Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Creating jobqueue scheduler-specific teams

See original GitHub issue

So I am the current bottleneck in Dask-Jobqueue, I am able to devote 4 hours per week max on this (when @guillaumeeb was the main person involved it was the same thing for him).

For some time I have been thinking that having a way to ping jobqueue scheduler-specific people easily could help because they can help debug cluster-specific things and help figure out whether this is a problem in Dask-Jobqueue, a jobqueue scheduler specific thing, a site local cluster-specific thing etc …

OK so here is my first attempt to make a list of volunteers by jobqueue scheduler let me know if I have missed some obvious candidates:

SLURM @rcthomas @willirath @kathoef @jhamman (maybe more on Kubernetes these days in which case feel free to say no) @lesteve
LSF @d-v-b @stuarteberg (hopefully you haven’t changed opinion since the Dask workshop a few months ago 😉)
PBS @guillaumeeb actually is there anyone else there was a MoabCluster implementation that is supposed to be PBS-like but I have never heard back since probably that means it “just works” or it is rather useless …
SGE @ericmjl @lesteve
HTCondor @mivade @matyasselmeci @riedel
OAR @lesteve maybe I can find someone else in France, let’s see, same thing I know some (~5-10) people are using it and I have never heard too much back …

I guess the easiest if is you just reply quickly in this issue you are fine being added to such a list. Any other feed-back about the general idea more than welcome! Another thing

For more about how to get more people involved on Dask-on-HPC see https://github.com/dask/community/issues/43, don’t hesitate to get involved there.

This goes without saying but this is opt-in only, and there the only very tiny moral contract is that from time to time (I would estimate less than once a week) I may ping you when I feel it is a jobqueue scheduler specific thing and if you have some spare bandwidth at this particular moment to help debug the problem that would be greatly appreciated!