Exchange 2010 transaction log

[aeccles]

I just started backing up Exchange 2010 and have run into this issue as well:

Event ID 9782: "Exchange VSS Writer (instance 05a7ede3-75f0-4d14-bed9-1eabb05d611e:30) has completed the backup of database 'Users E-K' with errors. The backup did not complete successfully, and no log files were truncated for this database."

This is a single Exchange 2010 server. It has 4 mailbox databases but this is the only one that an error is reported on.
I am not truncating the logs as Backup Exec performs a full backup each night and that process truncates the logs. When Backup Exec runs it's full backup, there are no errors.

So to recap, this appears to be specific to Veeam and specific to just one of my four databases.

The above event log is followed immediately by
Event ID 2007: Event Description: Information Store (5856) Shadow copy instance 30 aborted.

[zoltank]

As per my previous post this appears to be expected behavior when backing up Exchange without truncating the transaction logs.

I'd also ask why you're still backing it up using Backup Exec when you're also using Veeam? Veeam is superior to BE when backing up a virtualized Exchange 2010 server. The best solution here would be stopping your BE backup of Exchange and allowing Veeam to truncate the logs.

[aeccles]

Using BE to tape for compliance. That will change once v7 is out, hopefully.

Were you also getting "Event ID 2007: Event Description: Information Store (5856) Shadow copy instance 30 aborted." ?

I mispoke earlier when I said I was only getting event ID 9782 for a single database. I am in fact getting the same event ID for each of the databases.

[zoltank]

You can use BE to back up the Veeam backups files to tape. That way you're still compliant, but you can stop your BE Exchange backup and shrink your backup window. It's what we've been doing for a couple years now and works well. It's a more elegant solution with fewer potential issues.

I'm fairly certain I was also getting the 5856 event, but I can't check for sure because my event log doesn't go back that far. Try running your Exchange Veeam job once with truncation turned on and see if all the errors don't go away.

Also, if you haven't already, set up a SureBackup job to check your Exchange backups. It's very cheap insurance.

[Dima P.]

This is a single Exchange 2010 server. It has 4 mailbox databases but this is the only one that an error is reported on.
I am not truncating the logs as Backup Exec performs a full backup each night and that process truncates the logs. When Backup Exec runs it's full backup, there are no errors.

Please refer to this thread as well. This investigation clarifies that when you got Backup Exec agent installed on the VM which needs to be backed by Veeam B&R up using VSS. To cut a long story short the issues may arise due the components (not native VSS writers and so on) that are deployed during agent installation.

Cheers!

[aeccles]

I'm backing Veeam up to a WAN target and my tape library is in a different location so it's not possible without creating another set of jobs. Adding another set of Veeam jobs doesn't make sense, if I'm just going to put them on tape.... Restore time goes up significantly and you need two products installed to restore. I hate BE, but it works flawlessly *knock on wood* for me in this scenario...

BTW, as we have been posting I was working with Veeam support and they confirmed that two event logs I posted about are normal behavior when you are not truncating log files.

[veremin]

BTW, as we have been posting I was working with Veeam support and they confirmed that two event logs I posted about are normal behavior when you are not truncating log files.

Yep, it’s totally expected to get this kind of messages regarding Exchange logs with a log truncation option disabled.

Hopefully, the upcoming release will answer your requirements and you will get all-in-one solution for your environment.

Thanks.