Skip to main content
Announcements
NEW: Seamless Public Data Sharing with Qlik's New Anonymous Access Capability: TELL ME MORE!
cancel
Showing results for 
Search instead for 
Did you mean: 
Frode
Contributor II
Contributor II

QMC failing jobs , error message (System error: The handle is invalid.)

We are running QV 12.40 April 2019 SR3 on a cluster with five servers , two for Distribution Service sharing the load. We have been experiencing intermittent failing jobs with the above error message. We initially ran "Max number of simultaneous QlikView engines for distribution:20" but lowered it to 7 and we saw less failing jobs. But since a while back we are experiencing (System error: The handle is invalid.) again , Distribution Services have been restarted after system has been hanging and then jobs have executed orderly. Today in QMC , ten jobs have failed with  (System error: The handle is invalid.) , and 37 jobs are in Queued , but when checking logfiles , they have finished successfully. Checking the DS servers , there are 141 processes and 4430502 handles on one server and 189 processes and 4890779 handles. Any suggestions of what to look for? Any configuration? 

 

4 Replies
Chip_Matejowsky
Support
Support

Hi @Frode,

Typically there is more information in the QDS Task log when Error: Connection aborted (System error: The handle is invalid.) is returned. Can you post the full QDS Task log for the failed task, or at least a larger snippet of the log showing the error?  Does the task fail each time it is run? Does it fail intermittently?

Regarding running more than 7 "Max number of simultaneous QlikView engines for distribution", have a look at the Qlik Support article How many reload engines (QVB) to allow in an environment as it details the considerations you need to understand when adjusting this setting.

Best Regards

Principal Technical Support Engineer with Qlik Support
Help users find answers! Don't forget to mark a solution that worked for you!
Frode
Contributor II
Contributor II
Author

The jobs fail intermittently, it's not the same jobs that fail (usually) and we've had to restart one server more often than another.

Frode
Contributor II
Contributor II
Author

 
Frode
Contributor II
Contributor II
Author