Zero to Hero 09 · Queues, jobs and reliability

Keep long work running safely in the background. Part 9 of 14 — 20 video lessons.

What it covers

1. Why long jobs need a queue
2. A simple job table
3. Workers that pick up work
4. Retries with backoff
5. Dead letters: when a job gives up
6. Idempotency: doing it twice safely
7. Locks and one-at-a-time work
8. Resuming after a crash
9. Health checks
10. Alerts that don't cry wolf
11. Backups before every big change
12. Restoring from a backup (practise it!)
13. Logs you can search
14. Rate limits and being a good API citizen
15. Cost ceilings per film and per day
16. Quotas per user
17. Scheduled clean-ups
18. Load testing on a budget
19. Checkpoint: a studio that survives a bad night
20. Part 9 recap

20 lessons, written and kept up to date by the community.

The lessons are for members, and joining is free.

Join free Sign in

All the courses →