AFAIK, the current EDG release does *not* fix this.
MDS reports an ETT of zero if the number of free cpu's exceeds the
number of queued jobs. So, if you have a queue with a restricted number
of nodes (say 4 out of 10) it will report an ETT of zero until there are
4 running jobs and 4 queued jobs.
There is a single line in ce-pbs that is responsible for this -- I have
frequently been tempted to change it, as one of my queues goes to a farm
that is a shared resource, and so is not allowed to take all the nodes
in the farm.
Andrew McNab wrote:
> On Fri, 21 Mar 2003, Ian Stokes-Rees wrote:
>
>
>>>You seem to have only one queue with no limits on cpu time or
>>>wallclock time. It's probably more useful to have several
>>>queues with different limits - even an "infinite" queue
>>>should probably have some sort of limit. If you put maximum
>>>job limits on the long queues you limit the hogging from long
>>>jobs (although with 4 cpus you don't have a lot to play with).
>>
>>PBS is setup the way the EDG installation guide told me to set it up.
>
>
> Do we know if the current EDG release fixes the mulitple queue reporting
> problems? ie that if you have multiple queues, some with empty slots but
> the total number of all-queue slots full, you can suck in jobs which wait
> at your site until a slot becomes free (this is why we only have one
> queue here.)
>
> Cheers,
>
> Andrew
--
#########################################
Dr O J E Maroney [log in to unmask]
Research Associate in eScience
HH Wills Physics Laboratory
University of Bristol
Tyndall Avenue, Bristol, BS8 1TL
Tel. 0117 928 8769
|