Exchange 2010 Granular Backup Failing, Status 71/4207

[Tim_K88]

Good day all, 

I just deployed a new Netbackup environment and need some assistance please. I apologise if this has already been asked but I cant seem to find anything with the same issue as what I am experiencing. 

I have 2 x Exchange 2010 SP3CU11 boxes running on top of Windows Server 2008 R2 x64. The initial granular backup worked perfectly with no errors. I have done all the necessary config with the "Replace a level token" and added the user to the correct groups etc, however my Incrementals been trying to run and failing with the following errors in the job log;

<LOG1>

03/03/2016 16:20:06 - Critical bpbrm (pid=6424) from client server1.company.here: FTL - snapshot preparation failed - Error attempting to find volumes to snap., status 130

03/03/2016 16:20:06 - Critical bpbrm (pid=6424) from client server1.company.here: FTL - snapshot preparation failed - Error attempting to find volumes to snap., status 130

Could not fetch snapshot metadata or state files  (4207)

</LOG1>

<LOG2>

03/03/2016 17:27:53 - Critical bpbrm (pid=5312) from client server1.company.here: FTL - snapshot preparation failed - Unable to backup Exchange database. It may not be mounted., status 71

03/03/2016 17:27:53 - Critical bpbrm (pid=5312) from client server1.company.here: FTL - snapshot preparation failed - Unable to backup Exchange database. It may not be mounted., status 71

Could not fetch snapshot metadata or state files  (4207)

</LOG2>

<LOG3>

03/02/2016 18:00:27 - Critical bpbrm (pid=124166) from client server2.company.here: FTL - snapshot preparation failed - Unable to backup Exchange database. It may not be mounted., status 71

Could not fetch snapshot metadata or state files  (4207)

</LOG3>
 

I have done the following;

- Checked the location "C:\Program Files\Veritas\Netbackup\online_util" to ensure that permissions are correct on the folder

- Restarted VSS on the client

- Checked that the DBs are in fact mounted in Exchange

- Forward and reverse records working ok with bpclntcmd

- Check the credentials for the Service Account that runs the Exchange Backup

- Checked that its in the correct groups and has the correct Local Security Settings

I dont understand how one backup performs perfectly and then anything after that does not complete successfully.

I look forward to your replies.

Thank you kindly!

[Mark_Solutions]

Is this a MS-Exchange policy type?

The reason i ask is that it looks like a VMware policy type .. and they do not support incremental backups, only full ones

Even an MS-Exchange policy type will not do any GRT during an incremental backup - it backs up only the log files ... that being the case if your exchange uses circular logging it will cause it to fail

[Tim_K88]

Hey Mark, 

Yeah it is an MS-Exchange Policy type.

Circular logging is currently off as well. 

I will keep reading the admin guide and see what I can come up with. Ideally I just want a weekly/monthly full with diff/incremental in between.

Thanks!

[Marianne]

What is configured as client name in the policy? Hopefully the virtual hostname associated with the DAG and not individual node name? Do you have bpresolver log on both nodes to see if location of Information Stores are correctly resolved?

[Tim_K88]

The client name in the policy is the FQDN.

I do not have a DAG in place so is just a straight forward backup of the mailbox databases on individual hosts. 

I havent checked the bpresolver log at all, I will investigate where they are and whats in them. 

Thank you for the advice! :)

[Marianne]

So, no cluster? 2 standalone Exch servers? 
With separate policies?
What is in Backup Selection of Exch policy/policies?
Are you seeing this error for one or for both Exch servers?

bpresolver log is for DAG.

Other logs needed on Exch servers:

bpfis 
bpbkar
nbfsd

About logs - see the Troubleshooting section of  NetBackup for Microsoft Exchange Server Administrator's Guide 
 

[Tim_K88]

Hey Marianne, 
Thank you for your quick responses :)

That is correct, I have 2 backup domains with AIR between the two for replication. My Exchanges are not clustered and are indeed both standalone.

They have separate policies and has 2 different masters as well. 

Backup selection I just changed right now but I had 3 x Exchange DBs with a New Stream between them, and the other policy had only Micosoft Information Store with a single stream.

I am seeing this as a very intermittent issue as sometimes after a reboot it works fine with the exact same settings and everything. I get a feeling this is a VSS/Snapshot issue since I see the following in the logs, see attached.

I have removed any specific identifiers for safety reasons. I hope this helps.

Thank you kindly

[Marianne · Accepted solution]

For sure VSS snapshot errors. All I can suggest is to check Event Viewer logs for VSS errors. And maybe use vssadmin commands to troubleshoot VSS.

[Tim_K88]

Hey Marianne, 

Thank you so much for your assistance. With diskshadow and vssadmin I was able to get rid of the vss issues. It turns out that my vss on C: was stuck and I couldnt alter it at all. I deleted all the disk shadows and then changed the shadow copies manually.