1) Check to see if there are long-running backups in progress. The active log files cannot be moved into the archive log directory until all the parts of the transaction involved have completed.
2) If you have no active backups sessions, do: q actlog begindate=today-2 search=reorg See if you can tell if there is reorg of a DB table in progress. Those will fill the log really quickly, and it won't empty until the reorg is complete. 3) Check to see if your archivelog directory is full. If it is, the files can't get moved from the activelog to the archivelog directory, and the activelog will grow. Run 2 DB backups to empty the archivelog directory, and DB2 will start moving the activelog files into archivelog dir again. 4) At least on a Windows TSM server, the active log will always be the size you specify for ACTIVELOGSIZE in dsmserv.opt . The oldest files just get renamed and reused, but they stay there. The archivelogdir actually empties out, after you run 2 DB backups. Hope one of those helps! -----Original Message----- From: ADSM: Dist Stor Manager [mailto:[email protected]] On Behalf Of duholm Sent: Thursday, November 29, 2012 3:01 AM To: [email protected] Subject: [ADSM-L] Active log Hi All My TSM server have trouble commiting the Active log to the database. At the moment I have an active log at 55 GB, so I had to cancel all my tasks running, and it has not stop the log from growing. So can anybody tell me, how I empty my Active log and what the normel bottlenecks are. My DB2 database and active logs are located on 7 200 GB SSD discs and I am running TSM ver 6.3.2.0. The CPU and RAM of the server are not used up nor are there any big disc queue lengths on any disc's. Hope somebody can help my Tanks Bo +---------------------------------------------------------------------- |This was sent by [email protected] via Backup Central. |Forward SPAM to [email protected]. +----------------------------------------------------------------------
