Additional Event Options: Like "when workflow hangs" or "if workflow fails to run"

[patrick_mcauliffe]

I utilize the Workflow Events quite a bit, but the standard options fall short in a few critical cases.

For example: I use the events to send me an email after the workflow runs with errors.

Usually that is all I need.  But, I have run into a few situations where the workflow didn't run at all - thus, no email was sent.

One such instance was on our server.  While still initializing (before it even hit tool #1) the job hung.  It sat like that for over a day before I noticed that there were no jobs running through a particular worker at all and saw this one sitting there. 

A second such instance was recently when I changed Persistence Options and accidentally deleted the source for a scheduled job.  The workflow marked itself complete in the scheduler and gave an error message of "Corrupt: Internal Error in PersistenceContainer......"

I realize that these are unexpected events and application errors (which I've worked with support on); but it would add great value to have that fallback mechanism to let us know the error happened.

Understandably, it would probably take a lot of work to engineer a schedule monitor that would catch this - but I think it would be worthwhile.

One option to do this: have an Event to email or run a command if a workflow hasn't started within a certain grace period after its scheduled time.

[ThizViz]

Agreed! I ran into a situation where I "broke" a workflow and what used to take less than an hour still wasn't done overnight. I'd love to see a timeout feature incorporated. 

[awilkey]

---

---

The workflow marked itself complete in the scheduler and gave an error message of "Corrupt: Internal Error in PersistenceContainer......"

I've experienced this as well...the job doesn't fail, but rather it simply isn't run - so there is no entry in the 'results' tab of the scheduler - and also no indication whatsoever that anything is amiss, unless you happen to expand the original workflow schedule...

[SeanAdams]

Interesting idea - thinking about it this may need to be a server event rather than a workflow event (since the workflow doesn't know that it didn't run; if didn't run; but the server / gallery would be in a position to know this)

[Joe_Lipski]

Another ACE stumbling across this issue...

I have a workflow being executed each morning by cmd and as part of the workflow it executes a batch file to call tabcmd. Intermittently this batch file hangs at different steps of the cmd and the workflow doesn’t fail or error, just sits there. This has also happened when I’ve ran directly in the designer interface hanging at 51% on the macro.

Does anyone have a solution to make a workflow fail after X amount of time?

[levell_x_dunn]

Yeah, I have a few jobs now stuck in "Initializing" status. Which as the admin, I can't cancel or remove without restarting the entire system, which is also not an option during business hours.

Server Admins need more control over our own servers and notifications. Jobs stuck since yesterday afternoon

[patrick_mcauliffe]

@levell_x_dunn  you can still cancel those jobs without restarting the server.  You'd have to terminate them from the scheduler window on Alteryx Desktop.

[levell_x_dunn]

@patrick_mcauliffe 

Unfortunately, because they never officially start and are stuck in initializing, you aren't able to cancel them on the scheduler via Desktop. 😕

[LNGRIFL]

@levell_x_dunn  or @patrick_mcauliffe 

Has anyone found a way to delete the jobs stuck in initializing OTHER THAN restarting the service?  We cannot bring the service down on Monday thru Saturday.