain all of the above messages, continue to the next step (step 2)

Step 2: Check the Client Log (Clog) of your FastBack Server for the following message:

FBSS7531E Cleanup on Policy: ‘<policy name>’ Volume: XXX:\ on <FB Client name> Failed.

Note: XXX is a volume number

If the Clog contains this message, you are affected.


If the Clog does not contain this message, continue to the next step (step 3).

Step 3: Run Cleanup process manually (via FastBack manager). Once the Cleanup ends, check the Clog for the error message described above (step 2).

If the Clog now contains this message, you are affected.


If the Clog still does not contain this message, you are not affected.

Solution:
Take a full snapshot of the affected chain, delete the previous snapshots of the chain, and apply the fix by installing FastBack 6.1.6.1 or higher.

Availability of Fixes:
ReleaseAffected levelsFirst level that includes the fix
Tivoli Storage Manager FastBack 6.16.1.0.0 through 6.1.6.0Version 6.1.6.1

Available in Fix Central

Tivoli Storage Manager FastBack 5.5All 5.5 levelsCustomers that use FastBack version 5.5.x need to upgrade to FastBack version 6.1.6.1 or higher to fix this issue

[{"Product":{"code":"SS9NU9","label":"Tivoli Storage Manager FastBack"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Component":"--","Platform":[{"code":"PF016","label":"Linux"},{"code":"PF033","label":"Windows"}],"Version":"6.1","Edition":"","Line of Business":{"code":"LOB26","label":"Storage"}}]

Document Information

Modified date:
26 September 2022

UID

swg21613910