A mystery event sometime between Nov 1 - 3 seems to have caused many (not all) of my scheduled tasks to stop firing. Not sure if it may be related, but another bug I was dealing with seems to have been resolved at the same time (where the "Reporter" was being populated from the "Created By" user instead). A new update doesn't seem to have been installed.
I checked and all of my scheduled tasks are still enabled.
So I tried to fire the problematic ones manually but nothing happens (no "Fired" notice appears, and no new tickets exist). The routines that were able to fire automatically worked manually.
If I edit a broken one and re-save it (not making any changes), it is able to fire again. However, I have 100's of scheduled tasks and doing this manually is not practical. Plus, this bug (and others like the one above) are really starting to shake my confidence in this plugin.
For a more practical solution I tried reindexing and also upgraded to the latest version and reindexed again. Finally I disabled and reenabled the plugin - no luck.
I'd love to get an answer to this today as our next batch of scheduled tasks (~100 tickets) will fire tomorrow morning and spending my evening manually opening 3-4 pages and renavigating to fix each scheduled task (multiplied by about 300) doesn't sound very fun... At all.
Havey you manipulated of system date/time while testing TheScheduler functionality?
I have identified one problem with firing scheduled issues after date/time manipulation performed for testing purposes. For example when the date was shifted 2 days ahead to test whether scheduled issue will be fired and then such issue was fired I noticed that after I restored correct date (acctual date) such scheduled issue was not being fired anymore. So I think that after firing such scheduled issue for the first time there must be some kind of information set that such issue have been already fired.
I consider this rather unusual. Most of the time date will be the same. On production servers there is really no reason to first set date to let say 10 days formard and then going back.
Could you confirm that date/time in your case was remained untouched?
Thanks for the reply. I'm not the system admin. If we made any change to date it only would have been a matter of hours, not days. Or, perhaps the way our sys admin upgrades JIRA caused the issue - I'm not very familiar with the process.
If that is the case, would a reinstall of The Scheduler resolve the issue, or is there something else that might fix this?
Based on the first dozen I've reveiewed, I believe the problem may be related to scheduled issues with matching Summary fields (althrough the Name field was already different). In other words, this was not a result of upgrading to The Scheduler 3.0 because I did that ~ 1 month ago.
For lack of a response and an upcoming fire of tickets, I resolved the issue (temporarily?) by manually firing every ticket - which spammed and confused my staff. For those that did not successfully fire, I resaved each scheduled task.
Took about 3 hours.
Unfortunately, I'm not sure what caused the issue, and feel unsettled knowing this may strike again.
0 0 12 ? * 2-6
0 0 12 ? * 6
But not all scheduled issues with these triggers were broken. Also, only 1 project turned out to be effected... Even though the settings were identical with another project with similar routines. Even within the problematic project, some routines were broken while exact copies (other than "assignee") were working fine.
Knowing this won't happen again would be great.
And while I've got the audience these improvements (in addition to a main page with quick access to all scheduled routines rather than having to go to each project to find routines) would be amazing:
Connect with like-minded Atlassian users at free events near you!Find a group
Connect with like-minded Atlassian users at free events near you!
Unfortunately there are no AUG chapters near you at the moment.Start an AUG
We're bringing product updates and pro tips on teamwork to ten cities around the world.Save your spot