Swiss Grid Operations Meeting on 2016-12-01 at 14:00

• Place: Vidyo (room: Swiss_Grid_Operations_Meeting, extension: 10537598)
• External link: https://vidyoportal.cern.ch/flex.html?roomdirect.html&key=FAEn4zjAba7BqoQ11TGZu66VSDE
• Phone gate: From Switzerland: 0227671400 (portal) + 10537598 (extension) + # (pound sign)
• IRC chat: irc:gridchat.cscs.ch:994#lcg (ask pw via email)
• Switch Vidyo SIP IP: 137.138.248.204

Site status

CSCS

• Systems:
  • Working hard with Fabio to have a fully pupetized VO box
  • dm01 (ARC data mover) installed for Brisi, to be tested
  • investigating random crashes of arched on our ARC CEs
  • Fair share is improving and looks fine
• Accounting numbers (from scheduler) from last month

• Storage
  • dCache: 2.10 to 2.13 upgrade was successfull in pre-production environment. Next week will upgrade the production
  • GPFS: even with the load of only two ARC we can see a huge loads on the servers. Next week will increase the servers from 4 to 8 and change the data-metadata distribution.
  • Brisi: brisi will be moved back to GPFS (the new one with 8 servers)

PSI

• Xxx
• Accounting numbers (from scheduler) from last month

UNIBE-LHEP

• Some instabilities due to a large campaign of quite I/O heavy jobs
• ARC upgraded to 5.1.3 then to 5.2.0 . Needs to increase the limit on open files (add to the a-rex init script ulimit -n 32768 )
• Campus wide power cut 29-30 Dec. Recovery in progress, some overdue hardware maintenance and network connectivity extension.
• HammerCloud status

http://dashb-atlas-ssb.cern.ch/dashboard/request.py/siteviewhistorywithstatistics?columnid=562&view=Shifter%20view#time=720&start_date=&end_date=&use_downtimes=false&merge_colors=false&sites=multiple&clouds=all&site=ANALY_CSCS,ANALY_CSCS-HPC,ANALY_UNIBE-LHEP,ANALY_UNIBE-LHEP-UBELIX,CSCS-LCG2,CSCS-LCG2-HPC,CSCS-LCG2-HPC_MCORE,CSCS-LCG2_MCORE,UNIBE-LHEP,UNIBE-LHEP-UBELIX,UNIBE-LHEP-UBELIX_MCORE,UNIBE-LHEP_CLOUD,UNIBE-LHEP_CLOUD_MCORE,UNIBE-LHEP_MCORE,UNIGE-DPNC,UNIGE-DPNC_MCORE

• Accounting numbers (from scheduler) from last month (core-hours November 2016):

ATLAS: 816207 ; T2k: 3 ; OPS: 8

• Accounting numbers from ATLAS dashboard from last month (core-hours November 2016) [1]

CSCS / UNIBE 65% / 35% - 1543403 / 715707

• Efficiency WT ok/fail [2]:

CSCS/UNIBE: 81.57/64.07

• CPU/WT efficiency [3]:

CSCS/UNIBE 0.67/0.71

[1] http://dashb-atlas-ddm-acc.cern.ch/dashboard/request.py/dailysummary#button=cpuconsumption&sites%5B%5D=CSCS-LCG2&sites%5B%5D=UNIBE-LHEP&sitesCat%5B%5D=All+Countries&resourcetype=All&sitesSort=2&sitesCatSort=0&start=null&end=null&timerange=lastMonth&granularity=8+Hours&generic=0&sortby=0&series=All

[2] http://dashb-atlas-ddm-acc.cern.ch/dashboard/request.py/dailysummary#button=successfailures&sites%5B%5D=CSCS-LCG2&sites%5B%5D=UNIBE-LHEP&sitesCat%5B%5D=All+Countries&resourcetype=All&sitesSort=2&sitesCatSort=0&start=null&end=null&timerange=lastMonth&granularity=8+Hours&generic=0&sortby=0&series=All

[3] http://dashb-atlas-ddm-acc.cern.ch/dashboard/request.py/dailysummary#button=cpuefficiency&sites%5B%5D=CSCS-LCG2&sites%5B%5D=UNIBE-LHEP&sitesCat%5B%5D=All+Countries&resourcetype=All&sitesSort=2&sitesCatSort=0&start=null&end=null&timerange=lastMonth&granularity=8+Hours&generic=0&sortby=0&series=All

UNIBE-ID

• Xxx

UNIGE

• Operations
  • Running smoothly: Local users and ATLAS Production jobs (Multi-Core jobs started around 1 month ago)
• Storage
  • Mainly used by ATLAS local UniGE -DPNC users: ATLASLOCALGROUPDISK SpaceToken
  • Some cleaning need from time to time
• Accounting: November 2016 (last month)
  • Total WallTime = 14361.2 days ; Grid WallTime = 5035.4 days ; Local UniGe -DPNC users WallTime = 9325.8 days
  • Accounting numbers (from scheduler) from last month

NGI_CH

• Funding for NGI_CH liaiason role (operation manager, security officer, etc) extended until end June 2017 (Sigve's NeiCH project). It has been agreed I will continue.
• Feedback on VAPOR passed on to EGI operations. Some issues might have been corrected - There's a ticket tracking globally the issues:

https://ggus.eu/index.php?mode=ticket_info&ticket_id=124872

• NGI-CH Open Tickets review
  • AFS related:
    • https://ggus.eu/index.php?mode=ticket_info&ticket_id=124815 - (UZH) Roland lookign after it
  • LHCb CSCS-LCG2:
    • https://ggus.eu/index.php?mode=ticket_info&ticket_id=125207 - arc01 "unresponsible" - progress?
  • CMS CSCS-LCG2:
    • https://ggus.eu/index.php?mode=ticket_info&ticket_id=125283 - PHEDEx Agents down for T2_CH_CSCS, updated this morning, CSCS reply pending
  • ATLAS UNIBE-LHEP-UBELIX:
    • https://ggus.eu/index.php?mode=ticket_info&ticket_id=124518 - Large fraction of failign jobs. Some issue corrected, more work needed, ongoing
  • ATLAS UNIBE-LHEP:
    • https://ggus.eu/index.php?mode=ticket_info&ticket_id=117899 - Storage dumps, on hold
  • Accounting related:
    • CSCS-LCG2: https://ggus.eu/index.php?mode=ticket_info&ticket_id=123765 - Sync tests to be checked, awaiting site feedback
    • UNIBE-LHEP: https://ggus.eu/index.php?mode=ticket_info&ticket_id=124320 - Similar as above

• Views on new EGI accounting portal
  • https://accounting-next.egi.eu/egi/site/UNIBE-LHEP/normelap_processors/SubmitHost/DATE/2016/1/2016/11/egi/onlyinfrajobs/
  • https://accounting-next.egi.eu/egi/site/UNIBE-ID/normelap_processors/SubmitHost/DATE/2016/1/2016/11/egi/onlyinfrajobs/
  • https://accounting-next.egi.eu/egi/site/CSCS-LCG2/normelap_processors/SubmitHost/DATE/2016/1/2016/11/egi/onlyinfrajobs/

Other topics

• Topic1
• Topic2

Next meeting date:

A.O.B.

Attendants

• CSCS:
• CMS:
• ATLAS: Gianfranco
• LHCb:
• EGI: Gianfranco

Action items

• Item1