Hi
If you do a
qmgr -c 'print server'
what does it say about the acl_groups on the queues?
JT
David Garcia Aristegui wrote:
> I think i have discovered the error
> more test.sh
> #!/bin/sh
> echo hola
>
> VO biomed
> [root@mallarme biomedsgm]# su - biomedsgm
> [biomedsgm@mallarme biomedsgm]$ qsub -q biomed test.sh
> 9881.mallarme.cnb.uam.es
>
> VO geant4
> [root@mallarme biomedsgm]# su - geantsgm
> [geantsgm@mallarme geantsgm]$ qsub -q geant4 test.sh
> qsub: Unauthorized Request
>
> How can i fix this? i mean, i want to allow the geant pool acccount to
> use the geant4 queue.
>
> [root@mallarme var]# cd /opt/edg/var/info/
> [root@mallarme info]# ls -l
> total 16
> drwxr-xr-x 2 biomedsgm biomed 4096 Oct 18 16:50 biomed
> drwxr-xr-x 2 dteamsgm dteam 4096 Nov 26 23:06 dteam
> drwxr-xr-x 2 geantsgm geant 4096 Nov 25 15:36 geant4
> drwxr-xr-x 2 lhcbsgm lhcb 4096 Oct 18 16:50 lhcb
>
> [root@mallarme info]# cd /var/spool/pbs/server_priv/queues
> [root@mallarme queues]# ls -l
> total 16
> -rw------- 1 root root 791 Dec 7 12:32 biomed
> -rw------- 1 root root 711 Dec 7 12:32 dteam
> -rw------- 1 root root 711 Dec 7 12:32 geant4
> -rw------- 1 root root 794 Dec 7 12:32 lhcb
>
> [root@mallarme queues]# cd /var/spool/pbs/server_priv/acl_groups
> [root@mallarme acl_groups]# ls -l
> total 16
> -rw------- 1 root root 133 Dec 7 12:32 biomed
> -rw------- 1 root root 114 Dec 7 12:32 dteam
> -rw------- 1 root root 77 Dec 7 12:32 geant4
> -rw------- 1 root root 95 Dec 7 12:32 lhcb
>
> Everything seems to be correct. Any ideas?
> Thankx.
>
> Dan Schrager wrote:
>
>> Hi David,
>>
>> Let me tell you how I have recently added VO geant4 to WEIZMANN-LCG2:
>>
>> 1)add geant4 to the VOS line in yaim def. file
>> VOS="atlas alice lhcb cms dteam sixt zeus see geant4"
>>
>> 2)add this section in the yaim def. file:
>> VO_GEANT4_SW_DIR=$VO_SW_DIR/geant4
>> VO_GEANT4_DEFAULT_SE=$SE_HOST
>> VO_GEANT4_SGM="x"
>> VO_GEANT4_VOMS_SERVERS="vomss://lcg-voms.cern.ch:8443/voms/geant4?/geant4"
>>
>> VO_GEANT4_STORAGE_DIR=$CE_CLOSE_SE1_ACCESS_POINT/geant4
>> VO_GEANT4_QUEUES="geant4"
>>
>> the "x" is a trick bought from ROLLOUT, needed.
>>
>> 3)add to users.conf a section like that:
>> 30000:geant4sgm:2088:geant4:geant4:sgm:
>> 30001:geant4001:2088:geant4:geant4::
>> 30002:geant4002:2088:geant4:geant4::
>> ...
>> 30050:geant4050:2088:geant4:geant4::
>>
>> 2088 is the new group number I chose for geant4; 30000-30050 is my
>> chosen range of geant4 uids.
>>
>> 4) Redo the configuration step of yaim for all nodes.(use scripts...)
>>
>> Regards,
>> Dan
>>
>>
>> David Garcia Aristegui wrote:
>>
>>> Yes, we are a YAIM compliant site. Maui was restarted, and we have
>>> the same error.
>>> The lcgpbs.rvf is
>>> [root@mallarme root]# more
>>> /opt/globus/share/globus_gram_job_manager/lcgpbs.rvf
>>> Attribute: email_address
>>> Description: "Set the email address to receive notifications. See the
>>> email_on_abort, email_on_execution, and emailontermination attributes."
>>> ValidWhen: GLOBUS_GRAM_JOB_SUBMIT
>>> (...)
>>> Attribute: queue
>>> Values: dteam lhcb biomed geant4
>>>
>>> Ideas? Thank you very much.
>>>
>>> Steve Traylen wrote:
>>>
>>>> On Fri, Dec 09, 2005 at 11:56:50AM +0100 or thereabouts, David
>>>> Garcia Aristegui wrote:
>>>>
>>>>
>>>>> In my site, with a LCG2.6 Torque CE
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Hi David,
>>>>
>>>>
>>>>
>>>>> [root@mallarme maui]# qstat -Q
>>>>> Queue Max Tot Ena Str Que Run Hld Wat Trn Ext Type
>>>>> ---------------- --- --- --- --- --- --- --- --- --- --- ----------
>>>>> dteam 0 0 yes yes 0 0 0 0 0 0 Execution
>>>>> lhcb 0 1 yes yes 0 1 0 0 0 0 Execution
>>>>> biomed 0 0 yes yes 0 0 0 0 0 0 Execution
>>>>> geant4 0 0 yes yes 0 0 0 0 0 0 Execution
>>>>>
>>>>> But the geant4 jobs never run in a WN, but we have no problems at
>>>>> all with the othe VO jobs (dteam, lhcb, biomed)
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>> If you add a queue you should probably restart maui. But also the file
>>>> in /opt/globus/share/globus_gram_job_manager/lcgpbs.rvf
>>>> must contain the new queue which is create by
>>>> /opt/globus/sbin/globus-initalization.sh which is called by YAIM
>>>> if you are using that.
>>>>
>>>> Steve
>>>>
>>>>
>>>>> *************************************************************
>>>>> BOOKKEEPING INFORMATION:
>>>>>
>>>>> Status info for the Job :
>>>>> https://gdrb03.cern.ch:9000/L9m-YZ5oaOIilL-jra_0Ww
>>>>> Current Status: Aborted Status Reason: Job RetryCount (3) hit
>>>>> Destination: mallarme.cnb.uam.es:2119/jobmanager-lcgpbs-geant4
>>>>> reached on: Wed Dec 7 23:20:09 2005
>>>>> *************************************************************
>>>>>
>>>>>
>>>>> What should i check to fix this?
>>>>>
>>>>> Thankx.
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>> +++++++++++++++++++++++++++++++++++++++++++
>>> This Mail Was Scanned By Mail-seCure System
>>> at the Tel-Aviv University CC.
>>
>>
>>
>>
|