1

I am new for torque and maui.

I was checking for node state to looking for which nodes are free and which nodes are in use.

For torque one command is pbsnodes. Which gives status and other info related to node.

When I was checking for maui then I found command diagnose -n which also shows status of the node.

I was wondering between these 2 status. Both are giving different status for the same situation.

When I do man pbsnodes I got the possible states for node "free", "offline", "down", "reserve", "job-exclusive", "job-sharing", "busy", "time-shared", or "state-unknown" But this type of different state I cant find for diagnose -n.

How pbsnodes and diagnose -n get the status for node. Is there any database like xCAT use for torque or maui?

Thx in advance for your valuable time.

warren
  • 18,369
  • 23
  • 84
  • 135
Nilesh
  • 255
  • 1
  • 6
  • 18

1 Answers1

1

While Torque and Maui are related, they do maintain separate information about the compute nodes which they manage.

Torque is the cluster's resource manager: it keeps track of the list of compute nodes, what hardware each one has, etc. Torque has its own simple job scheduler built-in, but this is often turned off in favor of another scheduler. Maui then sits "on top" of Torque and schedules jobs on its nodes. It supports much more flexible scheduling than the built-in Torque scheduler, and has to keep track of different information than Torque -- so the status it reports may be different.

I don't know xCAT well, but neither one has what you'd really call a database. Torque has a list of nodes in a text file, server_priv/nodes, and Maui just uses whatever nodes are supplied by Torque.

ajdecon
  • 1,301
  • 4
  • 14
  • 21
  • Thx @ajdecon, I have the list of the nodes but just I want to know that which node is on/off. For `job` there is `epilogue` and `prologue` script in `/mom_priv/` directory which will be called when job change its status. That type of any trigger is available for node which will invoke when node is going to down or up. – Nilesh Apr 04 '12 at 04:45