[gridengine users] Weird jobs are shown up
reuti at staff.uni-marburg.de
Thu May 2 19:26:51 UTC 2013
Am 02.05.2013 um 18:13 schrieb Steven Du:
> Checked accounting file, and found these lines.
> UNKNOWN:UNKNOWN:crsusr:cranky:sen_KAR_SOD.20130328.PWPS:6268203:sge:0:0:0:0:21:0:0:0.000000:0.000000:0.000000:0:0:0:0:0:0:0:0.000000:0:0:0:0:0:0:NONE:defaultdepartment:NONE:1:0:0.000000:0.000000:0.000000:-U batchuser -l db=db_12,virtual_free=4G:0.000000:NONE:0.000000:0:0
And the new job information is gone for the recent jobs?
Was there any backup restored and/or the qmaster machine restarted in an unsafe way?
Do you have a logrotation installed for the SGE's logfiles?
Normally the restart of the qmaster is supported without any issues.
> it should be like:
> batch.q:crspbv05.com:crsusr:kar:sen_KAR_SOD.20130501.PWPS:9113775:sge:0:1367502180:1367502210:1367507697:0:0:5487:1698.202131:98.826176:3084432.000000:0:0:0:0:823523:568:0:235224.000000:476416:0:0:0:2312820:201558:NONE:defaultdepartment:NONE:1:0:1797.028307:0.000000:0.000000:-U batchuser -l db=db_05,virtual_free=4G:0.000000:NONE:0.000000:0:0
> So, where are these coming from? Why is it shown up after one month? Are they cached somewhere, and re-triggered by restart sge_qmaster?
> Any thoughts are appreciated. Thanks!
> On Thu, May 2, 2013 at 12:07 PM, Steven Du <edgefree at gmail.com> wrote:
> Hi Gurus,
> I ran into a weird issue occasionally.
> Some very old (1 month ago) jobs suddenly are shown up in GRID after restarted sge_qmaster. Sure that nobody submitted this job at this moment. And of course, those jobs' parameter to GRID is not exactly good or suitable for our GRID.
> Did google, but I am not able to find any valuable stuff.
> Can anyone help me out?
> users mailing list
> users at gridengine.org
More information about the users