Board Issues API not returning all issues for a board

[Megha Mittal]

Hi there, 
I am using the Jira Agile API to fetch all issue keys from a board:

GET /rest/agile/1.0/board/{boardId}/issue?fields=id,key

The endpoint returns the issues correctly but not all, it is not returning certain issues that are clearly visible under an Epic.

These same missing issues do appear when I call the Epic endpoint:
GET /rest/agile/1.0/epic/{epicKey}/issue


So the issue exists, is linked to the Epic, and I have permission to view it — but the /board/{boardId}/issue API does not return it.

I need clarification on why this is happening and how to make sure the Board Issues API returns all issues that belong to the board, including those under epics.

Questions:

• Under what conditions does /board/{boardId}/issue exclude issues?

• Is this caused by board filter JQL?(also all the status are mapped in the column settings)

• What is the correct way to fetch all issue keys that belong to a board?

Thank you.

[Trudy Claspill]

Hello @Megha Mittal 

Welcome to the Atlassian community.

Are you working with a Scrum board or a Kanban board?

Are you working with Jira Cloud or Jira Data Center? I see the tags on your post say Deployment=Cloud, but there is also a tag that says "data center".

Per the documentation for the Cloud endpoint the issue returned are all issues within the scope of the board that meet these criteria:

1. This only includes issues that the user has permission to view.
2. An issue belongs to the board if its status is mapped to the board's column.
3. Epic issues do not belongs to the scrum boards.
4. Note, if the user does not have permission to view the board, no issues will be returned at all.

It does not explicitly state it, but I would assume that another criteria is that the issue is within the scope of the board's filter.

When you examine the issues missing from the API output are those issues within the scope of the board's filter?

There is also a limit to the number of issues that can be returned by the endpoint. If more issues exist than that, you have to retrieve them a page at a time:

maxResults

integer

The maximum number of issues to return per page. See the 'Pagination' section at the top of this page for more details. Note, the total number of issues returned is limited by the property 'jira.search.views.default.max' in your Jira instance. If you exceed this limit, your results will be truncated.

[Megha Mittal]

Hello @Trudy Claspill

First of all, thank you for replying promptly.

I am working with a Scrum board and using Jira Cloud.

The user I am using has the necessary permissions.The issue keys that are not visible via the API have their statuses correctly mapped to the board columns (I have verified this under Board Settings → Columns).

My understanding from your response is that Epic issues themselves do not appear on the Scrum board - but the child issues (stories, tasks, bugs) under an Epic should appear, as long as they match the board filter and have mapped statuses.

However, the missing issues in my case are children of Epics, and they are within the scope of the board filter and have mapped statuses, yet they still do not appear when I query:

 
GET /rest/agile/1.0/board/{boardId}/issue

I am aware of pagination and am using startAt and maxResults to page through all results. Could you clarify why these Epic-child issues might be excluded from the Board Issues API?

Thank you!

[Trudy Claspill]

I don't have any projects with that many issues to test against, so I will only be able to ask questions and offer untested advice.

Child issues of Epics should be returned by the API. I have confirmed that by testing against my own instance.

Do the issues missing from the API results display on the board or in the backlog? Are they issues that are still incomplete or issues that have already been completed in some previous sprint and no longer display in either the backlog or board?

Are sub-task type items included in the results?

How many issues are you actually able to get through the API? There is a limit on boards, at least in the UI display, that no more than 5000 issues can be shown. If you are only able to get up to 5000 issues then the API may be subject to that limit also.

[Megha Mittal]

• Missing issues are displayed on the board (More specifically, on the all issues tab, where the board filers are automatically applied, there I have searched the missing issue, and it is present there.
• These missing issues are typically are of type BUG and has status cancelled, and some of them was never been assigned to any team and sprints. 
• Yes, sub-task type issues are coming in the API response. 
• Through API , I am getting around 8k issues. On the board (under all issues tab), in UI it is showing around 1k issues.

[Trudy Claspill]

Thank you for that additional information.

I thought that I read in your post that you have about 8k of issues within the scope of your board, but now I don't see that information.

You said in the API you are getting "around 8k issues", but that you are not getting all the issue, even considering that Epics won't be included.

Are you able to access all the 8k issues returned by the API using pagination?

How many issues, excluding Epics, are within the scope of the board filter? On the All Work tab only 1000 issues will be displayed but there could be more in the project. You will see "1000+" at the bottom of the list. If you click on that it will show you the actual total number of issues. Make sure to change the filter to exclude Epics.

[Megha Mittal]

Yes I am able to access all the 8k issues returned by the API using pagination. 

And, there are around 11k records in the scope of board filter (excluding the epic issue type).

Also, on my scrum board, the tab says All issues (instead of All Work ), is it a normal behaviour to have?

[Trudy Claspill]

The difference in tab name may just be a delay in roll out of the terminology change from "issues" to "work item" for your instance.

At the moment the only other limiter I can think of is the one I pasted in my first response concerning maxResults where it mentions truncating the results based on the property 'jira.search.views.default.max'

I'll keep digging to see if I can find another explanation.