[GE users] Help: Job is running but no respond

Lee Amy openlinuxsource at gmail.com
Sun Jul 13 06:06:42 BST 2008


    [ The following text is in the "ISO-8859-1" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

Hello,

I submit 4 parallel jobs at the same time by using the Open MPI tight
integration parallel environment, of course I take up slots is larger than
the cluster slots. After several hours, 3 jobs have been finished and one is
still running. But when I use top command to observe the process I find that
there's no any parallel program running. And after 12 hours, the job is
still "running" when I use qstat to show.

Qstat says that the job is running and from that the slots have been
allocated.

Could you tell me what happened on earth?

Thank you very much~

Regards,

Amy Lee



More information about the gridengine-users mailing list