Retry Transactions on Deadlock with Jitter
Detect a deadlock error, wait with jitter, and retry the transaction so concurrent operations do not permanently fail.
Task
Apply catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); } so the schema change is safe, reversible, and does not lock the table during peak traffic.
When to use this
A migration deploys catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); } during peak hours, takes a 40-second lock, and causes a connection pool exhaustion.
Prerequisites
- A database with migration tooling configured
- A staging environment mirroring production schema
Steps
1. Configure catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); } in the project configuration file with the recommended defaults.
catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); }
Expected output: The configuration is accepted and catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); } runs without errors on the first invocation.
2. Run the check locally to verify catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); } produces the expected result before pushing.
pnpm test
Expected output: Local run matches CI behavior; no surprises after push.
3. Add the gate to the CI pipeline so catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); } blocks merge on failure.
- name: Run add transaction deadlock retry
run: pnpm test
Expected output: Pull requests with failing catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); } are blocked from merging.
Troubleshooting
Problem
The check passes locally but fails in CI
Solution
Pin dependency versions and the runtime to match CI exactly; reproduce with the same Docker image or act.
Problem
Configuration is rejected with a schema error
Solution
Check the tool version matches the config schema; older versions may not support newer fields.
Limitations
- The gate only runs on events that trigger the workflow; locally pushed commits bypass it.
- Configuration drift is possible if someone edits settings outside version control.
Make migrations safe
Start by measuring the problem. Detect a deadlock error, wait with jitter, and retry the transaction so concurrent operations do not permanently fail. The most common mistake is adding the gate without first confirming it catches the issue you care about. Run catch (e) { if (e.code === ‘40P01’) await sleep(jitter()); return retry(tx); } against the current codebase and inspect the output before wiring it into CI.
catch (e) { if (e.code === '40P01') await sleep(jitter()); return retry(tx); }Verify and roll back
Once the gate is in place, make failures actionable. A failing catch (e) { if (e.code === ‘40P01’) await sleep(jitter()); return retry(tx); } should point to the exact file and line, not a generic error. Pin the tool version in the lockfile so a upstream release does not silently change behavior. Review the gate output in every pull request just like a test result.