Worker: after disconnect, runs don't seem to recover
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 21
- Forks
- 23
- Avg merge
- 1d 22h
- Merged PRs (30d)
- 17
Description
Not sure about this, and it may just be a duplicate of #1178
In a flurry of lost runs yesterday, one pattern I notice is:
- Connection to lightning is lost
- The worker reconnects
- Any messages on the wire are lost: the message in flight when connection drops doesn't seem to ever get delivered (citation needed)
- If that message happened to be step:complete, we're dead
There's also a risk at this point that the run has been marked as Lost by lightning.
Actually, I think the problem we have is that once a run is marked Lost, that's it, game over. If the worker was just held up, and all the events come home an hour later for /reasons/, the run will still be lost - even though all the information gets piped in.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by tracing the worker reconnect flow and the lifecycle of runs marked Lost, focusing on what happens when a step:complete message is in flight during disconnect. Reproduce the lost-connection scenario and determine whether late events can still affect the run; done means the recovery behavior and its regression coverage are clear.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- distributed-systems
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 45/100