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.

Sign up for GitHub

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

Jump to bottom

PMIX ERROR: NOT-FOUND in file src/client/pmix_client.c at line 205 #1323

Closed
robnagler opened this issue Aug 29, 2018 · 1 comment
Closed

PMIX ERROR: NOT-FOUND in file src/client/pmix_client.c at line 205 #1323

robnagler opened this issue Aug 29, 2018 · 1 comment

Comments

@robnagler
Copy link
Member

Happens on dev in WarpVND and Synergia. Restarting the server fixes it

[v.radia.run:01024] PMIX ERROR: NOT-FOUND in file src/client/pmix_client.c at line 205
[v:01024] *** Process received signal ***
[v:01024] Signal: Segmentation fault (11)
[v:01024] Signal code: Invalid permissions (2)
[v:01024] Failing at address: 0x2b6e810
[v:01024] [ 0] /lib64/libpthread.so.0(+0x121b0)[0x7f9382dd91b0]
[v:01024] [ 1] [0x2b6e810]
@robnagler robnagler mentioned this issue Sep 17, 2018
Closed
@robnagler
Copy link
Member Author

This is related to #1151 and #1311. h5py starts MPI and populates the environment. open-mpi/ompi#3158 (comment) suggests clearing the environment of OMPI_* and PMIX_* environment on os.exec (subprocess). Doing this seems to fix this issue. I'll add this to new runner code, not the old.

robnagler pushed a commit that referenced this issue Sep 17, 2018
fix #1323 eliminate OMPI_ and PMIX_ from environ before exec
5207ed3 
h5py is starting MPI but it shouldn't really so this trick is necessary
@hvasbath hvasbath mentioned this issue Jun 9, 2020
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@robnagler