[HTCondor-users] condor submitting to slurm

[Michael Di Domenico <mdidomenico4@xxxxxxxxx>]

i basically have two pool of compute in our environment, one condor
based and the second slurm based

currently i have the condor daemons installed on all the nodes both
condor and slurm.  what i'd like to do is remove from the condor
daemons from the slurm nodes and have condor submit its jobs to the
slurm scheduler

my understanding is that condor can now do this using the grid universe, so...

first question, is anyone doing this now?  and if so, is there any
clear documentation that explains all the config work needed in condor
and slurm.  i found a couple of presentations online, but they weren't
very clear.  what i need is kind of a tutorial end-to-end

second question, some of the examples i saw seemed to indicate that
changes to the submit files would be required.  is there any way to
avoid this?  i have users that submit extremely basic jobs to condor
into the vanilla universe.  i'd like condor to submit these jobs to
slurm as is as if the slurm node was just another condor slot in the
pool