Priority not assigned based on automation

[Alex]

Hi all,

I created a very simple priority assignment rule based on impact field. This part worked for a few days but then stopped. Please see attached screenshot for explanation. The incident submitted two days ago by same person/same form/same fields worked (highlighted in green above), however, since yesterday the match is no longer found (highlighted in red).

Anyone had any idea why it stops working? Note that I had another more complex rule initially implemented which worked for a few days and then stopped for no reason. I thought that logic was too complex so I created a plain rule which worked for a few days and then stopped as well. Any ideas what's goin on? This is a simple 1 input field and I had no changes to anything. I also set 5 seconds delay to re-fetch the data just in case. 

Totally at loss what's going on and what am I missing. 

[Walter Buggenhout]

Hi @Alex and welcome to the Community!

Is there any reason why you have a Refetch issue data action in your rule? You might have of course, but just looking at your rule I can't really figure out why you would need that here. This may impact the data that's on your issue, so I would suggest to remove this and see what happens.

Hope this helps!

[Alex]

Hi Walter, thanks for your reply. I have no idea why it's there but this was a part of the schema used by our corporate to assign priority. This schema works for corporate team, but not for me. 

Anyway, I removed that part and will monitor what happens. 

[Alex]

Just received a few new incidents. Still doesn't work even with data re-fetch deactivated. 

Not sure what's going on. I had two rules and both of them worked for a few days and then, suddenly, stopped, the rules are based on a single field.

I guess the field passed via the form is not the same as used by automation? The name is the same though but for some reason it doesn't accept the value passed in it. 

[Walter Buggenhout]

Aha, ok. Is this an advanced form where you can decide to map the form field to a Jira field (one thing to check). And if so, is it mapped correctly.

Next, do you have multiple custom fields with the same name in Jira and indeed - in this case you are referencing the wrong one?

Or is there only one field, but does it have different contexts configured for different projects?

All these options can indeed impact the data being different than what you expect.

[Alex]

Yes, I'm mapping the form field to Jira field. The field seem to be the same as I don't see any other options with the same name. See attached. 

Could you please elaborate on the context for different projects? I do see this field set up in Fields section but I can't dive deeper into that part as I have no access, that seem to be managed by our corporate jira administrator.

[Walter Buggenhout]

Could you please elaborate on the context for different projects? I do see this field set up in Fields section but I can't dive deeper into that part as I have no access, that seem to be managed by our corporate jira administrator.

That is correct, that part is indeed managed by your Jira admin. But since you can see the options of the field displayed, I would guess that is not the issue.

I did initially not pay attention to the fact that you're using an advanced form to capture the data. It could be that the refetch issue data action we talked about earlier is actually there to make sure that the data that is coming through the form is already saved to Jira before performing any actions on it. It could be that there is some sort of delay happening there (somewhat unpredictably) so the data is not saved to the database yet when your automation rule is trying to process it.

While I think this may be a probable cause, I don't really have immediate ideas on how to diagnose or work around that issue. In automation for cloud, a delay action (see this community post) was introduced and the request is also described in feature suggestion AUTO-238. While I don't have a data center instance available myself, everything I read seems to suggest that this delay option is not available there. In the AUTO-238 suggestion, the workaround is proposed to add a few consecutive refetch issue data actions after another. It is a pretty ugly workaround, but it might help determine if this really is the issue.

If it isn't, I'm about to run out of inspiration. In that case, you might want to work with your corporate Jira Admin to see if the Jira logs disclose more details about what is going wrong and work with Atlassian support from there to get this resolved. 

[Alex]

I'll take ugly automated ugly workaround over manual one anyday :)

I've added three re-fetches. Let's see what happens now. 

[Alex]

Ok, still doesn't work even with three re-fetches. I'm checking with corporate admins, looks like there is something else in the background which I'm either missing or can't access. 

[Alex]

Just an update on this. There was only 1 option which didn't work so I had to rework the logic around that to assign prio for lowest severity in a different manner. I have no idea why 2 options from the same exact field work as expected and another one is not assigned. I reworked the logic and it now works. No idea what's wrong with that particular option, probably something in the back end.

Thanks for your help .