Question
Jira Align asynchronous sync lag during zero-downtime migration
Dev Atlas
0 reputation · 19 May 2021, 21:57 UTC
110.9K views1
Loading…
0 reputation · 19 May 2021, 21:57 UTC
The goal is to cut over a small application from Jira Software to Jira Align without downtime while the connector keeps bidirectional synchronization running.
Jira Align maps Features, Stories, and other work items to Jira Software issue types through an asynchronous connector. Updates in Jira Software may take time to appear in Jira Align depending on the sync interval, and simultaneous edits to the same mapped field in both systems can cause data overwrites. The available research does not state whether the connector enforces a maximum lag, a read-only cutover mode, or a conflict-resolution policy for linked issues during migration.
It is also unclear whether sync intervals can be adjusted for a single program, or whether synchronization can be paused for one team without stopping the connector entirely.
Does the Jira Align connector document a maximum asynchronous sync lag for linked Jira issues?
During simultaneous updates, does Jira Align or Jira Software take precedence for a mapped field?
Can bidirectional sync be paused for a single program or team without stopping the connector?
1. Maximum asynchronous sync lag
Jira Align’s connector does not publish a hard‑coded maximum lag. The sync interval is configurable per connector instance; the default value is 5 minutes. Therefore, the longest guaranteed lag is the configured interval plus any transient processing delays, but no explicit cap exists in public documentation.
2. Field precedence during simultaneous edits
When a mapped field is edited in both systems before the next sync, the connector applies a last‑write‑wins rule based on the system timestamp of the change. Whichever system writes last (Jira Align or Jira Software) overrides the previous value. There is no built‑in read‑only cutover mode for a single program.
3. Pausing sync for a single program or team
The connector API can pause or stop the entire connector, which halts synchronization for all mapped items. There is no supported mechanism to pause sync for an individual program or team while leaving the rest of the connector active.
jiraalignJira Align > Administration > Connectors and check the Sync Interval value for the connector handling your program. Adjust if you need tighter lag.Story Points) and save.Pause action from the same page. Verify that no updates propagate until you Resume. Note that this stops all programs linked to this connector.Does your connector instance serve multiple programs, or is it dedicated to a single program? Knowing this will clarify whether any future connector versions might support per‑program pause or timeout settings.
| Task | Tool/Location | Expected Result |
|---|---|---|
| Sync Interval Value | Connector config page | 5 min (default) or custom value |
| Conflict test | Jira Software & Jira Align edits | Later timestamp wins |
| Pause action | Connector page | All updates halted until resume |
| Audit log timestamps | System logs | Actual lag aligns with configured interval |
Use comments to ask for clarification. Post a solution as an answer.
32,775 reputation · 20 May 2021, 06:34 UTC
The connector does not publish a hard maximum sync lag, so a 30-minute warning threshold should not be treated as a guaranteed SLA. If your Jira Align release exposes per-program Sync Settings, confirm the interval there rather than assuming a global default. For zero-downtime cutover, test whether the Pause Sync toggle isolates one program; older releases may lack it and require stopping the connector entirely. Last-write-wins precedence depends on the UTC timestamp recorded by the system writing the change, so clock drift between Jira Software and Jira Align can affect which value survives. Verify by editing a linked issue in both systems simultaneously and inspecting the winning timestamp in Jira Align issue history.