[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [HTCondor-users] how to stop matching new job to a specific worker node

On Mon, 2022-03-28 at 11:06:53 +0800, jiangxw@xxxxxxxxxx wrote:
> Is there a way to stop matching new jobs to some specific worker nodes while
> waiting the running job completed itself?

Quite traditionally, I'm keeping a config file local to the node, or/and on a
shared filesystem, that's included into the overall condor_config (which is an
executable script here, combining multiple input files).

to the config this way, and running condor_reconfig to activate, does the trick
(already did before condor_drain was even invented).
To restore the node to general availability, clear the file and reconfig.

> It would be better if only centrally do a re-config on CM side, excluding
> condor_drain/condor_off or direct actions on worker node.

I know that other schedulers keep lists of available/drained nodes, but I'm
not aware of such a feature inside Condor. You can still maintain individual
config files (per node) on a shared filesystem, but there's no central means
of making the worker node reread its configuration except condor_reconfig
(... that I'm aware of - if this has changed sometimes in the past years,
I appreciate pointers)


Steffen Grunewald, Cluster Administrator
Max Planck Institute for Gravitational Physics (Albert Einstein Institute)
Am Mühlenberg 1 * D-14476 Potsdam-Golm * Germany
Fon: +49-331-567 7274
Mail: steffen.grunewald(at)aei.mpg.de