[GE users] qrsh /bin/bash error mark all Queue to Error state

Reuti reuti at staff.uni-marburg.de
Thu Jul 3 21:51:20 BST 2008


Am 03.07.2008 um 17:34 schrieb Angel Arancibia:

> 2008/7/3 Reuti <reuti at staff.uni-marburg.de>:
>> Is it still putting the queues in E state (also when the priority  
>> is zero)
>> when you close the window?
>
> Yep,
>
> queuename                      qtype used/tot. load_avg  
> arch          states
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q1.cluster.ifir.ed BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q10.cluster.ifir.e BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q11.cluster.ifir.e BIP   0/4       0.00     lx24- 
> amd64    E
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q2.cluster.ifir.ed BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q3.cluster.ifir.ed BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q4.cluster.ifir.ed BIP   0/4       0.00     lx24- 
> amd64    E
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q5.cluster.ifir.ed BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q6.cluster.ifir.ed BIP   0/4       2.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q7.cluster.ifir.ed BIP   0/4       -NA-     lx24- 
> amd64    au
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q8.cluster.ifir.ed BIP   0/4       0.00     lx24-amd64
> ---------------------------------------------------------------------- 
> ------
> sistint at era-q9.cluster.ifir.ed BIP   0/4       0.00     lx24- 
> amd64    E
>
> ###################################################################### 
> ######
>  - PENDING JOBS - PENDING JOBS - PENDING JOBS - PENDING JOBS -  
> PENDING JOBS
> ###################################################################### 
> ######
>    9483 0.25000 bash       aarancibia   qw    07/03/2008  
> 12:30:13     1

What is "qacct -j 9483" saying - is there any record with an  error  
code? Are you getting any eMails with "-m bea" with a reason? The  
setting of "loglevel" is "log_info" in SGE's configuration and still  
no output about the cause of reason in any of SGE's messages in  
$SGE_ROOT/spool/qmaster/messages or era-q9/messages and so on files  
(job rescheduled because of...)?

As I can't reproduce it (openSUSE server & terminal, besides Mac  
terminal) - maybe it's not related to SGE but to Debian/Ubuntu (you  
mentioned to use this).

Best would be, if someone from the Debian world would step in.

-- Reuti

---------------------------------------------------------------------
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