[GE users] SGE noobie

craffi dag at sonsorol.org
Tue May 26 12:00:36 BST 2009

Looks like you have created two standalone clusters.

There should only be one sge_qmaster on the entire cluster, and an  
sge_execd on any node expected to run jobs. I have no idea why qhost  
shows both machines though!

In short: redo the installation making sure you kill off any old  
daemons as well, ideally with a shared $SGE_ROOT directory between the  
2 hosts on the LAN; pick one to be the qmaster, install execd on all  


On May 25, 2009, at 2:08 PM, biostat wrote:

> I am trying to get SGE working on my work LAN. I installed both  
> sge_qmaster and sge_execd on two machines. However when I try and  
> run a job on either, the other machine seems to have no knowledge of  
> the jobs existence. The job is only seen - and run - by the computer  
> that submitted it. When I try "qsub -l 'hostname=host2'" on host1,  
> the job is just listed as qw forever (on host1; host2 seems to have  
> no knowledge of the jobs existence).
> There are no errors or anything. The messages files are devoid of  
> errors. The two computers are just acting as if the other computer  
> does not exist. Yet they list eachother with qhost.
> I'm new to SGE, so I feel there is a good chance I'm just missing a  
> fundamental step or underlying layout. The only thing that could be  
> unique about this situation is that each of my computers, though on  
> a LAN, also has a unique global IP. However, if that were causing  
> the problems, I would think there would at least be error messages.
> Thanks for your time
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=198852
> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net 
> ].


To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].

More information about the gridengine-users mailing list