User scheduled workflow status moving to complete

[eddieyak]

With an Alteryx user workflow on the server, the schedule is moving from active to complete, even though though the end date is blank and never is also selected.  Below is a picture of the schedule that is moving to complete 

[AngelosPachis]

Hi @eddieyak ,

Do you mind going to the Schedules page in Gallery and attach a screenshot of the record created for this workflow please?

[eddieyak]

[raychase]

Any chance your host machine was restarted? I've seen this happen if a scheduled job is in the queue (not running, but actually waiting in queue) at the time of a hard restart of the AlteryxService.

[eddieyak]

Looking at that, the last reboot was on 10/14 while the jobs got moved to complete on 10/22. 

[mehtas0311]

Hi

We are also facing this issue in our organization.

server version 2020.3

We take a backup every Sunday, and on Monday many of the schedules shows complete.

Please share if there is any root cause information and fix to be applied.

Regards

[raychase]

@mehtas0311 - any jobs that are queued up to run while your AlteryxService is in a 'stopping' state will automatically get marked as complete.

For example, you initiate a backup at 10:00pm, but there are some workflows actively running.  The AlteryxService won't be fully stopped until those active jobs are completed (let's say they run until 10:10pm).  At 10:05pm, another scheduled job was added to the queue.  This schedule will now be marked as complete.

[mehtas0311]

This is strange.

Why should the schedule (with No Ending Date) be marked complete if there is Alteryx Server restart.

cancelling the queued jobs during Server restart is appropriate and I can understand but why the schedule itself is marked as complete.

Once the server is available, these schedules dont run and cause delay in workflow processing.

Is there any workaround such that when there is a server restart the schedule for queued jobs can be made active.

[raychase]

I agree.  I've brought it up with Alteryx more than once.  My workaround has been to include auto-kill functionality in my backup script to cancel actively running jobs.  This effectively shortens the 'stopping' time of the AlteryxService, which reduces the likelihood of the 'completed' schedules bug.  It's not ideal, but it works for us.

[mehtas0311]

hi @raychase 

Could you share the command you use to auto-kill active jobs.

Thanks @raychase