r4 - 28 Apr 2009 - 14:36:29 - KaushikDeYou are here: TWiki >  Admins Web > MinutesDataManageApr28



Minutes of the US ATLAS Data Management meeting, Apr 28, 2009
  • Previous meetings and background : IntegrationProgram
  • Coordinates: Tuesdays, Noon Central
    • (309) 946-5300, Access code: 735188; Dial *6 to mute/un-mute.


  • Meeting attendees: Pedro, Bob, Saul, Armen, Patrick, Wei, Charles, Hiro, Wensheng, Tom, Michael
  • Apologies: Shawn, John
  • Guests: None

Topics for this week

  • HPSS/dcache status - Pedro
    • Note special dcache optimization meeting tomorrow at BNL: MinutesBNLdcacheApr29
    • new drives have been installed
    • currently ATLAS has 20 drives (drive usage)
    • only 500 stage requests @ 9h30
    • reduce pnfs machine load during vacuum is still a work in progress
  • Storage cleanup - All
    • cleanse only does dq2-delete-replicas, not bulk deletion - so why are we seeing _sub deletion from deletion monitoring page? Need to follow up.
    • checked list of 60k obsolete datasets against LFC and pnfs space - only found few hundred files? Strange. Let this scan finish. Follow up by email.
    • ESD central cleanup will start soon at Tier 2's for Tier 0 datasets on DATADISK
  • DQ2 adler32 plugin - Hiro
    • DQ2 never retries. Follow up with DQ2 team.
    • Will become part of DQ2 release soon
  • SCRATCHDISK deployment
    • BNL - done, AGLT2 - 9TB (45TB total, done), MWT2 - 10 TB (done), NET2 - flexible TB (done), SLACT2, SWT2 - later, after cleanup
    • Maybe problem for XRootD? sites - name of SCRATCHDISK too long. Wei will follow-up with Andy.
  • BNL migration to space tokens - Armen, Pedro, Hiro
    • space has been allocated :: RT-12748 and RT-12749
    • Hiro finished all permission changes (as much as possible) - will keep an eye next few weeks
    • Panda team working on migration since 10 am today
  • Hot issues
    • Completing datasets at T2 - consistency checking, ongoing at AGLT2
    • Existing file problem - could be central deletion problem (sometimes file, sometimes LFC, never both?)? GGUS ticket 48200.
  • AOB
    • Hiro doing consistency check at BNL, comparing LFC against pnfs, followup next week
    • >17M pnfsid's stored now in LFC
    • Hiro - still ongoing checksum problems from BNL network card drivers (?) - two thumpers have lots of errors
    • Saul - still checking files at NET2 (will delete and resubscribe)

-- KaushikDe - 27 Apr 2009

About This Site

Please note that this site is a content mirror of the BNL US ATLAS TWiki. To edit the content of this page, click the Edit this page button at the top of the page and log in with your US ATLAS computing account name and password.


Powered by TWiki
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback