Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Exited workers block the DistributedRunner #4

Closed
maleadt opened this issue Feb 4, 2021 · 2 comments · Fixed by #5
Closed

Exited workers block the DistributedRunner #4

maleadt opened this issue Feb 4, 2021 · 2 comments · Fixed by #5
Assignees

Comments

@maleadt
Copy link
Contributor

maleadt commented Feb 4, 2021

If a worker dies, the test run never finishes:

      From worker 3:    ~> Running CUDA/nnlib/batched_mul test-case (on pid=3)...
      From worker 3:    julia: /home/tim/Julia/src/julia/src/subtype.c:1887: jl_subtype_env: Assertion `obvious_subtype == 3 || obvious_subtype == subtype || jl_has_free_typevars(x) || jl_has_free_typevars(y)' failed.
      From worker 3:
      From worker 3:    signal (6): Aborted
...
      From worker 4:    Process 4 is done with handling task after running 432 tasks (out of 1336)
      From worker 5:    Process 5 is done with handling task after running 309 tasks (out of 1336)
      From worker 2:    Process 2 is done with handling task after running 436 tasks (out of 1336)
[hang]
@maleadt maleadt changed the title Segfaulting workers block the DistributedRunner Exited workers block the DistributedRunner Feb 4, 2021
@mdashti
Copy link
Contributor

mdashti commented Feb 5, 2021

Hmm. Peculiar. It shouldn't really happen. Let me look into it.

@mdashti mdashti self-assigned this Feb 5, 2021
@maleadt
Copy link
Contributor Author

maleadt commented Feb 5, 2021

I have a patch already, I'll create a PR in a while.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants