[GE users] SGE jobs in "qw" state

Chris Dagdigian dag at sonsorol.org
Mon May 22 21:32:42 BST 2006


Stranger and stranger.

Lets see if you can run the simplest type of job, type "qrsh  
hostname" (you want to run the 'hostname' command so don't substitute  
your machine name)  and "qrsh date" and other simple things like  
"qrsh uptime".

The qrsh command is a way to run a quick command on the least loaded  
available node in the system. This is why running "qrsh hostname" a  
few times is a good way to test out the basic SGE setup as for it to  
work, many different cascading things all have to be working.

> workgroupcluster:~ root# qrsh hostname
> node002.cluster.private

A variation of that command is to run the 'hostname' or any other  
simple unix command on a chosen remote node that you know should be  
available.

In your setup this may be something like:

   qrsh -q all.q at compute-0-97.local  "/bin/hostname"

That command should run the "/bin/hostname" command on node-097 and  
report back a hostname of "compute-0-97.local"





---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list