Alteryx Server Issue: Last Run Suddenly Changed to Never and State is Complete

[ashleychen2737]

Hi All,

My team and I have been running into sporadic issues when scheduling our workflows through Alteryx Gallery. From time-to-time our stakeholders will report that dashboards and reports have not updated, and when we do some research it appears that the scheduled workflow's Next Run suddenly changes to "Never" and the state goes from "active" to "complete". We have not been able to identify why this happens, and why only certain workflows. This problem began occurring when we upgraded to 2018.1. Has anyone run into this issue, or better yet, have a solution for it?

[StephenRuhl]

Do those scheduled workflows have an End Date to the schedule?

 

[ashleychen2737]

The end dates for the workflows are not set. They are also scheduled to run hourly on the Alteryx Server.

[cdgogan]

Hey ashleychen2737,

Did you get any further correspondance on this issue?  We're currently running Alteryx Server 2018.4 and I've received a similar report from one of our users who schedules a workflow on an hourly interval as well.  They too had no End Date configured and claim their scheduled job abruptly stopped and that they hadn't made any changes.

Cheers

[tmclaughlin]

We also are experiencing this issue under 2018.4 and have opened a case with support which has been open for weeks with no solution.  I would like to see some escalation of this issue as it can be very time-consuming to reset all of the schedules affected by the change in status.  It almost appears to be connected to the restart of the Alteryx Service.  Some weeks are fine and other weeks when the service is restarted we see the issue.  If we get an answer from support I will post on this thread.

[raychase]

Any update on this?  I've seen it happen on several versions of Alteryx Server.

[tmclaughlin]

No fix from support yet.  Can't get them to respond.  I must be the only customer using the gallery scheduling.  🙂

[raychase]

Not sure if this helps your situation, but there is a known case where jobs that are in queue while the server is suddenly rebooted will flip from active to completed.

This isn't something I was aware of until recently.