How to clone an epic with associated stories to another space while keeping the order of the stories

Question

Based on an example in the Jira automation template library I created an automation rule that works as intended…. almost.

What I want is to keep the order of the associated stories under the epic, but when I run the automation rule, it clones the epic as intended and also clones the associated stories, but in a random order - not what I want.

What must I change in my automation rule to preserve the order?

Arkadiusz Wroblewski · Accepted Answer

Hi @Paul Schluck

Atlassian explains the limitation here:

https://support.atlassian.com/cloud-automation/docs/branch-automation-rules-to-perform-actions-on-related-issues/

Either you must rerank or use Rest Api

Cheers,

Arek🤠🧙‍♂️

Gabriela - LeanZero · Accepted Answer

Hi Paul,

You lose the order because the clone branch runs on all your stories at once, and the branch doc says "Branches on multiple work items (such as 4 sub-tasks) will run in parallel with no guarantee one will finish before the next one starts." When I put a Lookup with ORDER BY Rank in front of the branch, the lookup came back in order, but the branch over it still ran in parallel and the clones landed at the bottom of the target in random order.

What worked for me on a test site was Arkadiusz's rerank, done after the clone in a second manual rule. Your first rule runs on the source epic, clones it, does a Lookup with parent = {{issue.key}} ORDER BY Rank ASC, and then clones each story in a branch with this in More options, so every clone keeps its source position in a number field you pick:

{"fields":{"parent":{"key":"{{createdIssue.key}}"},"customfield_NNNNN":{{lookupIssues.key.join(",").concat(",").indexOf(issue.key.concat(","))}}}}

The comma on the end matters, because without it a key like ABC-4 matches inside ABC-46 and the two swap places. I checked that one on the test site too.

Your second rule runs on the new epic, does a Lookup ordered by that number field, and then sends a web request PUT to https://your-site.atlassian.net/rest/agile/1.0/issue/rank with a secure Authorization header set to Basic and the base64 of your email:API token, Content-Type application/json, and this body:

{"issues":[{{#lookupIssues}}"{{key}}"{{^last}},{{/}}{{/}}],"rankAfterIssue":"{{issue.key}}"}

I got a 204 back and the stories came out in source order, and you can send up to 50 issues per call according to the rank endpoint doc. Editing the Rank field directly through an edit action won't help you here, since I saw it return 204 and change nothing.

For a one-off, dragging the stories into order once by hand in the new epic does the same job, and you can also vote on JRACLOUD-93667. Is your target space company-managed with a board, and how many stories does a typical epic have?

Sarah Marenchino | Devoteam · Accepted Answer

Hi Paul, I agree with @Arkadiusz Wroblewski and @Gabriela - LeanZero that the parallel execution of automation branches is the key limitation here. One additional point regarding the proposed configuration: I would avoid relying directly on {{createdIssue.key}} as the new parent inside the story-cloning branch. Atlassian documents that subsequent clone actions can change the value of {{createdIssue.key}}. The recommended approach is to: * Clone the Epic. * Immediately store the new Epic key in a variable, such as NewEpicKey. * Use {{NewEpicKey}} as the parent when cloning each Story. This ensures that the cloned Stories reference the intended Epic, independently of the order in which the branches execute. Atlassian Support: Clone an Epic with its child issues For the ordering itself, a separate re-ranking step using the Jira Agile REST API is a possible approach, but I would validate the final order on the destination board before treating the automation as reliable.