[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Condor-users] testing condor-G (Non Local Crashed)
- Date: Thu, 5 Jun 2008 09:48:27 -0700 (PDT)
- From: Sean Manning <seangwm@xxxxxxx>
- Subject: Re: [Condor-users] testing condor-G (Non Local Crashed)
The machine I tested this on runs Globus 4.0.3 and Condor 6.8.5. It
is a metascheduler of two different clusters. I'm new to Globus and
Condor too, so I can't give any other advice.
>Thank's for the answer Sean
>But why my globus works normal when there's no globus-job-manager
>i can still submit job and the output is good. Which version of Globus
>toolkit do you have ?
>Is the globus-job-manager daemon optional ? I still don't understand.
>> txcom2003@xxxxxxxxxxxxxxxxxxx wrote:
>>>I'm trying to test condor-G fault tolerant for non local crashed
>>>this article "Condor-G: A Computation Management Agent for
>>>Multi-Institutional Grids" by terminating globus job manager daemon. but
>>>when i searched the pid of the daemon with command ps ax | grep globus.
>>>26652 ? S 0:00
>>>/opt/globus/libexec/globus-scheduler-event-generator -s fork -t
>>>20439 ? S 0:00
>>>/opt/globus/libexec/globus-scheduler-event-generator -s condor -t
>>>Are they job manager daemon ?
>>>I use condor-G, globus toolkit 4.0.7 and condor 7.0.1.
>> Hi Tonny,
>> I think that the daemon is called globus-job-manager, like this:
>> [someuser@somehost ~]$ ps -A | grep gl
>> 5958 ? 00:05:01 globus-schedule
>> 5974 ? 00:04:57 globus-schedule
>> 5981 ? 00:00:36 globus-schedule
>> 30451 ? 00:00:03 globus-job-mana
>> 30491 ? 00:00:03 globus-job-mana
>> 30836 ? 00:00:03 globus-job-mana
>> 30857 ? 00:00:03 globus-job-mana
>> In the long form (ps -Af = _a_ll processes, long _f_ormat) that would
>> mcgridws 30857 1 0 May20 ? 00:00:03 globus-job-manager
>> -conf /usr/local/globus/etc/globus-job-manager.conf -type condor -rdn
>> jobmanager-condor -machine-type unknown -publish-jobs -condor-arch
>> INTEL -condor-os LINUX