I just have to say in response to all this interest in query events ... it has problems.
You can have an event that says "Completed", but you didn't get a backup (the TDP script started, but then failed later.) You can have an event that says "Missed", but you DID get a backup (operations or the user kicked it off manually). You can have lots of nodes that don't even HAVE events (they're scheduled by cron, or Autosys, etc., not by TSM) So we use the filespace's backup-complete date to see whether all filespaces are up to date on their backups. This is harder to do, but it has a VERY USEFUL side effect: it finds wasted space. If you look at all your backup-complete dates, you'll find some filespaces that are VERY OLD, i.e., haven't been backed up in a year. These are usually junk - like, the Oracle database you moved from node A to node B a year ago. YOU deleted the 200 GB of old filespace on node A... ...but TSM never did, and it's still sitting on the last version of that 200GB worth of files. We had one customer find almost FOUR TERABYTES of wasted space in their overloaded library, that they didn't know they were sitting on. I bet other people have this situation too. If you're running out of space in your library, take a look at this possibility. -------------------------------- Mr. Lindsay Morris Lead Architect, Servergraph www.servergraph.com 859-253-8000 ofc 425-988-8478 fax
