Qlik Sense App not updated after reload

[Katlitipkins]

Hello everyone! We've encountered an unexpected issue. We have a large application (50 GB in RAM) that updates several times a day. However, from time to time, we notice that the reload task completes without errors, but the application properties still show the previous reload date. We have no idea where to look for the cause. It might be related to issues with the file storage where application files are stored, or the autosave mechanism might be overwriting something. Can you suggest where to investigate?

Best wishes,

Eugeny

[mpc]

Hi, 

Which tool do you use ? 

Quite strange, did you try to create a new reload task ? 
Nothing in the reload task log ? Did you change the SetErrorMode in the script ? 

Best

[Bhushan_Mahajan]

@Katlitipkins Check source refresh

[Katlitipkins]

Hi!

source refresh?

ErrorMode is Default (=1).

It’s a really strange issue. There’s nothing in any logs (Engine, Scheduler, Repository). I see the message "App saved", and everything seems absolutely fine, but I still had this problem.

I suspect it might be related to the Qlik Sense database because it happened during the morning hours when many apps are refreshing with fresh data, and a lot of users are logging into their apps. It could be putting a heavy load on the database. However, there’s nothing concerning in the repository logs.

[Katlitipkins]

Hi!

Tool? We use Qlik Sense Client Managed.

Checked it later in the day— the task worked correctly. No problems were reported, but it's still confusing. We have a couple of hundred apps, yet there’s no clue at all—not a single word in the logs. What happened then? Should we monitor such situations on a regular basis?

[mpc]

Ok; if you suspect the Qlik Repository Database overload, maybe you can increase the Database Max Pool : Recommended practice on configuration for Qlik Sen... - Qlik Community - 1711222 (first with the recommended settings and the more).
You can enable debug logs for Rep. and Engine Service. 

[Katlitipkins]

Thank you.

I suspect the database because the issue correlates with the start of the workday, but it could also be related to the file share. A couple of times, I’ve seen situations where the app’s reload date was updated, but the app file itself wasn’t. I found that it was locked on the file share… Maybe this case is something similar.

We set the Max Pool Size as recommended in the article you provided—900 for 8 nodes. However, maybe we should consider increasing it.

I enabled debug logs for the Repository Service, and searching through them was interesting. I tried to identify an autosave event but didn’t have any success. Then, after three hours, some users experienced issues with the visibility of certain apps and streams, so I decided to switch back to INFO level.

When you mentioned enabling debug logs for the Engine, did you mean the System logs?