Skip to main content
Announcements
Have questions about Qlik Connect? Join us live on April 10th, at 11 AM ET: SIGN UP NOW
cancel
Showing results for 
Search instead for 
Did you mean: 
steeefan
Luminary
Luminary

Task finishes successfully, is registered as "Failed" (SaveApp caught extended exception)

I'm having trouble with our QS Prod server. Tasks that have actually finished with the message "Execution finished." are still registered as "Failed" in the QMC. This is of course a major problem with regards to jobs that should follow the one that just failed, especially during nightly reloads.

Using this guide, I was able to trace the the error:

5974    20220607T085505.289+0200    DEBUG   SERVERNAME  System.Engine.Engine    272     DOMAIN\qlikuser ExtendedException: Type '-1' thrown in file 'C:\Jws\engine-common-ws\src\ServerPlugin\Plugins\PluginApiSupport\PluginHelpers.cpp' in function 'ServerPlugin::PluginHelpers::ConvertAndThrow' on line '138'. Message: 'Unknown error' and additional debug info: 'File error for file \\internal.domain\Data\00_Qlik\Qlik_Share_Prod\Apps\Transactions\SERVERNAME.internal.domain\AAAAAAAA-BBBB-CCCC-DDDD-EEEEEEEEEEEE, message: sharingViolation \\internal.domain\Data\00_Qlik\Qlik_Share_Prod\Apps\Transactions\SERVERNAME.internal.domain\AAAAAAAA-BBBB-CCCC-DDDD-EEEEEEEEEEEE'

5975    20220607T085505.305+0200    ERROR   SERVERNAME  System.Engine.Engine    272     DOMAIN\qlikuser AppSerializer: SaveApp caught extended exception -1: Unknown error

Event-blurred.png

As can be seen, the Qlik Sense apps are stored on a network share and not on the local Qlik Sense server. We already excluded the "Transactions" folder from our virus scanner but to no avail, tasks still keep failing although they were actually finished successfully. It appeared for the first time last Sat after two servers were rebooted after patching, QS Prod and the file server where these files are hosted. I was told that the patches were not in regards to file management or virus scanning which I did also verify myself.

Traffic on that folder "Transactions" is very high so I assume the file server sometimes has problems keeping up with writing files and then deleting them again immediately while they still might be locked / not fully transferred or written to the file server's disk.

Has anyone dealt with the same behavior before? Would you recommend moving at least the "Apps" folder back to the local disk of the server to avoid locking / sharing violations with processes on the file server?

 

Server specs:

  • Windows Server 2019 Standard (x64)
  • Intel Xeon Gold 6254 @ 3.10GHz (32 vCPUs)
  • 156gb RAM
  • Qlik Sense Feb 2022 Patch 3
Labels (1)
1 Solution

Accepted Solutions
Albert_Candelario

Hello @steeefan ,

Kindly install the Feb 2022 Patch 4.

For such issue we made an improvement in the Engine and added a retry mechanism where a locked transaction file would cause Engine to fail when saving the app.

Using - How to troubleshoot the AppSerializer: SaveApp cau... - Qlik Community - 1930186 as you mentioned you should be able to see other process interacting with such folder, sometime the antivirus is excluded by the Windows Defenders online processes are still active. We  have seen some of such situations.

Cheers,

Albert

Please, remember to mark the thread as solved once getting the correct answer

View solution in original post

1 Reply
Albert_Candelario

Hello @steeefan ,

Kindly install the Feb 2022 Patch 4.

For such issue we made an improvement in the Engine and added a retry mechanism where a locked transaction file would cause Engine to fail when saving the app.

Using - How to troubleshoot the AppSerializer: SaveApp cau... - Qlik Community - 1930186 as you mentioned you should be able to see other process interacting with such folder, sometime the antivirus is excluded by the Windows Defenders online processes are still active. We  have seen some of such situations.

Cheers,

Albert

Please, remember to mark the thread as solved once getting the correct answer