Hello,
We are using version Data center, the structure of our projects in jira has two ways:
1) EPIC -> Sub-Task
2) EPIC -> Story -> Sub-Task
and I noticed that every time someone logs time/adjusts the original estimation on level of the sub-task is not being updated on the EPIC level. I managed to automate the first scenario where in case of an adjustment on the Sub-Task level automation will update the parent. This works well.
I struggle with the second scenario, because the automation only updates the story, not the sub-task.
What would you suggest to implement?
Thank you and wish you an amazing day! :)
It sounds like you've already got a good handle on the simple Epic -> Sub-task scenario, so kudos! The tricky bit is that Epic -> Story -> Sub-task structure.
Here's what I'd suggest, speaking as someone who's wrestled with Jira automations a few times:
You're on the right track with automation, but it needs a little tweak. The problem is it's stopping at the Story level and not going all the way up to the Epic. Think of it like a relay race – you need to pass the baton (the updated time/estimate) one more leg.
The most elegant way to do this is with "Advanced branching" in your automation rule. If you have that feature available, it's like having a smart traffic director. It checks the issue type and sends the update to the right place.
So, your rule would look something like this:
Imagine it like this: The Sub-task whispers its updated time to the Story, and then the Story shouts it up to the Epic.
If you don't have advanced branching, you can still do it, but it's a bit more like using duct tape and string. You'd have to use smart values and conditional logic within a single "Edit Issue" action. It's doable, but it gets messy fast, and it's harder to maintain down the road.
One other thing to watch out for is circular updates. Make sure updating the Epic doesn't trigger the rule again and cause an endless loop. That's like a dog chasing its tail – lots of activity, but not much progress.
If you're comfortable with scripting and have ScriptRunner, that's another option. It's a bit more powerful, but it requires some coding.
Finally, always test this on a staging environment first. You don't want to accidentally mess up your production data.
Let me know if that makes sense.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.