Mark Grondona

There's also #3346 and the need to rework job-exec to support restart with running jobs. I'm wondering if we should defer minor cleanup of the prototype-y nature of the current...

> if the coral2 plugins are introducing problems with epilog reference counts, perhaps that package could provide a shutdown.d scriptlet to fix, until a better solution is found? FYI -...

Does the gitlab CI run in a multiple brokers per node configuration? If not, I guess would could add that because having that working does aid in testing I suppose.

Hm, it seems like we should have gotten a slightly more descriptive error message here: https://github.com/flux-framework/flux-core/blob/7d61efd0096f80dad82a61f64dcf7ebdf8257f48/src/modules/resource/status.c#L82-L85 Maybe the issue is on the Python side. More investigation required.

We've had another user request this support. It seems a first step could be to support a wrapper script that reads a "config file" and can be used as a...

More context for this use case from a user: > We use the multi-prog feature to execute two different codes in the same MPI allocation; these codes communicate through MPI...

No, there is no native support in jobspec V1 for MPMD, but the script above should work using a Slurm `--mulit-prog` input file. The user can copy this script to...

One idea is to expose the quorum progress via an RPC, then responsibility for indicating progress can be handled by `flux job attach` (or any other tool that is interested)....

It is kind of nice for coredumps but not critical. Not much we can do about it anyway. There is the old sendmail method, e.g. https://kernel.googlesource.com/pub/scm/utils/util-linux/util-linux/+/v2.23.1/lib/setproctitle.c But I'm not sure...

> This turns out to be poorly optimized for the case described since idset_range_set() ends up interating over the entire trie. I'm possibly misunderstanding, but the testcase is only setting...