Hi Juan,
according to your maui.cfg, dteam should get 1/130 share of your resources.
If you have 20 nodes, do the math...
I am not 100% sure that this causes the problem, but if it is, try to put
all queues except for dteam in one group, say 'all', and dteam queue in a
separate group, say 'reserved', adding ADEF=reserved and ADEF=all at the end
of appropriate lines in maui.cfg:
GROUPCFG[dteam] FSTARGET=100 MAXPROC=10,20 ADEF=reserved
GROUPCFG[atlas] FSTARGET=30 MAXPROC=10,20 ADEF=all
GROUPCFG[gvmuam] FSTARGET=30 MAXPROC=10 MAXJOB=1 ADEF=all
GROUPCFG[zeus] FSTARGET=30 MAXPROC=10 MAXJOB=1 ADEF=all
GROUPCFG[short] FSTARGET=30 MAXPROC=10 MAXJOB=1 ADEF=all
(I would also suggest you to remove GROUPCFG[DEFAULT] line if you do not use
it).
After this, just add
ACCOUNTCFG[reserved] FSTARGET=5 MAXPROC=19
ACCOUNTCFG[all] FSTARGET=95 MAXPROC=20
Now, the FSTARGETs will apply only within the groups, i.e. dteam will have
100% within the group 'reserved', which will get 5% of the overall resources
(i.e. 1 node); real atlas FSTARGET will be 25% of 95% etc. You can as well
adjust FSTARGETs for the rest of the queues to 25 (so that they sum up to
100), although it is not obligatory...
Hope this helps,
Antun
-----
E-mail: [log in to unmask]
Web: http://scl.phy.bg.ac.yu/
Phone: +381 11 3160260, Ext. 152
Fax: +381 11 3162190
Scientific Computing Laboratory
Institute of Physics, Belgrade
Serbia and Montenegro
-----
---------- Original Message -----------
From: Juan Jose Pardo Navarro <[log in to unmask]>
To: [log in to unmask]
Sent: Sun, 27 Nov 2005 12:20:13 +0100
Subject: [LCG-ROLLOUT] torque + maui
> Hi all,
>
> I have torque and maui and when I submit some jobs, all jobs always are
> with queue status:
>
> Some details:
>
> A)
>
> # qstat -q
>
> Queue Memory CPU Time Walltime Node Run Que Lm State
> ---------------- ------ -------- -------- ---- --- --- -- -----
> dteam -- 02:00:00 99:00:00 -- 0 28 40 E R
> --- ---
> 0 28
>
> B)
>
> the configuration of server pbs and dteam is:
>
> create queue dteam
> set queue dteam queue_type = Execution
> set queue dteam max_running = 40
> set queue dteam resources_max.cput = 02:00:00
> set queue dteam resources_max.walltime = 99:00:00
> set queue dteam enabled = True
> set queue dteam started = True
> # Set server attributes.
> #
> set server scheduling = True
> set server acl_host_enable = False
> set server managers = root@serverpbs
> set server operators = root@serverpbs
> set server default_queue = dteam
> set server log_events = 511
> set server mail_from = adm
> set server query_other_jobs = True
> set server scheduler_iteration = 600
> set server node_ping_rate = 300
> set server node_check_rate = 600
> set server tcp_timeout = 6
> set server default_node = lcgpro
> set server node_pack = False
> set server job_stat_rate = 30
>
> C)
>
> I have 20 nodes:
>
> *************************************
>
> node1
> state = free
> np = 1
> properties = lcgpro
> ntype = cluster
> status = arch=linux,uname=Linux node0 2.4.21-32.0.1.EL #1 W
> ed May 25 16:02:04 CDT 2005 i686,sessions=? 0,nsessions=?
> 0,nusers=0,idletime=26
> 881,totmem=993140kb,availmem=875488kb,physmem=479068kb,ncpus=1,
> loadave=0.00,netl oad=120720384,state=free,rectime=1132633775
>
> node1
> state = free
> np = 1
> properties = lcgpro
> ntype = cluster
> status = arch=linux,uname=Linux node1 2.4.21-32.0.1.EL #1 W
> ed May 25 16:02:04 CDT 2005 i686,sessions=? 0,nsessions=?
> 0,nusers=0,idletime=26
> 881,totmem=479068kb,availmem=361576kb,physmem=479068kb,ncpus=1,
> loadave=0.00,netl
oad=120698957,state=free,rectime=1132633785 .................................
.....................
>
> D) I send /var/spool/maui/maui.cfg file
>
> E)
>
> #qstat -a
>
> [root@gridce01 root]# qstat -a
>
> gridce01.ft.uam.es:
> Req'd Req'd
> Elap
> Job ID Username Queue Jobname SessID NDS TSK Memory Time
> S Time
> --------------- -------- -------- ---------- ------ --- --- ------ -----
> - -----
> 283.serverpbs dteam27 dteam STDIN -- 1 -- --
> 02:00 Q
> --
> 284.serverpbs dteam27 dteam STDIN -- 1 -- --
> 02:00 Q
> --
> 285.serverpbs dteam27 dteam STDIN -- 1 -- --
> 02:00 Q
> --
> 286.serverpbs dteam27 dteam STDIN -- 1 -- --
> 02:00 Q
> --
> 287.serverpbs dteam27 dteam STDIN -- 1 -- --
> 02:00 Q
> --
> .............................................................
> ...........................................................
>
> any idea ?
>
> --
> ========================================================================
> Juan Jose Pardo Navarro e-mail: [log in to unmask]
> Dpto Fisica Teorica. C-XI.
> Laboratorio de Altas Energias
> Universidad Autonoma de Madrid. Phone: 34 91 497 3976
> Cantoblanco, 28049 Madrid, Spain. Fax: 34 91 497 3936
> ========================================================================
------- End of Original Message -------
|