Fwd: [torqueusers] recovery behavior question
tfreeman at mcs.anl.gov
Wed Feb 13 20:23:34 MST 2008
On Wed, 13 Feb 2008 21:57:57 -0500
"Glen Beane" <glen.beane at gmail.com> wrote:
> forgot to reply to list...
> ---------- Forwarded message ----------
> From: Glen Beane <glen.beane at gmail.com>
> Date: Feb 13, 2008 9:57 PM
> Subject: Re: [torqueusers] recovery behavior question
> To: Tim Freeman <tfreeman at mcs.anl.gov>
> queueing a job or having a job complete triggers a scheduling cycle
> also see the scheduler_iteration setting in qmgr (the default is 10 minutes)
OK, thankyou for explaining.
"Node(s) joining/leaving pool" seems like an event to consider triggering that,
scheduler_iteration looks like it could work for current issue.
More information about the torqueusers