Installing & Configuring a Batch Scheduler: Torque + Maui

Installing Torque (PBS)

1. Download and extract the source from Cluster Resources
2. > ./configure --prefix=/usr/local/torque --set-cflags=-O2
3. > make
4. > make install
5. > make packages (will generate .sh files for distribution)
6. Create a system account TORQUEADMIN
7. Add /usr/local/torque/bin and /usr/local/torque/sbin to path.
8. Initialize PBS server files and create default queue.
1. > ./torque.setup TORQUEADMIN
   Note that "pbs_server -t create" is running in background.
   torque.setup is similar to following:
1. > pbs_server -t create
2. > qmgr -c "set server scheduling=true"
3. > qmgr -c "create queue batch queue_type=execution"
4. > qmgr -c "set queue batch started=true"
5. > qmgr -c "set queue batch enabled=true"
6. > qmgr -c "set queue batch resources_default.nodes=1"
7. > qmgr -c "set queue batch resources_default.walltime=3600"
8. > qmgr -c "set server default_queue=batch"
9. > qmgr -c "set server operators += TORQUEADMIN@SERVERNAME"
10. > qmgr -c "set server managers += TORQUEADMIN@SERVERNAME"
    
9. Check pbs_server running status.
1. > qstat -q
2. > qmgr -c 'p s'
3. Stop the pbs_server, runs "qterm -t quick"
   
10. Install pbs_mom into all compute nodes by running the generated script torque-package-mom-linux-ia64.sh in all compute nodes.
11. Add server node information to compute nodes.
1. Create /usr/spool/PBS/server_name with the server hostname.
• > cat /usr/spool/PBS/server_name
  shannon
  
2. Create /usr/spool/PBS/mom_priv/config
1. Create the file with the following lines.
   $pbsserver shannon1 # note: IP address of host running pbs_server
   $logevent 255
   $restricted shannon1 # note: IP address of host running pbs_server
   $usecp shannon1:/home /home
   
12. Add compute node information to server node
1. Create /usr/spool/PBS/server_priv/nodes
1. Create the file with the hostnames. e.g.
   shannon2 np=2
   shannon3 np=2
   shannon4 np=2
   shannon5 np=2
   
13. Start pbs_server on server node, and pbs_mom on all compute nodes.
1. > qterm -t quick
2. > pbs_server (in server node)
3. > pbs_mom (in all compute nodes)
   
14. Verify torque
1. > qstat -q
2. > pbsnodes -a
3. > echo "sleep 30" | qsub
4. > qstat
   

End of Torque installation.

Installing Maui

1. Download and extract the source from Cluster Resources
2. > CFLAGS=-O2 ./configure --with-pbs=/usr/local/torque --with-spooldir=/usr/spool/maui --prefix=/usr/local/maui-3.2.6p13
3. > make
4. > make install
5. Create a system user mauiadmin.
6. Edit /usr/spool/maui/maui.cfg
1. Set ADMIN1 mauiadmin
7. Add mauiadmin to PBS manager and operator list.
1. > qmgr -c "set server managers += mauiadmin@shannon"
1. > qmgr -c "set server operators += mauiadmin@shannon"
8. Change owner of /usr/spool/maui and /usr/local/maui/sbin to mauiadmin.
   

End of installing Maui.

Running Torque and Maui
1. pbs_server (on server node only, must be started using root).
2. pbs_mom (on compute nodes only, must be started using root).
3. maui (on server node only)

# Prepared at 16 March 2006