SamuelMuellerKMS opened a new issue, #9405: URL: https://github.com/apache/devlake/issues/9405
### Search before asking - [x] I had searched in the [issues](https://github.com/apache/incubator-devlake/issues?q=is%3Aissue) and found no similar issues. ### What happened Since upgrading to a version that contains #8931 ("collect issues via search API to avoid kanban sub-filter exclusion"), no new rows are written to `_tool_jira_sprint_issues` / `sprint_issues`. Issues and sprints are collected correctly, but the relation between them is missing: - All sprints started after the upgrade have **zero** issues (every active sprint in our instance). - Issues collected before the upgrade keep their old sprint relations, which become stale when issues move between sprints. - `_tool_jira_issues.sprint_id` is `0` for all issues. Root cause: - `issue_collector.go` now uses `api/2/search` (Server) / `api/3/search/jql` (Cloud) with `filter = <boardFilterId>`. - `apiv2models/issue.go` → `Issue.ExtractEntities` still builds the sprint list **only** from `fields.sprint` and `fields.closedSprints`: ```go if i.Fields.Sprint != nil { sprints = append(sprints, i.Fields.Sprint.ID) } for _, sprint := range i.Fields.ClosedSprints { sprints = append(sprints, sprint.ID) } ``` - `fields.sprint` and `fields.closedSprints` are only returned by the Agile API (`agile/1.0/board/{id}/issue`). The search API returns sprint membership only through the Sprint custom field (schema custom `com.pyxis.greenhopper.jira:gh-sprint`, e.g. `customfield_10004`). Raw data comparison from our instance (`_raw_jira_api_issues`): | collected | URL | `fields.sprint` / `fields.closedSprints` | Sprint custom field | |---|---|---|---| | before upgrade | `rest/agile/1.0/board/98/issue?...` | present | present | | after upgrade | `rest/api/2/search?expand=changelog&fields=*all&jql=filter = ...` | **absent in 100% of rows** | present | On Jira Server/DC the custom field value is an array of legacy strings: ``` ["com.atlassian.greenhopper.service.sprint.Sprint@14e33bf4[activatedDate=2026-11-04T13:44:13.760+01:00,autoStartStop=false,completeDate=<null>,endDate=2026-11-18T13:30:00.000+01:00,goal=,id=2424,incompleteIssuesDestinationId=<null>,name=Team A Sprint 42,rapidViewId=148,sequence=2376,startDate=2026-11-04T13:30:00.000+01:00,state=ACTIVE,synced=false]"] ``` On Jira Cloud it is an array of objects with an `id` property. ### What do you expect to happen Sprint–issue relations are extracted for issues collected via the search API, i.e. `_tool_jira_sprint_issues` and `sprint_issues` contain every sprint listed in the issue's Sprint field, as before #8931. ### How to reproduce 1. Jira Server/DC connection (likely Cloud as well), board scope with active sprints. 2. Run a blueprint with a DevLake version that includes #8931. 3. `SELECT COUNT(*) FROM _tool_jira_sprint_issues WHERE sprint_id = <active sprint id>;` → `0` 4. Check the raw issue: `JSON_EXTRACT(data, '$.fields.sprint')` is `NULL`, while `JSON_EXTRACT(data, '$.fields.customfield_<sprint field id>')` contains the sprint. ### Anything else Suggested fix: in the issue extractor, resolve the Sprint custom field id from `_tool_jira_issue_fields` (`schema_custom = 'com.pyxis.greenhopper.jira:gh-sprint'`), the same way story point / epic link fields are resolved, and parse it from `Fields.AllFields` when `fields.sprint` / `fields.closedSprints` are empty: - element is an object → take `id` - element is a string (Server legacy format) → parse `,id=(\d+),` Note: the Sprint custom field already contains both closed and current sprints, so it can fully replace `sprint` + `closedSprints`. Workaround we use in the meantime: a SQL script run after each sync that parses the Sprint custom field from `_raw_jira_api_issues` and fills `_tool_jira_sprint_issues` / `sprint_issues`. ### Version v1.0.3-beta19@210dcd4 ### Are you willing to submit PR? - [ ] Yes I am willing to submit a PR! ### Code of Conduct - [x] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
