cancel
Showing results for 
Search instead for 
Did you mean: 

Unable to retrieve items

Maynard_K
Level 6
Employee
Greetings,

I am having issues retrieving items. The R2 queue will backup and the items will not move unless the "Task controller service" has been restarted. I saw another posting where this issue was fixed with 6SP3 but it still continues. The odd thing is that if a dtrace of storagerestore and storageserver are running a restart of the task is not needed and the R2 queue continues to flow. Any ideas are welcome.
16 REPLIES 16

Maynard_K
Level 6
Employee
One interesting point is if I go into task manager and kill all retrievaltask.exe's the items will be removed after the task starts again.

Micah_Wyenn
Level 6
Weird...is the msmq's tossin' any events or a dtrace showing any errors?

micah

Maynard_K
Level 6
Employee
There are no errors being logged for MSMQ. As far as Dtrace shedding any light, it doesn't and it keeps the queues flowing. I would venture to say that by running Dtrace concurrently, the communication channel stays open. Possibly timing?

Micah_Wyenn
Level 6
I just saw something similiar to this in the field. Turns out that a .net2.0 hotfix was needed to bring stability to the host server. If you're running 2.0, i'd give it a shot.

http://support.microsoft.com/kb/913384

Erik_Kuipers
Level 3
I have the same problem on our second EV server but it is running .NET 1.1.

Can someone tell me what is best practice: using .NET 1.1 or .NET 2.0?

Micah_Wyenn
Level 6
7.0 has .net2.0 as a pre-req now i believe...so if you plan on upgrading it might not be a bad idea to install it. Of course, even if you do the upgrade, the earlier versions will continue to use the .net version they're preconfigured for. So 6.0 will always use 1.1

micah

Erik_Kuipers
Level 3
Thanks Micah for clearing this matter.
Still leaves a question open why the retrievaltask.exe hangs randomly and freezes the according R2 queue.
Dtrace doesn't give any clues as does the app even logs.

I am upgrading to SP3 tonight but if the problem remains, I have to find a solution for this issue.

Micah_Wyenn
Level 6
Yeah,
I've got a client right now who's seeing something similiar. What I think is going on, is that one of the services may be leakin'. In our case, it seems to be the admin service (whic h kills us everytime it happens, which is nearly nightly). I'm not sure how to fix it, but I figure with enough evidence maybe we can troubleshoot it better.

micah

Erik_Kuipers
Level 3
Update:

Yesterday I upgraded to SP3 and it looks like the retrieval problem has getten worse. We had to kill the RetrievalTask.exe 3 times this morning already.
We have very little issues with Archiving but Retrieving is not stable at all.

I logged a call at Symantec Support.

MarkBarefoot
Level 6
Employee
Have you thought about DTRACEing the retrievaltask only??

Does this just happen randomly or at any specific times??

Erik_Kuipers
Level 3
Hi Mark,

My apologies for the late reply. I was absent last week.
Like experienced by Maynard, if I let Dtrace running constantly, the R2 queues keep flowing .

Erik_Kuipers
Level 3
Update:
The r2 (retrieval) queues also keep flowing setting the diagnostic levels on the Storage Service to minimum. Not setting it to Maximum minimizes excessive logging to the Application event log.

Erik_Kuipers
Level 3
Update:
The r2 (retrieval) queues also keep flowing setting the diagnostic levels on the Storage Service to minimum. Not setting it to Maximum minimizes excessive logging to the Application event log.

EDIT: please remove this message! My apologies for double posting.Message was edited by:
Erik Kuipers

Dennis_Visser
Not applicable
We experienced the same problem on a netapp filer while we run the ev server on w2k3 sp1. EV was version 6.0 SP3

In the local internet zone of the enterprise vault server we added (without http) the enterprise vault servers.

We also upgraded the EV servers to SP4. All running well now.

Paul_Grimshaw
Level 6
Employee Accredited Certified
We currently have 3 customers who are experiencing the same isue as yourself and this particular issue has been escalated to our engineering department.
It may be worth logginga support case so that you can have your case added to this list

Erik_Kuipers
Level 3
To make sure that truning on Diagnostic logging on the Storage service did the trick to keep the R2 queues flowing, we turned off logging and rebooted the server.
Immediately R2 queues stalled again and we turned on logging again.
The only dowside with turning on Diagnostic logging is that it competely fills up the Enterprise Vault Application log. We did not experience any performance downgrade yet.
 
Paul: any progress yet on this issue?