[GE users] Shutting down SGE 5.3
Reuti
reuti at staff.uni-marburg.de
Fri May 20 13:11:32 BST 2005
Hi William,
as you will lose all of your computing time anyway this way, you can
just issue a:
qdel -uall
and then shutdown all the machines. This way all created $TMPDIR dirs on
the nodes will be safely removed. IMO it's mostly not possible to qalter
the jobs to get them rerun after the shutdown, as input/output files
maybe in an inconsistent state (at least for our jobs). Hence the (our)
users have to look at the files and resubmit the jobs in such a case.
Cheers - Reuti
William Hay wrote:
> We may have a planned power outage to our SGE cluster next week.
> We probably wont have enough time to drain the cluster before we shut it down.
>
> What is the recommended method for shutting down SGE 5.3 in this sort of circumstance
> (ie with running non-checkpointable jobs)?
>
> William Hay
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net
More information about the gridengine-users
mailing list