NVIDIA / NVIDIA/raft

[BUG] Comms not robust to worker failures

Open
#237 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug inactive-30d inactive-90d
Dominant language
Cuda
Stars
1k
Forks
251
Avg merge
1d 8h
Merged PRs (30d)
31

Description

I'm unsure whether this is really a problem with the comms or whether this information needs to be better provided within Dask, however if a worker disconnects, the scheduler seems to still return the worker as a candidate for scheduling tasks and this can cause the comms initialization to hang while the client tries to run a task directly on that worker (since client.run() is used)

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing a worker disconnect during communications initialization, then trace how the scheduler still returns that worker as a candidate and how client.run() is used. Done means initialization no longer hangs when a candidate worker has disconnected.

Written by the indexing model from the issue text.

Assessment

Domain
distributed-systems
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.