From: Johnston M. J C. AFRL/D. <mic...@ki...> - 2005-12-21 20:37:57
|
Hiya all... I'm having some strange problems with PBS. I've tried to google the errors, but didn't come up with anything helpful. I did find one post by Bernard about only having LAM or MPITCH installed. I had both installed and removed MPITCH. Anyway, I submit my job with "qsub" and the job sits in the queue but goes nowhere. I'm looking in my logs and I'm seeing the following errors. ####################### PBS_Server;Svr;cluster;Scheduler sent command new PBS_Server;Svr;cluster;Scheduler sent command recyc PBS_Server;Svr;cluster;Scheduler sent command term ####################### And ####################### PBS_Server: Connection refused (111) in contact_sched, Could not connect Scheduler - port 15004 ####################### I've tried maui and pbs_sched and neither of them will send the job to a node to run. When I run the OSCAR test, it comes back and says pass. Is there somewhere else that I should look? Thanks for the help! If I figure this out I'll be sure to post it since I don't' see too many people with this problem. Mike |