Does Jira Align’s sync retry mechanism prevent duplicate writes during transient failures?

Answer

No, Jira Align’s sync retry mechanism does not inherently prevent duplicate writes; it relies on a last‑write‑wins (LWW) timestamp comparison rather than unique transaction IDs, so a retry after a transient failure can re‑apply the same change.

Confirmed facts

• The sync engine uses LWW based on detected timestamps to decide which version wins.
• No unique transaction identifier or version token is attached to each change attempt for idempotency checks.
• If a transient failure occurs after the target record is updated but before the acknowledgment is received, a subsequent retry may see the same timestamp and re‑apply the update.

Steps to verify in your environment

1. Pick a non‑critical field that is synced bidirectionally.
2. Introduce a brief network interruption or timeout during a sync cycle while modifying that field.
3. After the sync resumes, check the History tab in Jira Software for multiple entries showing the same value change.
4. Compare the Updated timestamps in Jira Align and Jira Software immediately after the retry; identical timestamps with a new history entry indicate a duplicate write.

Missing diagnostic detail

Whether the system clocks of Jira Align and Jira Software are tightly synchronized affects the reliability of the LWW logic; if clocks drift, the retry may incorrectly treat an older update as newer and cause overwrites. Please confirm if clock synchronization (e.g., NTP) is enforced between the two instances.