Tier1 and Tier-2 Weekly Report 2017-08-08

Tier1 and Tier-2 Weekly Report 2017-08-08

General News

ADC Technical Coordination Meeting [20170807]
- Stress testing Object Stores (Cloud and Object Store Benchmark - COSBench)
- Resources utilisation monitoring in job dashboard
- ATLAS HC Benchmarking test
- Studies of different CPUs at sites from ES data
ADC Weekly Meeting [20170801]
- MC Production
  - Using 200-300k slots, big spikes from NERSC
  - Grid went emptying last Thursday so applied the merge EVNT step and submitted all of MC16a physics samples + MC16c evgen + MC16c CP digi+reco
  - Essentially filled up grid resources
  - About 860M events in queue and more samples being prepared
- Group Derivation Production
  - No report
- Reprocessing
  - Still waiting on fix for 2017 data for missing trigger info
  - Still waiting on 2 data runs with low-pt di-tau reconstruction
  - Running local tests for Heavy Ion 2015-16 data in release 21
  - Running 2016 data reprocessing test with R20.7 and new ID alignment
  - Running 2016 data reprocessing test with R21 and track error scaling override
- CRC/ADCoS Report
  - Problems with Frontier services with Overlay jobs limited by max jobs 1200 (currently 800 due to some issues with RAL and IN2P3 servers)
  - Some fixes/enhancements to EOS after some file loss issues.
  - Been some issues running group production
  - Were some DDM issues yesterday after new code released
  - Australia-ATLAS should be on LHCONE since Monday
  - Squid service at RAL is dual-stack ipv4/6
  - Agreement reached on merging everything to 10k event files, merge all future EVNT in mc16
  - Adding filebeat monitoring to Frontier - also at TRIUMF
- Pilot Update v70.3 - release a.s.a.p.
  - Dropped 'here' from asetup options as was causing issues
  - Replaced pilot usages of deprecated schedconfig fields 'maxmemory' and 'memory' with 'maxrss'
  - Added missing verification of asetup command
  - Removed PFC generation for event service jobs since it is no longer needed
  - CPUTime fix for event service jobs due to problematic os.times()
CA Cloud Squad Actions
- User DaTRI requests approved / processed (last week)
Open Issues and Ongoing Discussions
- Availability & Reliability Reports [20170328]
  - 2017 Reports - click down to find reports of format WLCG_All_Sites_ATLAS_MMMYYYY.(html|pdf)
- Sites should be providing the 2016 storage pledge.
  - REBUS Site Capacities
  - REBUS Federation Resource Pledges
- ATLAS Canada HammerCloud Tests
WLCG and General Computing News
- Security News
- gLite/EMI News
ROC_Canada News
- Current shift