PS1 IPP Czar Logs for the week YYYY.MM.DD - YYYY.MM.DD

(Up to PS1 IPP Czar Logs)

Monday : YYYY.MM.DD

Tuesday : YYYY.MM.DD

Wednesday : 2014.03.26

  • 22:25 MEH: stdsci poll is sluggish, <50-100 being kept loaded.. restarting -- throughput nearly doubled
    • also creating this wiki.. -- is there another one?
    • might as well restart pstamp for mops in the morning as well
    • notice ps_ud_MPE not have prio so top label in stdsci, taking out of stdsci until defined

Thursday : 2014.03.27

mark is czar

  • 09:30 MEH: clearing fault 5 WS diffim with bad quality (all ref FWHM nan) -- 536632,536671 skycell.2546.063; 536692,536712,536728,536749 skycell.2606.023
  • 11:00 MEH: ps_ud_MPE given prio 300 and added back into stdsci
  • 11:20 MEH: doing regular restart of remaining long running pantasks -- summitcopy, registration, distribution
  • 12:00 MEH: MOPS stamp backlog, adding 2x c2
  • 20:40 MEH: MD.PV2 updates starting, will monitor, not necessary to do anything with it unless remove the label as necessary

Friday : 2014.03.28

mark is czar

  • 11:40 MEH: since staticsky not running, using the nodes in stdsci to finish MD.PV2
  • 12:00 MEH: near 100k past Njobs, not able to really keep the poll up and definitely not with extra nodes -- regular restart of stdsci needed

Saturday : YYYY.MM.DD

Sunday : YYYY.MM.DD

  • 11:20 Chris restarted LAP
    According to the database, we stopped on g/skycell.1914.  This is
    entry 618 in the pv2.queue.ra18-20.  I've symlinked this queue back to
    the current.queue file.
    
    To restart processing, I've done:
    
    head -640 current.queue | ./trickle_add.pl
    
    which should push 22 lapRuns to be active.  
    
  • 14:30 MEH: adding in nodes not being used for staticsky
  • 17:10 MEH: split some compute nodes used in staticsky to stack as well