We have a workflow step with a timeout of 1 hour that uses the Auto Advancer. This is because we have a step that needs to check on a remote service every hour to see if an external process has completed. What we're experiencing is this works fine for a while but will stop working intermittently. It stops polling. The workflow history shows that it just quit polling. Sometimes it goes days without breaking, sometimes it goes only a few hours. There are no errors in the logs that we can see.
I'm trying to reproduce it in a dev enviornment and I tested various scenarios like throwing exceptions in our workflow execute() methods but that doesn't reproduce this behavior for us. It simply reports the workflow exception message as the subtext for the job in the inbox.
Does anyone have any ideas or seen similar behavior?
Thanks in advance.
Not sure but it looks like your sling eventing is getting messed up some time. What version of CQ are you using ?
Thanks for the response. The CQ version is 5.4. I'll look into debugging options for sling eventing, any more ideas are appreciated.
I you are able to reproduce this issue in local, Then I would suggest to run same code in CQ5.5 update1 (Or get latest sling eventing bundle) to see if it is core issue. If that is the case, You can raise issue with daycare to investigate further.