[Gate-users] Re : Condor grid

Samuel BURG samuel_burg at yahoo.fr
Tue Dec 21 09:54:38 CET 2010


we are running condor+gate on a 50~100 node cluster, I will send you a sample of 
our config file this afternoon.

can you give us more details on your install ?
system, version, directory tree, DNS server, NFS sharing ....

did you successfully submit some test jobs like bash scripts to the "idle" nodes 
did you check the condor log files of each nodes to see what's the issue ?

Samuel Burg

Mail : <samuel_burg at yahoo.fr>
"Aw, come on, who would be so stupid as to insert a cast
to make an error go away without actually fixing the error?"
Apparently everyone. 
-- Raymond Chen --

De : Albert Hirtl <albert.hirtl at meduniwien.ac.at>
À : gate-users at lists.opengatecollaboration.org
Envoyé le : Lun 20 décembre 2010, 9h 53min 04s
Objet : [Gate-users] Condor grid

Dear all,

We are trying to run gate on a condor gird (8 nodes with 8 CPUs each). All 
machines in the grid are visible and available. However, condor is not 
distributing the jobs among all machines. In fact, jobs are only submitted if 
started on the master, and then only the master is used. All other machines are 
idle. If ones tries to start a job on some other machine (not being the master), 
the job is not executed at all. We tried to find out what the problem in the 
condor_config might be, but I think we came to a dead end.
Does anybody have a working condor_config file and is willing to share it?

Any help is very much appreciated.


-- Medizinische Universität Wien - Allgemeines Krankenhaus Wien
Universitätsklinik für Nuklearmedizin
Währinger Gürtel 18-20
1090 Wien

Tel.: +43 (0) 1 / 40400 5545



More information about the Gate-users mailing list