A couple of days ago our exchange server died due to lack of disk space, turns out for over 2 years the transaction log files have not been purged after each successful backup.The backup date on the database is 31/08/2007. I am using Backup exec 10d and it is backing up the info store.The only thing I can think of is that there are usually 2 or 3 emails that backup exec says are corrupt so the backup reports as being successful with exceptions. I am wondering if the presence of these "corrupt" emails is stopping the backup reporting as being successful and as such Exchange is not seeing the backup as being successful and doesn'tchange the backup date on the database.Anyone got any ideas?

ok..update...I just ran a backup where the only thing selected was the info store, backup ran fine and logs purged, not sure why the normal backup isn't doing this though, is it possible that since the backup is running at the same time as the maintenance schedule that this is causing some kind of conflict?

Yes make sure that Backup is not interfering with Online Maintenance. Database date will not be changed unless and until the full backup was completed successfully. While taking fullbackup please double check that you have selected the Storage Group and not just database to purge logs. You can also check the same via running below command over chk file eseutil /mk e00.chk also event logs can give you info about if full backup was completed or not http://www.msexchange.org/tutorials/Exchange-2003-Backup-Restore-NTBACKUP.html Vinod |CCNA|MCSE 2003 +Messaging|MCTS|ITIL V3|

I have the storage group selected so I think the issue is with the backup and Online Maintenance conflicting, it looks like the backup date hasn't changed since the servers were installed so that would make sense.

On Wed, 23-Sep-09 11:17:11 GMT, Richard Bruce wrote:>ok..update...I just ran a backup where the only thing selected was the info store, backup ran fine and logs purged, not sure why the normal backup isn't doing this though, What you just did is a "Normal" backup. What you've _been_ doing issomething called a "brick level backup" (i.e. making a copy of themailbox contents, not a backup of the database).Only "Normal" (i.e. "Full") or "Incremental" backups remove log files. the backup is running at the same time as the maintenance schedule that this is causing some kind of conflict? No, it's becasue you've not been making a backup of the database.---Rich MatheisenMCSE+I, Exchange MVP--- Rich Matheisen MCSE+I, Exchange MVP

Actually Rich, I was doing a backup of the First Storage Group, NOT just the mailboxes. The solution was actually very very simple! The backups had been detecting 3 corrupt mails in mailboxes, these were selected for backup prior to the First Storage Group, due to these corrupt mails, Backup Exec was not (and I have no idea why but...) telling Exchange that the backup of the First Storage Group was successful so the logs woul dbe purged.I re-ordered the backup, putting the First Storage Group as the first item to be backed up, nothing else was changed, so the same emails are still being seen as corrupt, but since the First Storage Group is backed up before any exceptions are encountered by BackupExec, Exchange gets the message that the storage group was backed up and the logs are purged.Simples!!

And that is a problem with BE, not Exchange. If BE is reporting corrupt emails, than you are also doing mailbox level backups. ( BE has had that "issue" for years). Mailbox Level Backups are not a Microsoft solution nor directly supported by Microsoft.

On Fri, 25-Sep-09 08:47:05 GMT, Richard Bruce wrote:>Actually Rich, I was doing a backup of the First Storage Group, NOT just the mailboxes. The solution was actually very very simple! The backups had been detecting 3 corrupt mails in mailboxes, If you were receiving warnings about "corrupt mails" then you were notbacking up the databases in the storage group. A database backupdoesn't examine the contents of mailboxes, it only verified that themorethan that, but it's really only interested in the integrity of thedatabase, not the data stored within the file).>these were selected for backup prior to the First Storage Group, due to these corrupt mails, Backup Exec was not (and I have no idea why but...) telling Exchange that the backup of the First Storage Group was successful so the logs woul dbe purged.Sounds like a problem with your backup software, not Exchange.>I re-ordered the backup, putting the First Storage Group as the first item to be backed up, nothing else was changed, so the same emails are still being seen as corrupt, but since the First Storage Group is backed up before any exceptions are encountered by BackupExec, Exchange gets the message that the storage group was backed up and the logs are purged.Simples!! Exchange "gets the message" from your backup software, nothing else.When the backup completes successfully the backup software makesanother call to the system attendant to purge the log files. If theog files are intacts, and the checkpoint file is correct, theunnecessary log files are removed.---Rich MatheisenMCSE+I, Exchange MVP--- Rich Matheisen MCSE+I, Exchange MVP