[torqueusers] Strange problem with job ids

Nick Anderson nick at anders0n.net
Wed Sep 13 09:02:58 MDT 2006


> Looks like a DNS issue to me - the machine can't decide if it's head oder
> head_2.
> Check that the hostnames and reverse lookups are all OK.
>
> Cheers,
> Ronny

Ronny,

In the last day we have made some changes. This is a multihomed head and we 
are trying to run pbs over head_2. Now we are getting unauthorized Request 
errors.  If you have any other ideas I would appreciate it.


qrun 109
qrun: Unauthorized Request  109.head_2


qstat
Job id              Name             User            Time Use S Queue
------------------- ---------------- --------------- -------- - -----
109.head_2          test_cpi         teamhpc                0 Q default



Qmgr: l s @head
Server head_2
        server_state = Active
        scheduling = True
        total_jobs = 1
        state_count = Transit:0 Queued:1 Held:0 Waiting:0 Running:0 Exiting:0
        default_queue = default
        log_events = 511
        mail_from = adm
        query_other_jobs = True
        scheduler_iteration = 600
        node_check_rate = 150
        tcp_timeout = 6
        pbs_version = 2.1.2
        submit_hosts = head,head_2
        server_name = head_2

host head
head.cluster has address 172.16.100.1
host head_2
head_2.cluster has address 172.18.100.1


Thanks,
Nick



More information about the torqueusers mailing list