04-02-2013 06:36 AM
Hi,
I have two complete IDENTICAL physical servers running Windows 2008 R2 with AD features.
Both servers host a global catalog, both servers are configured exactly the same.
Both servers run the same Job, with GRT for AD enabled.
One server backs up succesfull the other one fails.
Final error: 0xe0008442 - A memory allocation request failed, The computer may be running low on virtual memory.
Why ? How ? WTF ?
This backup product drives me crazy !!!
04-02-2013 06:46 AM
...are they in the same datastore or different datastores? If so, do you have enough disk space for snapshots?
Thanks!
04-02-2013 06:54 AM
Both servers are the same,
Diskspace usage C: 50Gb (21Gb free) D: 18,5GB (12Gb free)
Pagefile is on D:
Each server backs up to a different appliance.
btw: servers are Windows 2008 (without the R2 thingy)
04-02-2013 07:09 AM
Sorry, ignore my post above...should have been free disk space! Not datastore (busy with something else!).
Have you tried a server restart? Has the job been recreated?
Thanks!
04-02-2013 10:28 AM
Yep, already did that, same outcome.
Now I have changed the job to run without GRT for AD and it ran succesfully.
Both are global catalog for AD, the server that is successfull is also schemamaster, but I dont see how that could be of any difference for GRT backup.
04-02-2013 10:42 AM
...and the RAWS agent on the problematic server is the same level patch-wise as the successful server?
No AV installed perhaps causing issues? And have you checked to make sure there are no hardware issues on that server?
Thanks!
04-02-2013 10:52 AM
Both servers were deployed at the same time.
Both server run AV.
No hardware issues as far as I know.
The question is: what goes wrong on a Microsoft server to throw a 'memory allocation request failed' when a VSS snapshot is attempted on most probably the AD database ?
Because that should be the only extra action taken on the server when GRT for AD is enabled, right ?