Deduplication of large files on Tivoli Storage Manager Version 6.1 may cause may cause active database to run out of space.
Flash (Alert)
Abstract
Analyzing very large files in a deduplicated storage pool on Tivoli Storage Manager Version 6.1 might cause the active database log to run out of space or cause other sessions or server processes to stall.
Content
Problem:
The active database log for a Tivoli Storage Manager Version 6.1 server might run out of space when an Identify Duplicate process is analyzing a file that is hundreds of gigabytes or larger. These files can have several million extents. Each extent is represented in the database and is stored in the active database log until the database entries are committed. Because of the large number of extents, it is
likely that the active database log will run out of space.
A side-effect of this problem is that client sessions or server processes might stall until the Identify Duplicates process that is analyzing the large file finishes. Analyzing large files holds resources such as locks and volumes until the Identify Duplicates process is completed. This prevents other sessions and processes from acquiring these resources and causes them to have to wait. The stall could last anywhere from several hours to days, depending on the size of the file being analyzed.
Who is affected:
Tivoli Storage Manager Version 6.1 users who store very large files into storage pools that have deduplication enabled.
Recommendation:
Use the following circumventions until a fix is available.
Circumvention:
If the storage pool with deduplication enabled has a NEXTSTGPOOL defined, customers can use the MAXSIZE parameter to prevent very large files from being stored in the deduplicated storage pool. Bypassing the deduplicated storage pool prevents large files from being analyzed and causing the problems described in this flash.
If an Identify Duplicates process begins processing a very large file, issue the QUERY PROCESS command and find the process. Note the volume that the process is currently analyzing. Cancel all Identify Duplicates processes and update the volume to make its status UNAVAILABLE. This will prevent future Identify Duplicates process from selecting the volume. If the volume is needed for a client restore or retrieve, cancel any Identify Duplicates processes running at the time and update the volume status to READONLY. Once the client restore/retrieve has completed, update the volume status to UNAVAILABLE again and resume the Identify Duplicates processes as needed. Once the fix has been applied, update the status of any volumes that were made UNAVAILABLE to READONLY . The Identify Duplicates processes will then begin processing any files on the volumes that have not already been analyzed.
Cross Reference information
Segment
Product
Component
Platform
Version
Edition
Storage Management
IBM Tivoli Storage Manager
Server
Platform Independent
6.1
All Editions
Copyright and trademark information
IBM, the IBM logo and ibm.com are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at www.ibm.com/legal/copytrade.shtml.
Rate this page
Please take a moment to complete this form to help us better serve you.
