Skip to main content
Announcements
Accelerate Your Success: Fuel your data and AI journey with the right services, delivered by our experts. Learn More
cancel
Showing results for 
Search instead for 
Did you mean: 
bjendrick
Contributor III
Contributor III

QDF 1.7.1 w/ QS November - Tasks Failing

I have an app I'm testing to update hourly.  When I load data into the app as myself (as a RootAdmin), everything is 100% successful.  When the task runs, which is under the internal sa_scheduler account, the task fails.  I've used $(Must_Include) to force a failure if the file isn't found, and that is where the error is thrown.  I've added the "sa_*" accounts to our global data connections through security rules, but it's still failing.  What might I be missing to enable the job to run successfully?

Please help... thank you.

1 Solution

Accepted Solutions
Magnus_Berg
Employee
Employee

Looking at the logs I think the problem is that the service account can't fine Q:\git\qlik_qdf\_core\debug_on.qvs a mounted drive q: is probably not available for this service account.. Use UNC path instead \\servername\sharename\git\qlik_qdf\_core\debug_on.qvs

Hope that this helps

Regards

Magnus

View solution in original post

6 Replies
bjendrick
Contributor III
Contributor III
Author

After much reading, I think this is related to not running the reload server in "Legacy Mode" as "Standard Mode" is still active.  I'll include links below, and once I confirm this works, I'll report back.

File system access restriction:

https://help.qlik.com/en-US/sense/November2017/Subsystems/Hub/Content/LoadData/file-system-access-re...

Disabling standard mode:

https://help.qlik.com/en-US/sense/November2017/Subsystems/Hub/Content/LoadData/disable-standard-mode...

bjendrick
Contributor III
Contributor III
Author

Changing the Standard Mode setting in Qlik Sense didn't impact the ability for the scripts to run as an automated task.  Has anyone else run into this kind of situation?  We have a multinode system, with a shared single drive between the instances.

What I've tried:

  • I've changed the folder data connection from a "Q:" drive to using the UNC path.
  • I've added the sa_scheduler account as a RootAdmin.
  • I've even attempted to give explicit access to the drive from the qliksenseservices account in Windows.

None of this works.  Anyone have any ideas?

Magnus_Berg
Employee
Employee

Hi Ben, sorry for late reply. I do not think changing modes would impact, I believe its security related, that running manually you run as your self, when scheduling you run as sa_scheduler  Could you attach the scheduled reload log files? It should be available for download under the scheduled task.

Clipboard01.jpg

Best regards

Magnus

bjendrick
Contributor III
Contributor III
Author

I thought it might have been security related as well, however, I have set sa_scheduler as RootAdmin, I have allowed the connections to be globally available to all sa_* users, and I've even confirmed that the drive itself has permissions for the qliksenseservices user.  I've enclosed the log file while is entirely unhelpful in this regard.

Magnus_Berg
Employee
Employee

Looking at the logs I think the problem is that the service account can't fine Q:\git\qlik_qdf\_core\debug_on.qvs a mounted drive q: is probably not available for this service account.. Use UNC path instead \\servername\sharename\git\qlik_qdf\_core\debug_on.qvs

Hope that this helps

Regards

Magnus

bjendrick
Contributor III
Contributor III
Author

Yes, yesterday with our consultant, we determined that some of our folder connections use the IP, and others use the Q: designation.  As soon as I changed them all to IP for the UNC convention, everything started working.  Also, Qlik Support was no help in this process either.  Be aware that their knowledge is hit or miss at times.

Thank you for your help!