[GE users] Jobs waiting due to loss of ressources

reuti reuti at staff.uni-marburg.de
Mon Dec 28 10:53:38 GMT 2009


Hi,


Am 26.12.2009 um 16:38 schrieb fanou:

> Hello,
>
> I am using SGE 6.1u4 on Linux Redhat Enterprise 5.1.
>
> For 2 days, my jobs in the queue are not launched anymore.
> If I 'qstat' pending jobs, I get the following sheduling_info :
> queue instance "all.q at master.cluster" dropped because it is full
>                             (-l fluentall=1) cannot run globally  
> because it offers only gl:fluentall=0.000000

gl: means it's a load value. Is the process which returns this load  
still running (`ps` or alike) (defined in `qconf -sconf` entry  
"load_sensor")?

-- Reuti


> This is for a serial job on 1 core. For a parallel job, I get the  
> same but the following too :
> cannot run in PE "mpi" because it only offers 0 slots
>
> It is the first time it happens. To describe the configuration, I  
> have only one queue "all.q". All nodes that are quad cores have 4  
> resources named "fluent-par".
>
>
> Output of qstat -f :
> queuename                      qtype used/tot. load_avg  
> arch          states
> ---------------------------------------------------------------------- 
> ------
> all.q at node01.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node02.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node03.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node04.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node05.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node06.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node07.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node08.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node09.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node10.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node11.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at node12.cluster           BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> all.q at master.cluster    BIP   0/0       0.04     lx24-amd64
>
>
> I must say I am used to used SGE for submission but to new to  
> administrate it.
> Any help would be appreciated !
>
> Fanou
>
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do? 
> dsForumId=38&dsMessageId=235055
>
> To unsubscribe from this discussion, e-mail: [users- 
> unsubscribe at gridengine.sunsource.net].

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=235251

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list