Torque Setup

From athena

http://www.clusterresources.com/wiki/doku.php?id=torque:appendix:l_torque_quickstart_guide

After compile and install, we did this...

 [root@athena0 torque-2.1.10]# export PATH=$PATH:/opt/torque/bin:/opt/torque/sbin
 [root@athena0 torque-2.1.10]# ./torque.setup moab

Installed on the first eight nodes:

 [root@athena0 ~]# cluster-fork --nodes=compute-0-%d:0-7 mkdir /opt/torque
 [root@athena0 ~]# cluster-fork --nodes=compute-0-%d:0-7 /share/apps/src/torque/torque-package-mom-linux-x86_64.sh --install
 [root@athena0 ~]# cluster-fork --nodes=compute-0-%d:0-7 /share/apps/src/torque/torque-package-clients-linux-x86_64.sh --install

To access the config files:

 [root@athena0 torque]# cd /etc
 [root@athena0 etc]# ln -s /var/spool/torque 

We then created a nodes file in the configuration directory:

 [root@athena0 server_priv]# pwd
 /etc/torque/server_priv
 [root@athena0 server_priv]# more nodes 
 compute-0-0 np=8
 compute-0-1 np=8
 compute-0-2 np=8
 compute-0-3 np=8
 compute-0-4 np=8
 compute-0-5 np=8
 compute-0-6 np=8
 compute-0-7 np=8

Nodes will find server!

 [root@compute-0-0 ~]# cd /var/spool/torque/
 [root@compute-0-0 torque]# ls
 aux         mom_logs  pbs_environment  spool
 checkpoint  mom_priv  server_name      undelivered
 [root@compute-0-0 torque]# more server_name   
 athena0.npl.washington.edu

Starting mom:

 [root@athena0 server_priv]# cluster-fork --nodes compute-0-%d:0-7 /opt/torque/sbin/pbs_mom


Hacks we needed to do:

 [root@athena0 server_priv]# cluster-fork --nodes compute-0-%d:0-7 "echo athena0.local >  /var/spool/torque/server_name"
 [root@athena0 server_priv]# cluster-fork --nodes compute-0-%d:0-7 "echo \$\pbsserver athena0.local >  /var/spool/torque/mom_priv/config"

Debugging mom:

 [root@athena0 server_priv]# cluster-fork --nodes compute-0-%d:0-7 /opt/torque/sbin/momctl -d 3
 

Restarting mom:

 [root@athena0 server_priv]# cluster-fork --nodes compute-0-%d:0-7 /opt/torque/sbin/momctl -r