Hi what commands did you used to submit the job. Did you used ncpus option or nodes:ppn option
On 12/25/06, luo yi <[EMAIL PROTECTED]> wrote:
there have 11 nodes left,but the a job required 10 nodes (40 cpus) can be scheduled to run.I use checkjob command to check it ,but i don't understand the reason it tell me. [EMAIL PROTECTED] ~]# checkjob 741 checking job 741 State: Idle Creds: user:zhouy group:lnm4 class:dpool qos:DEFAULT WallTime: 00:00:00 of 99:23:59:59 SubmitTime: Wed Dec 20 09:58:49 (Time Queued Total: 5:00:36:45 Eligible: 2:16:52:43) Total Tasks: 40 Req[0] TaskCount: 40 Partition: ALL Network: [NONE] Memory >= 0 Disk >= 0 Swap >= 0 Opsys: [NONE] Arch: [NONE] Features: [dpool] IWD: [NONE] Executable: [NONE] Bypass: 172 StartCount: 0 PartitionMask: [ALL] Flags: RESTARTABLE PE: 40.00 StartPriority: 3892 job cannot run in partition DEFAULT (idle procs do not meet requirements : 4 of 40 procs found) idle procs: 60 feasible procs: 4 Rejection Reasons: [CPU : 4][State : 17][ReserveTime : 10] what is the reason mean? why only 4 feasibel procs found? In fact ,there are 10 nodes,40 cpus free at that time. what's the meaning of "job cannot run in partition DEFAULT (idle procs do not meet requirements : 4 of 40 procs found) idle procs: 60 feasible procs: 4 Rejection Reasons: [CPU : 4][State : 17][ReserveTime : 10] " _________________________________________________________________ 与联机的朋友进行交流，请使用 MSN Messenger: http://messenger.msn.com/cn _______________________________________________ mauiusers mailing list [email protected] http://www.supercluster.org/mailman/listinfo/mauiusers
_______________________________________________ mauiusers mailing list [email protected] http://www.supercluster.org/mailman/listinfo/mauiusers
