search

dmwm / CRAB2

CRAB2
2 stars 11 forks source link

StageOutSE attribute reported to Dashboard is always 'unknown' #295

Closed ericvaandering closed 11 years ago

ericvaandering commented 11 years ago

Original Savannah ticket 47788 reported by spiga on Thu Mar 5 04:59:11 2009.

StageOutSE attribute reported to Dashboard is always 'unknown'. Knowledge of ouput SE is important for analysing of the stageout problems. This info was requested by many users , including ASTF people.

ericvaandering commented 11 years ago

Comment by spiga on Mon Mar 9 04:02:37 2009

I checked the 2_5_0_pre6 and it seems to me the attribute has been sent.

Julia could you please check this task_id spiga_crab_0_090308_120905_mdn316

and let me know?

thanks Daniele

ericvaandering commented 11 years ago

Comment by julia on Mon Mar 9 05:46:49 2009

Hi Daniele, I've checked and I see that it is still reported as 'unknown'. Below is what I find in ML logs: 12:12:25 spiga_crab_0_090308_120905_mdn316 4_https://wms218.cern.ch:9000/gTiXMJLg0FCxBajqDWfNdw_192.135.30.132 StageOutSE java.lang.String:unkn own This is the same for all jobs of the task.

ericvaandering commented 11 years ago

Comment by julia on Mon Mar 9 06:18:50 2009

One more follow up. Looking more into details, I think that the SE is defined correctly, but sending is however screwed up. Just before report of StageOutSE, I see another report, which has the name of the SE in place of attribute name, not in the value field: 12:14:23 arda_cms_ml2 spiga_crab_0_090308_120905_mdn316 5_https://wms218.cern.ch:9000/8CJ6LxqIn9BGSEHgE14cAg_192.135.30.132 pccms2.cmsfarm1.ba.infn.it java.lang.String:unknown 12:14:23 arda_cms_ml2 spiga_crab_0_090308_120905_mdn316 5_https://wms218.cern.ch:9000/8CJ6LxqIn9BGSEHgE14cAg_192.135.30.132 StageOutSE java.lang.String:unknown May be it can help. Cheers Julia

ericvaandering commented 11 years ago

Comment by None on Mon Mar 9 11:31:46 2009

ok this is what I have on the log for one of my jobs.

could you check and confirm that these are not in your DB?

>>> info for dashboard: * Cat /export/data1/condor/execute/dir_7936/https_3a_2f_2fwms213.cern.ch_3a9000_2f4ipDl0dtF3B5Gndqqv9KPA/jobreport.txt * MonitorJobID=1_https://wms213.cern.ch:9000/4ipDl0dtF3B5Gndqqv9KPA MonitorID=spiga_crab_0_090308_190207_90k8zx ExeTime=20 ExeExitCode=0 StageOutExitStatusReason =Copy succedeed with srm-lcg utils StageOutSE = t2-srm-02.lnl.infn.it StageOutExitStatus = 0 CrabCpuTime=17.57 0.58 93% CrabWrapperTime=37 CrabStageoutTime=11 * End Cat jobreport * Parameters sent to Dashboard.

ericvaandering commented 11 years ago

Comment by spiga on Tue Mar 10 16:51:49 2009

ok this is what I have on the log for one of my jobs.

could you check and confirm that these are not in your DB?

>>> info for dashboard:

MonitorJobID=1_https://wms213.cern.ch:9000/4ipDl0d... MonitorID=spiga_crab_0_090308_190207_90k8zx ExeTime=20 ExeExitCode=0 StageOutExitStatusReason =Copy succedeed with srm-lcg utils StageOutSE = t2-srm-02.lnl.infn.it StageOutExitStatus = 0 CrabCpuTime=17.57 0.58 93% CrabWrapperTime=37 CrabStageoutTime=11

                       * End Cat jobreport *****
ericvaandering commented 11 years ago

Comment by spiga on Tue Mar 31 16:21:31 2009

Hi Julia

can I close the ticket?

cheers Daniele

ericvaandering commented 11 years ago

Closed by spiga on Thu May 28 05:00:40 2009

ericvaandering commented 11 years ago

Comment by spiga on Thu May 28 05:00:40 2009

Starting from 2_6_0_pre4 and 1_0_8_pre3 all the stage out problem in dashboard reporting seems fixed.

Daniele