Distribution Service Stopping without Reason

[bnichol]

Recently we upgraded to QV 11.2 SR12.

Every Sunday since upgrading to this version (4 weeks) the Server stops processing between 3:30am and 5:00am.  Upon investigation, the server displays as active, but the only option in the service screen is the 'Start' service button.  In the QMC, the message 'Publisher is down' is displayed.

After starting the service, everything is fine until the next week.

I've reviewed all the logs and nothing indicates why or when the service stopped.  The next entry is the 'Initializing QlikView Distribution Service... Please wait..', after the service is started.

Is this a known issue in SR12?  Has anyone else experienced similar symptoms? 

I'm upgrading to SR13 at the earliest availability, but am curious if this will address the issue.  Is anyone seeing this issue in SR13?

Thanks for your support,

B

[marcus_sommer]

Have you looked into the windows event-logs? Quite often you will find there the reason or at least hints to what is happens.

- Marcus

[bnichol]

There was an event from the Service Control Manager source with the following description 'The QlikView Distribution Service service terminated unexpectedly.  It has done this 3 time(s).'

These 2 TermDD source events were recorded a minute apart, prior to the Distribution Service failing...

'The Terminal Server security layer detected an error in the protocol stream and has disconnected the client.'

'The RDP protocol component "DATA ENCRYPTION" detected an error in the protocol stream and has disconnected the client.'

Doesn't really hint at a cause though.

Thanks for the suggestion,

B

[Anonymous]

Probably one for Qlik Support.

[bnichol]

I agree, but they haven't offered any suggestions yet.

[rwunderlich]

Set your service recovery settings to restart on all failures As a workaround.

Rob

[Anonymous]

Maybe contact your Account Manager to see if he can up the priority.  You are effectively System Dead every week.

Rob's suggestion of the auto restart also does sound like a good workaround.

[Peter_Cammaert]

The consistent failure at the same time on the same day every week may be an indicator for the reason. Is a reload task running at that time? Is something else grabbing excessive amounts of memory? Is there enough memory available for the QDS to operate?

Can you switch the QDS Log to debug logging and post the Root_YYYYMMDD.txt.file of the next failure?

Best,

Peter

PS Running QVS11.20SR12 EE without any complications.

[quwok]

Considering the consistency of the issue and the time it occurs, it might be worth checking if there are any other maybe non-QlikView activities happening at the time. Server maintenance, backup, windows update or some other routine out of hours task.

[bnichol]

Peter,

The server does not use significant amounts of memory during these issues, and the QDS has plenty of additional memory it can consume if necessary.

The QDS log is set to debug, but there wasn't any error or shutting down messages.  Presumably, since the service failed unexpectedly, a log entry wasn't generated.

Another symptom I just identified, is a reload task was aborting for several hours prior to each failure.  The task is different each time.  I've identified an instance that occurred during a different day and time, so I don't think this is a specific task related error.

Thanks for your response,

B