Forum Discussion

Ravi_Bhandari's avatar
11 years ago
Solved

Aborting restore: Error writing Exchange object: Microsoft Information Store

I have netbackup 7.5.0.6 master running on Windows 2008R2 backing up an Exchange 2013 cu 7 server running on windows 2012 R2 which is VMWare 5 virtual host.

I'm desperately trying to restore a Mailbox Database to a Recovery Database but keep getting status 2810 and "client restore EXIT STATUS 185: tar did not find all the files to be restored"

I'm sure that this has worked previously but can't for the life of me get to work when I really need it to.

Much appreciate any advise or strategies to diagnose/troubleshoot

  • If the backup was taken with an unsupported combination of versions of MS Exchange version and NetBackup Client version - then patching the NetBackup Client to a supported version in combination (with MS Exchange version) is not necessarily going to help - because the original backup appears to have been secured with an unsupported version combination - so, if the backup combination was originally unsupported, then any attempt to restore from an unsupported version combination is, by definition, also unsupported.

     

8 Replies

  • Can you try some of advice given in this tech note:

    http://www.symantec.com/docs/TECH146996

    Have you tried to restore a older or newer backup ?

    Please also take a look at the event viewer for Exchnage related errors

    If you are down/hurry - open a case with Symantec/Veritas right away.

  • My problem has moved forward, now I have this.

     

    16:52:07 (7279.001) WRN - MountDatabase failed with error(0x0) for machine EXCHSVR2, storage group Recovery Mailbox Database, database Recovery Mailbox Database
    16:52:07 (7279.001) ERR - An Exchange restore error was detected.  You may need to manually mount the Database stores to successfully finish the restore.
    16:52:07 (7279.001) INF - TAR EXITING WITH STATUS = 5
    16:52:07 (7279.001) INF - TAR RESTORED 4 OF 4 FILES SUCCESSFULLY
    16:52:07 (7279.001) INF - TAR KEPT 0 EXISTING FILES
    16:52:07 (7279.001) INF - TAR PARTIALLY RESTORED 0 FILES
    16:52:09 (7279.001) Status of restore from copy 1 of image created 03/05/2015 18:03:04 = file read failed
     

    I am unable to manually mount the database either.

  • 1)Kindly check the logs whether data is successfully restored, you can check by confirming

    completed reading backup image        
    EXITING with status 0

     

    2) In our enviroment exchange restore too fails with 2850 and 5 error code but we check the above status and then ask exchange admin to mount and extract data

  • I was unable to mount the database manually. I had to first repairr the recovery database and then remove the logs before I could mount it.

    I'm wondering why the restore left the database in an inconsistent state?

  • Thanks for all of the responses.

    I've applied the 7.5.0.7 patch but still have the same problem. I believe that the restore/recovery is failing because of corrupted logs. The log file it fails on is present but it's size is considerably smaller compared to other logs. Also the ESEUTIL tool highlights missing logs.

  • If the backup was taken with an unsupported combination of versions of MS Exchange version and NetBackup Client version - then patching the NetBackup Client to a supported version in combination (with MS Exchange version) is not necessarily going to help - because the original backup appears to have been secured with an unsupported version combination - so, if the backup combination was originally unsupported, then any attempt to restore from an unsupported version combination is, by definition, also unsupported.

     

  • :-) Ravi backup was taken with unsupported version . Symantec cannot guarantee data recovery.

    In short Restore will fail