02-24-2012 01:37 AM
Can anyone help me.
We have several Enterprise Vault archiving servers. Most of the wok fine, but there is one with a large ammount of saveset (close to 7.000.000).
Off course the event viewer now display a lot of 41021 and 41008 event-id's.
I have checked for the ignorearchivetrigger.txt, but this one is present and in the right location.
I dont see the .old file though. So it appears the backup doesn't change the extention of the file.
What can i do?
regards,
Rob Knegt
Solved! Go to Solution.
03-07-2012 10:49 PM
I think i'm going to open a case. the amount of savesets is now 6.800.000. How is this possible in report mode? Doesn't report mode mean it dosen't archive?
02-24-2012 02:18 AM
How old is the file ignorearchivebittrigger.txt ?
Is the file being created by your backup tool ? See this technote for a pre/post script:
http://www.symantec.com/business/support/index?page=content&id=TECH35610
02-24-2012 02:30 AM
I just checked again and found the .old file. so that looks allright. Ii am going to check the techfile
02-24-2012 04:36 AM
02-24-2012 04:43 AM
We are on 8.0.3
02-28-2012 11:45 AM
Since Enterprise Vault uses the created date of the file I would look at what the date of that file is as a starting point.
Also there were some issues on 8.0.3 are there any errors reported from the StorageFileWatch process?
02-28-2012 12:52 PM
I can testify that the process of backup and having the backup write the IgnoreArchiveBitTrigger.txt is ok, as I configured it at that customer.
I've told rob to check if it is not a new location, or perhaps a missed one. I also advised to check eventviewer. I hope he replies here to let us know if the issue is resolved.
02-28-2012 11:11 PM
Hi gertjan,
the ignorearchiveTriggerbit is not in another location. For now i have set in report mode, to see if the large amount drops. I'll keep you posted
02-28-2012 11:24 PM
Just checked teh event viewer and saw ther are a lot of errors regarding invalid savesets (event id 7109).
Could this be the reason?
And if so, how can i fix it?
02-29-2012 12:31 AM
From 2009, when I was there..
02-29-2012 12:59 AM
Way ahead of you ;)
Like i said i'm already in report mode. looks like the amount of savesets is dropping :).
But now i'm running into the event id 7109. And this causes the next issue.
And already got the 6265 id.
So i made i proactive reboot of the system before it goes in back-up mode
02-29-2012 01:47 AM
If you see the event 6265, it is indicating a lack of memory issue. Although it says 'not enough storage available for operation', this issue is related to RAM. On that server, check task-manager, performance. Check to see what the memory usage is.
If it is high, perhaps you can have the VM-team give you an extra 2 GB of Ram (machine now has 1.5 right?) for the time that is needed to work away the savesets.
If I recall correctly, last time this happened, we had the task in report mode for about a week before the savesets were processed.
This is (btw) another reason to upgrade to 9.03. In 9.03 some improvements have been done in regards to memoryusage/leaks/performance etc. Might be an extra handle to convince people to upgrade....
02-29-2012 02:18 AM
I did, and seemed fine. But event 6265 is often followed by event 4183. So to prevent ev from shutting down during back up mode i restarted it pryor to the back up. since than i haven't seen event 7109 as well.
fingers crossed
03-01-2012 06:29 AM
There was a new VM Best Practice guide published in January this year. It is mainly around EV10 but you may find it useful to at least read through to see if there are any nuggets of info.
http://www.symantec.com/docs/TECH180094
03-01-2012 01:46 PM
i'll look into it. TNX
03-07-2012 04:59 AM
The number of savesets appears to be going down, but very slowly. It's now at about 6.700000. But it's weird. It goes down, at one point we were at 6.400000. Although in report mode constantly sometimes the amount of savesets still adds. How is this possible in report mode?
03-07-2012 05:05 AM
Hi rob,
Are you sure the store remains in backup-mode? Is it not taken out of backup by the backup-software by accident?
If I recall correctly, there should be a batchfile that is called by the backup software to set/clear backup mode. Try to locate these, and either rename, or rem out the command that clears backup mode.
On the affected server, check the eventlog for event 7080 (clearing backup mode). If you see this happening, the command is run from somewhere.
03-07-2012 05:11 AM
It is in restore mode, back up mode is running daily.
Event 7080 is not in the logs
03-07-2012 06:07 AM
I found a bunch of batch files for pre and post back up. however i can not find the command that clears backup mode. But i'm going to check with the batch files on other ev servers to see if there are different command or settings
03-07-2012 10:49 PM
I think i'm going to open a case. the amount of savesets is now 6.800.000. How is this possible in report mode? Doesn't report mode mean it dosen't archive?