Skip to content

pmix/pmix3x: Fix internal PMIx discovery logic. #8832

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
Merged
jsquyres merged 2 commits into from
Apr 22, 2021
Merged

pmix/pmix3x: Fix internal PMIx discovery logic. #8832

jsquyres merged 2 commits into from
Apr 22, 2021

Conversation

@artpol84
Copy link
Contributor

@artpol84 artpol84 commented Apr 20, 2021 •
edited
Loading

See #8823 for the details.

bot:notacherrypick

@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

The fixes for other branches will follow.

@artpol84 artpol84 requested review from hppritcha and rhc54 April 20, 2021 18:50
@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

I'm waiting for the guys from LANL to confirm that this is fixing their problem (should be later today).
Once I'll hear from them, I'll remove the "WIP" label and open PRs to other branches.

@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

@rhc54 what is "PMIX_ID" and do we want to still check against it?

@artpol84 artpol84 force-pushed the topic/v4.1.x/fix_pmix_detection branch 2 times, most recently from 8a14505 to 52c8340 Compare April 20, 2021 19:02
@rhc54
Copy link
Contributor

rhc54 commented Apr 20, 2021

@rhc54 what is "PMIX_ID" and do we want to still check against it?

I have no idea - it isn't even something we deprecated. I'd remove it.

@artpol84 artpol84 force-pushed the topic/v4.1.x/fix_pmix_detection branch from 52c8340 to cc45161 Compare April 20, 2021 22:03
artpol84 added 2 commits April 20, 2021 15:03
@artpol84
pmix/pmix3x: Fix internal PMIx discovery logic.
0b3c1d9 
See open-mpi#8823 for the details.

Signed-off-by: Artem Polyakov <artpol84@gmail.com>
@artpol84
pmix: Fix detection of Externally-built PMIx
2210251 
See open-mpi#8823 for more details.

Signed-off-by: Artem Polyakov <artpol84@gmail.com>
@artpol84 artpol84 force-pushed the topic/v4.1.x/fix_pmix_detection branch from cc45161 to 2210251 Compare April 20, 2021 22:04
@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

Just got the confirmation that on the system where the problem was observed this indeed fixes the issue:

$> OMPI_MCA_pmix_base_verbose=100 srun -n 2 --mpi=pmix_v3 ./a.out
...
[sn001.localdomain:35359] mca:base:select: Auto-selecting pmix components
[sn001.localdomain:35359] mca:base:select:( pmix) Querying component [isolated]
[sn001.localdomain:35359] mca:base:select:( pmix) Query of component [isolated] set priority to 0
[sn001.localdomain:35359] mca:base:select:( pmix) Querying component [pmix3x]
[sn001.localdomain:35359] mca:base:select:( pmix) Query of component [pmix3x] set priority to 100
[sn001.localdomain:35359] mca:base:select:( pmix) Querying component [s1]
[sn001.localdomain:35359] mca:base:select:( pmix) Query of component [s1] set priority to 10
[sn001.localdomain:35359] mca:base:select:( pmix) Querying component [s2]
[sn001.localdomain:35359] mca:base:select:( pmix) Selected component [pmix3x]
...

@artpol84 artpol84 added this to the v4.1.1 milestone Apr 20, 2021
@ibm-ompi
Copy link

ibm-ompi commented Apr 20, 2021

The IBM CI (PGI) build failed! Please review the log, linked below.

Gist: https://gist.github.com/ibm-ompi/88b480fb98dfd2bdac939f023e1f8fe6

@jsquyres
Copy link
Member

jsquyres commented Apr 20, 2021

Which release branches is this PR relevant to?

@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

I think all of them

@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

I'm going to create PRs to 4.x and 3.x branches

@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

@jjhursey @gpaulsen please help with IBM failure

@jsquyres
Copy link
Member

jsquyres commented Apr 20, 2021

Ok. After this PR is merged, can you cherry-pick to all relevant release branches? Thanks.

@artpol84
Copy link
Contributor Author

artpol84 commented Apr 20, 2021

@jsquyres already opened.
I guess we don't want to go as far as v2.x, right?

@jsquyres
Copy link
Member

jsquyres commented Apr 20, 2021

@jsquyres already opened.
I guess we don't want to go as far as v2.x, right?

Correct.

@jjhursey
Copy link
Member

jjhursey commented Apr 21, 2021

bot:ibm:pgi:retest

@jsquyres
Copy link
Member

jsquyres commented Apr 21, 2021

/azp run

@azure-pipelines
Copy link

azure-pipelines bot commented Apr 21, 2021

Azure Pipelines successfully started running 1 pipeline(s).

@artpol84
Copy link
Contributor Author

artpol84 commented Apr 22, 2021

bot:mellanox:retest

1 similar comment
@artpol84
Copy link
Contributor Author

artpol84 commented Apr 22, 2021

bot:mellanox:retest

@jsquyres jsquyres merged commit 6ce9b9d into open-mpi:v4.1.x Apr 22, 2021
@it-is-a-robot it-is-a-robot mentioned this pull request Nov 30, 2021
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@rhc54 rhc54 rhc54 approved these changes

@hppritcha hppritcha hppritcha approved these changes

Assignees

No one assigned

Labels

bug RM approved

Projects

None yet

Milestone

v4.1.1

Development

Successfully merging this pull request may close these issues.

6 participants

@artpol84 @rhc54 @ibm-ompi @jsquyres @jjhursey @hppritcha