asked on

Backup Exec could not back up the Exchange database because it could not find a healthy copy of the database.

I have a client with a Windows 2008 Standard SP2 Server. It is a single server running AD and Exchange 2010.

About half the time I backup Exchange, it works correctly.

The other half the time it fails with the following messages:

Final error: 0xe000038d - Backup Exec could not back up the Exchange database because it could not find a healthy copy of the database.
Final error category: Resource Errors

For additional information regarding this error refer to link V-79-57344-909
Click an error below to locate it in the job log
Backup- 2008-FS1V-79-57344-906 - AOFO: Initialization failure on: "Microsoft Information Store". Advanced Open File Option used: Microsoft Volume Shadow Copy Service (VSS).

Snapshot provider error (0xE000038A): The specified Logon Account credentials do not have the necessary Powershell permissions that are needed to complete the job. Review the credentials to ensure that it has the Exchange Organization Administrator role and sufficient rights.

I assume it can't actually be a permissions issue, or it would fail consistently.

Backup Exec has all the latest Live Update patches.

I've tried backing up to different devices, and I've tried isolating the Exchange Backup so that it runs as a separate job. No change.

Any help would be appreciated.

Thanks!

sunnyc7

Restart VSSWriter service on Exchange box and try a test run ?

NEMC

ASKER

Just checked. That service is not running and it's set for MANUAL startup.

Is that the problem?

sunnyc7

yes
You have to run the service and set it to automatic.

After you do that - restart all backupexec services
and then run a Test backup

NEMC

ASKER

OK, test backup I just ran without restarted BEXEC Services failed with same error.

I'll restart and try again.

Thanks,

sunnyc7

Also run this from cmdline on Exchange box

start > run > cmd
vssadmin list writers

Chck if all writers have NO Error.

relevant kb
http://www.symantec.com/business/support/index?page=content&id=TECH27875

NEMC

ASKER

After restarting BEXEC Services, backup is running.

No errors are listed when running vssadmin list writers

I'll keep an eye on the reports in the next week to confirm we are able to backup consistently.

Is MANUAL startup the default setting for the VS Service?

sunnyc7

automatic I would say.
Please set it to auto.

Usually VSS Writer errors are because of the service / errors in the writers.

NEMC

ASKER

Thanks for the great response.

I'll leave the question open for the next week while I monitor the backups and then close with points.

sunnyc7

Thanks

NEMC

ASKER

Service is terminating itself. I'm going to set it to auto recover and check the logs for indications why.

As a result, backups are failing again.

sunnyc7

Which service is terminating itself ?

NEMC

ASKER

VSS

sunnyc7

Where exactly are you storing your shadow copies. Is it in a separate logical partition ?
Is that partition running out of space - and hence the service is terminating because it cant write to that ?

Let me know if you are getting any errors.
Assuming- Windows Server 2008 / Exchange 2010

NEMC

ASKER

Yes, Server 2008/Exchange 2010.

Shadow Copies are probably on the same drive. I had checked available disk space earlier, but will look again.

ASKER CERTIFIED SOLUTION

sunnyc7

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

sunnyc7

I tracked down some Service stopping issues to space constraints on the disk.
Trendmicro anti-spam would shutdown by itself and then not restart (space issue > directed temp folder to another location)
VSS issue > also space - directed to separate volume.

NEMC

ASKER

OK, thanks I'll give it a try.

NEMC

ASKER

I was unable to change to volume for storing the shadow copies.

Restarting the service and enabling the service to auto-restart made no difference.

I am going to assume that sunny's answer would have worked and award points since this question has dragged out for too long.