> Can you specify what you mean in detail by the owner group should have
> access to it's nodes? Should already running job from the other group
> being killed?

If necessary, yes.

I'd rather those jobs were hibernated somewhere, but if we have to kill
them, so be it.

> I could think of a cron job which changes either the user_lists when there
> are waiting job of the owner of this part of a cluster, or adjust a slot
> limit in an RQS in a similar way:
>
> limit users @group1 hosts @group2 to slots=0
>
> to block further starts. Already running job would continue.

That's not going to work either, I'm afraid. Each job typically lasts 4-6
hours.

The scenario I've got is that two groups have each bought computers, and
they own that resource. I'm trying to organise it such that each group can
use idle time on the other group's machines - but it is imperative that
the group owning the machines *must* have immediate (or thereabout) access
to the boxes it owns. If they've got to wait most of a day to get access
to their own hardware, they'll simply remove them from the grid.

Vic.

_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to