BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CERN//INDICO//EN BEGIN:VEVENT SUMMARY:Experiment problems/issues DTSTART;VALUE=DATE-TIME:20120424T100000Z DTEND;VALUE=DATE-TIME:20120424T102000Z DTSTAMP;VALUE=DATE-TIME:20130520T155833Z UID:indico-contribution-187805-1@cern.ch DESCRIPTION:Speakers: \nReview of weekly issues by experiment/VO\n\n- LHCb \n\n- CMS\n\n- ATLAS\n\n- Wider VOs\n-- Known issues\n-- Site support of t hese VOs (the 10% target)\n\nhttp://indico.cern.ch/contributionDisplay.py? contribId=1&confId=187805 LOCATION:EVO - GridPP Operations team meeting URL:http://indico.cern.ch/contributionDisplay.py?contribId=1&confId=187805 END:VEVENT BEGIN:VEVENT SUMMARY:Meetings & updates DTSTART;VALUE=DATE-TIME:20120424T102000Z DTEND;VALUE=DATE-TIME:20120424T104000Z DTSTAMP;VALUE=DATE-TIME:20130520T155833Z UID:indico-contribution-187805-0@cern.ch DESCRIPTION:Speakers: \n- ROD team update\n\n-- Also note: UKI ROC is bein g decommissioned. dteam groups/roles should be under NGI_UK.\n\n- Nagios s tatus \n\n- EGI ops\n\nBDII Instability: Did we observe problems with BDII on April 12? (One for RoD?)\n\nAgenda: https://wiki.egi.eu/wiki/Agenda-16 -04-2012\n\nEMI:\n Out now: BDII Core\, Gridsite\, VOMS\n Expected this we ek: BLAH\, DPM\, Hydra\, GFAL/lcg_util\, StoRM and WMS\n\n This WMS releas e should be the one to resolve the outstanding configuration issues we've noticed with it.\n\n The final list of what should be in EMI-2 will be rel eased soon (7th May current plan). This will give the expeted timelines f or SL6 and Debian releases. Current status is the SL5 release is looking fine\, the SL6 has a few sticky points\, but should be there. Debian: onl y the client parts\, UI and WN - and maybe not all clients in the UI for D ebian.\n EMI-2 release is due in 2 weeks. Advise against rushing to updat e production systems until they've been through staged rollout. On the ot her hand\, they'll need run through staged rollout. It looks like the UMD versions will migrate to the EMI-2 releases transparently for SL5\, so if your installed from UMD\, you should not need to do anything in particula r.\n\n Hyrda is expected to be included in EMI-2.\n\n UI/WN tarball: Ther e are testing releases of the tarballs\, inked off this ticket: https://gg us.eu/tech/ticket_show.php?ticket=74675\n Feedback to be placed back in th at ticket (for both UI and WN\, despite what the name implies).\n\nStaged Rollout:\n LFC\, and IGE SAGA adaptors.\n Mostly waiting for EMI-2\, at wh ich point there will be a lot of testing to do (for both SL5 and SL6).\n A call for staged rollout testers for Debian clients\, if we have anyone ca lled Chris who might be interested... or indeed anyone whose surname is no t Walker ....\n\nBDII instability:\n With the problems with GEANT networki ng on April 12th\, Ibergrid and NGI_IT both observed Site BDII's falling o ver\, hard.\n The suspicion is that with the GEANT problems\, there were l ots of clients connecting and dropping - leaving a pending connection on t he BDII ldap server. The cleints then reconnected\, and left another pend ing connection behind\, until this behaved as a DOS scenario. This was se en in both EMI and gLite 3.2 releases. (i.e this is seen with network in stability\, not with network outage).\n Did we observe similar problems in the UK?\n BDII Site 1.1.0 includes an update to OpenLDAP 2.4\, which shou ld prevent problems in this sort of scenario - not by directly addressing the problem\, but because it's noticably more performant\, it increases th e number of clients it can handle before this scenario causes service deli very problems.\n\n\n\n- GridPP middleware status [placeholder]\nhttp://www .hep.ph.ic.ac.uk/~dbauer/grid/staged_rollout.html\n\n- Tier-1 update\n\n1. On Thursday 12th April we had a series of Atlas disk servers loose networ k connectivity. Although not confirmed we believe this problem is fixed by a newer kernel (and network driver) and this was rolled out to the affect ed Atlas disk servers (those with a particular 10Gbit network card) that a fternoon and the following morning. We have just (yesterday evening) seen what looks like a similar thing on a LHCb disk server\, and are planning a further rollout of the newer kernel.\n2. We had a problem on one of the A tlas Castor headnodes caused by time drift. We have been checking for the ntp daemon running\, but that was not sufficient. We have now rolled out a nagios test for time drift - which has picked up a number of systems that were out by some seconds.