Kick off workers that don't respond to metrics requests
Nobody has claimed this yet.
- Dominant language
- OCaml
- Stars
- 42
- Forks
- 20
- PR merge metrics
- No merged PRs in 30d
Description
Sometimes workers get stuck in uninterruptable waits (due to btrfs bugs). In that case, the TCP connection remains alive but the worker never does any work. The scheduler should detect this (e.g. by a timeout when collecting metrics) and mark the worker as paused (or restarting), so that queued jobs for that worker get reassigned.
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 locating the scheduler's metrics-collection path and the worker state transitions for paused or restarting workers. Trace how queued jobs are reassigned when a worker becomes unavailable; done means a nonresponsive metrics request is detected despite a live TCP connection, the worker is marked unavailable, and its queued jobs can be reassigned.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- ocaml
- Domain
- distributed-systems
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100