Tier-2 Weekly Report 2008-11-18

SAM Results

• SAM-availability-20081118-01-ATLAS:
  

• SAM-availability-20081118-02:
  

• SAM-availability-20081118-03:
  

• SAM-availability-20081118-04:
  

• Source URL: http://lcg-sam.cern.ch:8080/reports/t2/site_avail.xsql

APEL Accounting Publishing Status

• apel-snapshot-20081118-01:
  

APEL Accounting Results

• apel-snapshot-20081118-02:
  

Site Reports

Alberta

Outstanding and Recent Issues

• corrected GlueSiteOtherInfo published by site-BDII [2008-11-12]
• doubled MCDISK area size to 8 TB [2008-11-14]
• occasional LFC connectivity errors for previously accessible guids [2008-11-15]
• corrected SE/SRM information in GOCDB for site [2008-11-18]
• intermittent SRM and CE alarms [2008-11-17]

• campus network outages scheduled for 2008-11-22 and could affect LCG-network access
• stderr/stdout direction with a handful of remaining WNs
• various gLite updates to CE, SE and MON pending

SFU

Outstanding and Recent Issues

• software release area installation full [2008-11-03; corrected 2008-11-05]
• failing panda and ganga test jobs ; zero pilot logfile sizes [2008-11-04 ; corrected 2008-11-05]

• SFU missing under CA cloud in dashboard [2008-11-05]
• Ganga job failures [2008-11-10]
• dccp errors [2008-11-06]

Victoria

Outstanding and Recent Issues

• corrected SE/SRM information in GOCDB for site [2008-11-14]
• globus-job troubles [2008-11-07; corrected 2008-11-12]
• incorrect information publishing for PhysicalCPUs and LogicalCPUs (see link) [corrected 2008-11-18]
• incorrect mappings to pool accounts for users with /atlas/ca and /atlas/Role=Production credentials [corrected 2008-11-06]

• apel accounting statistics investigation - why are they so low?

Toronto

Outstanding and Recent Issues

• release installation troubles [2008-11-17]

• incorrect information publishing for PhysicalCPUs and LogicalCPUs (see link)[corrected 2008-11-11]
• apel accounting reporting and publishing [corrected 2008-11-06]

Production

Summary

site	success	failure	success (walltime)	failure (walltime)	efficiency	efficiency (walltime)
TRIUMF-LCG2	24301	3654	345322197	19791549	86.9%	94.6%
SFU-LCG2	7945	513	137110270	2662013	93.9%	98.1%
VICTORIA-LCG2	4916	1112	159141636	9907860	81.6%	94.1%
TORONTO-LCG2	2696	133	120048311	1231444	95.3%	99%
ALBERTA-LCG2	2274	68	53548282	188058	97.1%	99.7%
total	42132	5480	815170696	33780924	88.5%	96%

Failed Jobs

ProductionFailedJobs-18Nov08.pdf

Notes

17 Nov: Received notification from shifters that jobs fail at TRIUMF; Rod responded that this was because of a jobdefs error which he had already reported.

-- BryanCaron - 18 Nov 2008