\n3. We had a problem with xrootd access to the AtlasStripDeg service class - traced to a configuration problem.\n4. We fo und an unnecessary restriction on our 4GB batch queue - a limit that we ha ve raised.\n5. We have added two new FTS front end systems on virtual mach ines. We backed out of this change at first as a number of problems were e ncountered. (One of these was that sites that had not updated their CA cer tificates since the new UK one was released were unable to submit FTS tran sfers). We have since re-applied the update (i.e. we do now have the two n ew FTS front ends in the alias).\n\n- Security update [to be placed as fir st item]\n\n\n- T2 issues\n\n- General notes.\nThere was a GDB last week: https://indico.cern.ch/conferenceDisplay.py?confId=155067. A summary for t he next ops meeting is being put together. \n\n- Documentation review [pla ceholder this week]\nhttps://www.gridpp.ac.uk/php/KeyDocs.php\n\n- Tickets \n\nSage Matt says: " I'd appreciate it if everyone checked to see if thei r site has any crusty looking tickets that need a spring clean. I'll be ch asing you from next week otherwise."\n\nNGI\nhttps://ggus.eu/ws/ticket_inf o.php?ticket=80259\nThe new neuro science VO nearly has a name. Nearly. Th e devil's in the details (as always).\n\nMANCHESTER\nhttps://ggus.eu/ws/ti cket_info.php?ticket=81449\nThis got sent to Liverpool by accident. John r allied it to the right place\, but it may have slipped under the radar. Ti cket is from lhcb\, sounds like cvmfs problems causing job failures.\n\nht tps://ggus.eu/ws/ticket_info.php?ticket=81343\nBiomed complaining about ne gative space advertised by the CE.\n\nCAMBRIDGE\nhttps://ggus.eu/ws/ticket _info.php?ticket=80732\nThis ticket can be put to bed\, the user doesn't s ee the problems anymore. I'm not sure what Santanu did to fix things thoug h.\n\nhttps://ggus.eu/ws/ticket_info.php?ticket=77008\nLooks like this old ticket can be closed to (with the appropriate saga recorded in the soluti on).\n\nGLASGOW\nhttps://ggus.eu/ws/ticket_info.php?ticket=80752\nHas the heavy load on the WMS evened itself out?\n\nhttps://ggus.eu/ws/ticket_info .php?ticket=80371\nIf the WMS has started to behave\, will you be able to look at enabling SNO+ soon?\n\nBIRMINGHAM\nhttps://ggus.eu/ws/ticket_info. php?ticket=80527 - CE stability\nhttps://ggus.eu/ws/ticket_info.php?ticket =81434 - CVMFS?\nhttps://ggus.eu/ws/ticket_info.php?ticket=80527 \n\nhttp: //indico.cern.ch/contributionDisplay.py?contribId=0&confId=187805 LOCATION:EVO - GridPP Operations team meeting URL:http://indico.cern.ch/contributionDisplay.py?contribId=0&confId=187805 END:VEVENT BEGIN:VEVENT SUMMARY:Actions DTSTART;VALUE=DATE-TIME:20120424T110000Z DTEND;VALUE=DATE-TIME:20120424T110500Z DTSTAMP;VALUE=DATE-TIME:20130520T155833Z UID:indico-contribution-187805-3@cern.ch DESCRIPTION:Speakers: \nTo be completed:\nhttps://www.gridpp.ac.uk/wiki/Op erations_Team_Action_items\n\nCompleted:\nhttps://www.gridpp.ac.uk/wiki/Op erations_Team_Completed_Actions\n\nhttp://indico.cern.ch/contributionDispl ay.py?contribId=3&confId=187805 LOCATION:EVO - GridPP Operations team meeting URL:http://indico.cern.ch/contributionDisplay.py?contribId=3&confId=187805 END:VEVENT BEGIN:VEVENT SUMMARY:AOB DTSTART;VALUE=DATE-TIME:20120424T110500Z DTEND;VALUE=DATE-TIME:20120424T110600Z DTSTAMP;VALUE=DATE-TIME:20130520T155833Z UID:indico-contribution-187805-2@cern.ch DESCRIPTION:Speakers: \n- Topics for next week: Review of GridPP28 discuss ion outputs (actions)\; Review of April GDB updates\n\n- Reminder: HEPSYSM AN 10th/11th May: http://hepwww.rl.ac.uk/sysman/May2012/main_as_before_mee ting.html. The agenda focus: site reports\; storage & MySQL.\n\n- Next cor e tasks review meeting Friday 27th (follow-up from GridPP28). There will b e a security team meeting in early May (those concerned will have received a poll request from DK).\n\nhttp://indico.cern.ch/contributionDisplay.py? contribId=2&confId=187805 LOCATION:EVO - GridPP Operations team meeting URL:http://indico.cern.ch/contributionDisplay.py?contribId=2&confId=187805 END:VEVENT BEGIN:VEVENT SUMMARY:Network monitoring (GridMon & Perfsonar) DTSTART;VALUE=DATE-TIME:20120424T104000Z DTEND;VALUE=DATE-TIME:20120424T110000Z DTSTAMP;VALUE=DATE-TIME:20130520T155833Z UID:indico-contribution-187805-4@cern.ch DESCRIPTION:Speakers: \n- Discussion on situation\n- Clarification of plan s\n- Timelines\n\nhttp://indico.cern.ch/contributionDisplay.py?contribId=4 &confId=187805 LOCATION:EVO - GridPP Operations team meeting URL:http://indico.cern.ch/contributionDisplay.py?contribId=4&confId=187805 END:VEVENT END:VCALENDAR