Veeam 9.5 and Exchange 2016 Cannot notify writers about the ‘BACKUP FINISH’ event.

This is post is more of a complaint than anything really useful, anyone that relies on snapshot based backups for Microsoft Exchange has seen VSS errors on any sizable environment.  Generally the job fails, Veeam will try again and all is well.  However the below “warning” came up recently.  Even though require successful backup completion is checked and Veeam is well aware of the error since it is the one reporting it. It still marked the job as successful.  I get it, the backup itself was complete Veeam as a backup product did its job.  However I now have logs building on my Exchange servers.  This error really has nothing to do with Veeam and if you call support they will likely point the finger at Microsoft and their VSS writers.

VSS: Backup job failed.

Cannot notify writers about the ‘BACKUP FINISH’ event.

A VSS critical writer has failed. Writer name: [Microsoft Exchange Writer]. Class ID: [{76fe1ac4-15f7-4bcd-987e-8e1acb462fb7}]. Instance ID: [{493f088b-00a2-4eb7-be5b-1c4c61dae2f2}]. Writer’s state: [VSS_WS_FAILED_AT_BACKUP_COMPLETE]. Error code: [0x800423f3].

I guess a feature request to Veeam would be to actually have it retry or restart the store service as an option instead of just marking the job as successful.

If you do run into this error  to validate:

Open and elevated command prompt and run the command

vssadmin list writers

Look for Exchange, if you see this its like any other time VSS has failed

Writer name: ‘Microsoft Exchange Writer’

Writer Id: {76fe1ac4-15f7-4bcd-987e-8e1acb462fb7}

Writer Instance Id: {493f088b-00a2-4eb7-be5b-1c4c61dae2f2}

State: [12] Failed

Last error: Retryable error

Sometimes you can just restart the Exchange Store service and that will clear it, more often than not when I see this I need to restart this node.

Now in my environment the servers being backed up are passive nodes that no active database copies currently run on so its not a big deal.

There are lots of reason why this happens, after years of backing up exchange in a virtual environment I just generally accept that the VSS writers are not a stable software and move on with life.

If you want to automate this if you see this a lot in your environment, run a script to as a Veeam pre task to restart the store service before the backup job runs.

Just to be clear, this error just means that the logs didn’t get truncated.  The backup was successful and really nothing needs to be done as it will probably run fine the next time a backup is run.  If you see this error repeatedly in your environment or it happens back to back multiple times then you may have a problem.

However if you have the Veeam console in front of you and you are worried about your exchange log volumes filling up, restart the exchange store and replication service on the problem exchange server start the backup job again.  It may take a few tries but your job will run faster since you just did a back recently.