Link to home
Start Free TrialLog in
Avatar of chrisp666
chrisp666

asked on

Exmerge problem

I use Exmerge to perform a brick-level backup of mailboxes from our Exchange store. This batch job has run happily for ages, but has suddenly failed:

[20:03:04] Copying data from mailbox 'Markxxx' ('MARKxxx') on Server 'FORSETI' to file 'C:\EXMERGEDATA\MARKxxx.PST'.
[20:03:04] Error configuring message service (MSPST MS) (UNKNOWN ERROR)  (CMapiSession::CreateEMSPSTProfile)
[20:03:04] Errors encountered. Copy process aborted for mailbox 'Markxxx' ('MARKxxx').

The exmerge.ini file specifies copying to Z:\EXMERGEDATA\MARKxxx.PST. This share exists and the permissions have not been changed. As always with exmerge, if you run it manually, it runs just fine! I have Googled around for this, but found nothing that works. One suggestion was to delete the share and recreate it, which I did, and miraculously, it worked once but then failed thereafter! Any ideas, anyone? Thanks in advance.
The exmerge.ini file:
 
; EXEMERGE.INI
; This file is for use with the EXMERGE.EXE program, version 4.00 or higher.
 
[EXMERGE]
LoggingLevel=0
LogFileName=ExMerge.log
DataDirectoryName=Z:\EXMERGEDATA
MergeAction=0
RestoreDB=0
SourceServerName=FORSETI
DomainControllerForSourceServer=
SrcServerLDAP-Port=
DestServerName=
DomainControllerForDestServer=
DestServerLDAP-Port=
SelectMessageStartDate=
SelectMessageEndDate=
ListOfFolders=
FileContainingListOfFolders=
FoldersProcessed=2
ApplyActionToSubFolders=0
FileContainingListOfMailboxes=
FileContainingListOfDatabases=
RemoveIntermediatePSTFiles=1
DateAttribute=0
DataImportMethod=1
ReplaceDataOnlyIfSourceItemIsMoreRecent=1
CopyUserData=1
CopyAssociatedFolderData=0
CopyFolderPermissions=0
CopyDeletedItemsFromDumpster=0
FileContainingListOfMessageSubjects=
SubjectStringMatchCriteria=0
FileContainingListOfAttachmentNames=
AttachmentNameStringMatchCriteria=1
MapFolderNameToLocalisedName=0
[International]
DefaultLocaleID=2057
UseLastLogonLocaleID=0

Open in new window

Avatar of jesusrulesme
jesusrulesme
Flag of United States of America image

A couple of things to try.  Point the batch file into a separate directory altogether or even the root of the drive, just to see if it works.  Also try running chkdsk on the drive.  
Avatar of chrisp666
chrisp666

ASKER

Yes, good idea...I will try it tonight when everyone has gone...thanks.
If ExMerge is working incrementally check and make sure the existing PST isn't read-only. I assume this is happening for all mailboxes?

The reason the file often runs differently when running it manually is the RUNAS from the job runs in a slightly different environment with less environment variables and from a different home location. If you're running as a different user in your scheduled job check the permissions for that user or use the user it works correctly with. A way to test this is to schedule a command line to launch then use that command line to run your batch file and see what errors are generated from the command line, if any.
renazonse, thanks...no, they are not read-only and the job runs (ran!) as administrator, just as you're not supposed to do it, but I had given administrator non-inherited 'send as' rights, and as I say, it used to work.

As regards your second point, I would never have thought of that in a million years and will try it tonight.

Thanks again...Chris
Well, changing the output directory didn't work - it still tried to write to the non-existent folder on the C:\ drive. Scheduling a command prompt and then running the batch job within that worked correctly, but for some reason seemed to use much more cpu than normal.

More confused than ever.
Do your existing PST files get opened from time to time by the Outlook clients? I've read that once these files are opened by a client it can piss off ExMerge preventing ExMerge from being able to write to it, which would explain the CPU usage. You may need to re-export the problem PST...unless this is happening for all of the PST files instead of just the one in the error?

I assume the Z: drive is a mapped drive? Is it possible the Z: drive is becoming unavailable at the time of the ExMerge job? I guess it's possible if the Z: drive is not available the ExMerge will try C: drive, not sure about that. What's the exact command you're using to run the ExMerge process? Any special switches?
Hi renazonse, no, these are purely archives and I am the only one with access. The Z: drive is permanently mapped...I, too, think this is the cause of the problem, but haven't a clue what. As I mentioned, I disconnected it and remapped it and the batch job ran...once! I think what I will do next is to disconnect the share and map it dynamically in the batch job - I wouldn't have thought of that without your reply, so let's hope! I will try it tonight as it generates a lot of network traffic as you can imagine.

Thanks again.
Makes me wonder if the server that hosts your Z: drive is having an issue...failing drive maybe? Or, maybe the drive just disconnects after being idle for a certain amount of time? jesusrulesme may have had the right idea with the chkdsk.

I guess another thing to look into would be licensing. If this worked for a long time and never had problems becoming near your license max may cause the mapped drive to be disconnected by the hosting server.
I've also seen issues where there is a lot of disk activity and corruption occurs from exmerge.  You might run a chkdsk on the Z: drive before running exmerge.
I'll certainly run chkdsk, but am slightly dubious as to whether that's the problem. The Z:\ drive is a Lacie NAS device and it is also used for the daily backups...there are no problems with those as I periodically restore one to another machine just to make sure the backup is working ok. I hasten to add, the backup job is not run at the same time as Exmerge.
Well! I never used to believe in magic, but now, maybe...I haven't got around to doing anything that I said I would in previous comments; I logged on to the server this morning to run chkdsk and found that Exmerge has run quite happily for the previous 2 nights! Confused?
I give up on logic. The job stopped running again as mysteriously as it started. I now connect to the share within the batch job and it has run successfully for a while. If it continues to do so, I will consider the matter closed. Thanks to all for your input.
ASKER CERTIFIED SOLUTION
Avatar of Britt Thompson
Britt Thompson
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
I can't think of much else, either - I am still pretty baffled, though; as I said earlier, the daily backup of the entire system goes to the same network device. Thanks anyway.