TSM error 11 when log archiving
Hi, we have old DB2 system with DB2 v8 on AIX. Due to financial constraints this hasn't been updated yet but works fine. However TSM password has expired and stopped backup and log archiving. TSM staff reset the password and now backup works but log archiving still failed with the following error. This happened long time ago and resetting the password resolved the issue. However the TSM staff is a new member for the company. Could you give some advice please ? Extract from the db2diag.log : PID : 2703600 TID : 1 PROC : db2logmgr (PRD001) 2 INSTANCE: db2prd1 NODE : 002 FUNCTION: DB2 UDB, data protection, sqlpgRetryFailedArchive, probe:4780 MESSAGE : Still unable to archive log file 150 due to rc 11 for LOGARCHMETH1 using method 2 and target STANDARD. MESSAGE : Failed to archive log file S0000150.LOG to TSM chain 0 from /db/db2prd1/db2prd1/NODE0002/SQL00001/SQLOGDIR/ with rc = 11. FUNCTION: DB2 UDB, data protection, sqlpInitVendorDevice, probe:1030 MESSAGE : sqluvint failed! Vendor rc:
TSM RC 11 means there isn't space available for storage. Possibly backups and logs belong to different TSM policies and are stored in different pools, this is why backups succeed and log archiving does not. 0011 E DSM_RC_ABORT_NO_REPOSIT_SPACE Explanation: The server does not have any space available to store the object. System action: Tivoli Storage Manager ended the current operation. User response: You can take any of the following actions: Request the system administrator to add space to the storage pool. For Tivoli Storage Manager client, set COMPRESSALWAYS=NO and COMPRESSIon=YES in the options file (DSM.OPT), then the file will be resent uncompressed if it grows during compression. For API Applications, consult the application's documentation for recommendations regarding compression. Turn off disk caching in the disk storage pool, and issue MOVE DATA commands to each disk pool volume to clear out the cached bitfiles.
rc = 11 is not the real RC coming back from TSM. You need to check the RC after the following line: MESSAGE : sqluvint failed! Vendor rc: See this technote for more info: IBM Interpreting Vendor API return codes from db2diag.log messages - United States Originally Posted by stj20 However TSM password has expired and stopped backup and log archiving. TSM staff reset the password and now backup works but log archiving still failed DB2 loads TSM API when the backup is initiated, but DB2 logger loads it when the instance is started and a portion of it is held in memory until instance recycle. You can try restarting the instance. The real RC from TSM will tell you why it's failing.
Hi, This is the TSM message based on 'sqluvint failed! Vendor rc' 2041 E: DSM_RC_BAD_CALL_SEQUENCE The sequence of calls is invalid. This does not say much. Backups and the Logs go to the same storage pool and it's unlikely that this is a Storage issue. Thanks in advance for your help.
Have you tried restarting the instance? Because backup is working and log archival is failing, I suspect that instance recycle will help here.
for backup you need backup stgpool and for logs you need archive stgpool usually that can not be the same we had a similar problem and killed the db2vend process for that db - it will restart for pw problem it is better to use the generated option and execute dsmapipw
Thanks all and the restart works and do not see any messages on db2diag.log. By the way how should I make sure archiving are happening into TSM without depend on DB2diag.log. In other words how can I prove the issue has resolved.
we use the db2diag command to find messages from last hour that have the text archive in it and ship mail whenever we found this message
Originally Posted by stj20 Thanks all and the restart works and do not see any messages on db2diag.log. By the way how should I make sure archiving are happening into TSM without depend on DB2diag.log. In other words how can I prove the issue has resolved. Hi, You can check if LOG files are going into TSM with this command : db2adutl query logs (with db2adutl query command, you can check also the backups)
I appreciate this is fixed now. I just wanted to add that when we got a RC 11 and archive logs failing (and backup failing as well) it was because the db2 instance owner could not write to the dsierror.log. (AIX server) There are two db2 instances on the server and only one of them could write to the log file as it was in it's home directory. In the end we put this line in dsm.sys: ERRORLOGNAME /var/log/tsm/dsierror.log ensured it was opened to both instances and it all worked again.
Thanks everyone for your valuable inputs .
Another way to to use "list history archive log" command. When you migrate to v9.x (I think this is not in v8), you will see messages about logs getting archived in the db2diag.log.
Is there any command in UDB/Linux to List all mount points for a given instance?
sql stored procedure default parameter
Z/os Db2 Unload Utility Question
getting current time in SQL
Get diagnostics exception - error line
I need to force application one bye one as script in Solaries
S04E for COBOL/DB2 stored procedure
Is Reorg needed after table LOAD?
AUTORESIZE in DB2 V8.2
question about IDENTITY
**Error while configuring DB2 client : URGENT**
about creat a table
How to move/cleanup active log files
Advanced Merge code
problem with creation of buffer pool
Arabic char conversion from CCSID1256 to 420