Operations team & sites

→ Europe/London
EVO - GridPP Operations team & sites meeting

EVO - GridPP Operations team & sites meeting

Jeremy Coles
Description
- This is the biweekly ops-team & sites meeting - The intention is to run the meeting in EVO: http://evo.caltech.edu/evoGate/. Join the meeting in the GridPP Community area. - The phone bridge number is +44 (0)161 306 6802 (CERN number +41 22 76 71400). The phone bridge ID is 77907 with code: 4880. Apologies: David, Stuart W, Stuart P, Duncan, Mark
Minutes
    • 11:00 → 11:20
      Experiment problems/issues 20m
      Review of weekly issues by experiment/VO - LHCb -- http://lhcbweb.pic.es/DIRAC/LHCb-Production/visitor/jobs/SiteSummary/display "1. srm problems over the weekend 2. Some issues with pilots aborting at various Tier-2 sites. Dealt with individually." - CMS "We dont have anything to report from cms this week." - ATLAS - Other - Experiment blacklisted sites - Experiment known events affecting job slot requirements - Site performance issues -- March availbility/reliability http://gvdev.cern.ch/GRIDVIEW/downloads/Reports/201103/wlcg/WLCG_Tier2_Mar2011.pdf. Requested corrections by tomorrow please! - Site downtimes
      Slides
    • 11:20 → 11:35
      Meetings & updates 15m
      - ROD team status (any points to raise to sites or issues to follow up?) The current rota ends this week. - Tier-1 update Operational security -- Checking results at https://pakiti.egi.eu - Escalated tickets https://gus.fzk.de/download/escalationreports/roc/html/20110411_EscalationReport_ROCs.html https://gus.fzk.de/ws/ticket_info.php?ticket= 64995 - RAL T1. No GlueSACapability defined for WLCG Storage Areas (hold) 65700 - Durham. lcg-cr error on se01.dur.scotgrid.ac.uk (hold) 65991 - Encrypted DN not passed to accounting data (reopened from 46024) I suspect most sites are broken, given that 95% of jobs are now 'DN unknown'. Especially the sites our users use a lot: Durham and Glasgow Data _has_ successfully been received from the following in either January / Feb / Mar 2011: UKI-NORTHGRID-LIV-HEP RAL-LCG2 261 UKI-LT2-IC-HEP UKI-LT2-QMUL UKI-NORTHGRID-LIV-HEP UKI-SOUTHGRID-RALPP All other UK sites should please check their setups. 66306 - Start of creation of the Irish NGI operational structure 66564 - WLCG sites not publishing the GlueCECapability shared. 67164 - Durham. lcg-tags command failing. YAIM issue?? 67322 - Wrong value for GlueHostProcessorOtherDescription. -> RAL-LCG2 67365 - WLCG sites not publishing the GlueCECapability Share. Can this now be closed? 68077 - RAL-LCG2. Mandatory WLCG InstalledOnlineCapacity not published. Needs a T1 update. - Upcoming meetings
    • 11:35 → 11:45
      GDB topics/discussion review 10m
      - Agenda http://indico.cern.ch/conferenceDisplay.py?confId=106643 - Summary by Pete: http://www.gridpp.ac.uk/wiki/GDB_13th_April_2011 Intro - dCache workshop in March http://indico.desy.de/conferenceDisplay.py?confId=3959 - WLCG Workshop, DESY, 11-13th July - May topics EMI-1 Release • Information Service • HEPiX • Virtualisation • Multicore and whole box scheduling • EOS • Glexec deployment and testing Installed capacity - EFDA & QMUL (in downtime?) - T2 report http://indico.cern.ch/materialDisplay.py?contribId=5&sessionId=0&materialId=0&confId=106643 CERNVMFS - Security review done. - Being deployed at CERN (plus 2 other sites - Tiered structure) - "Instead of NFS server and mounts you have cvmfs client mounting virtual filesystem using fuse and downloading files as required by http..." Middleware - "glite 3.2 fully supported till Oct 2011, WMS will not be released in glite 3.2 will be in EMI CREAM 1.6.x from glite 3.2 newer one in cream. Then only security till April 2012 3.1 fully supported, lcg-CE, torque_utils, LSF_utils, glite-Cluster WMS FTS/FTA/FTM Security only for UI, WN VOMS, DPM &LFC (except memory leak in VOMS fixed) glite 3.2 site BDII glue 2 case sensitivity prob fixed top BDII WN lots of updates: FTS clients, WMS and LB clients, Data management clients, VOMS-Admin, lcg-infosites and lcg-ManageVOTag glite 3.1 EMI-1 should be ready by end of April. PIC confirmed that they won’t be working on CONDOR_utils for glite 3.2.Link title Likely end of support for LCG-CE is sept 2011. EMI working on release candidate 3" CREAM - Almost there with availability calculations - ACE to be used for May onwards MUPJs - Feedback on UK input WN environment - "The HEPiX Virtualisation Working Group proposed the creation of 3 files in /etc/machinefeatures with machine information during virtual machine instantiation..." "Passing similar information to jobs in real machines is considered to be useful." Technical discussion - Proposal from Ian Bird
    • 11:45 → 11:50
      EGI input 5m
      "EGI have asked for priorities in certification and release of EMI-1 components. Can you give feedback at today's [meeting] please? I think glexec and Argus should be up there. Are people waiting for other things? Are sites desperate for an SL5 WMS? What about DPM? Storm? Anything else. I am not asking what people will install when it turns up but what we are really waiting for as fixes or new features that are required but not present in glite 3.2." SB: There's also no SL5 myproxy.
    • 11:50 → 12:00
      Site review and plans 10m
      - Updates from each site on current concerns and priorities.
    • 12:00 → 12:05
      AOB 5m
      - CERN asked for issues with Oracle. They are planning a meeting with them and want additional input. RAL has issues with database licensing and tape/robot maintenance but we do not have any Sun kit. Have any of you been hit by increased licensing costs or poor support for Sun kit?