Minutes 

Attendees: Dmitry A., Wayne B., Leve H., Jérôme L., Gene V.B.

ShiftLog [Leve]:

• Checked that links work both ways with RunLog
• Re-deployed to fix an issue with a function used to tell if we're in a shift-taking period
  • Dmitry suggested to modify this next year to not use dates

Online infrastructure [Wayne]:

• New webserver:
  • Database access now working fine
  • Watcher script needs turned on
  • cron jobs (and associated files) need to be copied
  • Announcement for a Monday swap:
    • 2pm (Wayne, Jerome, Leve, Dmitry should be available to handle whatever may arise - Dmitry already knows one configuration he needs to change during the swap)
    • NFS will be turned off on the old server, sync will be done, IPs swapped, and NFS on.
• Two online disks (noted last week) still need replacement
• Event Display
  • Not currently functional (we lost our maintainer; Jerome is trying to prop it back up for now)
  • Pieces in multiple locations for online display plus offline (images dumped for web display)
  • Not critical (old online event display still functional), but valuable for PR
• Long-standing request (not critical) to be able to read from HPSS online (via data carousel)
  • Jerome looking for any road blocks (e.g. firewall)
• Machine replacements:
  • emcsc done
  • starutilities and ftpcgas not yet done (scheduled for work with relevant parties early next week)
• All Windows nodes rebooted with security patches
• rts01 replacement request merged into a larger purchase request (RT 2070)
• Request for help restoring old FTPC event builders (RT 2066)
  • Possible solution of other hardware if this is too much effort
• FMS/Spin request for replacement machines (RT 2073 and 2074). Not critical, but latter is important for remote access (current machine is inside WAH and not on network)

Databases [Dmitry]:

• After new nodes go into production pool, 2 old nodes will be targeted for FastOffline + nightly tests
  
  • Assessment of new node speed/power still ongoing
• onldb1 and 2 were up for 1+ years, so rebooted on purpose to check state
  • Collector daemons did not automatically start, but otherwise OK
  • Dmitry suggested leaving it this way since reboot need is so rare and automatic can take a while to get caught up (has to look through and skip all runs from the year)
  • Gene suggested making it automatic in case Dmitry's not around to restart daemons (though shift leaders can potentially do it themselves). We'll go with this.
• QA files (OnlinePlots + Jevp) stored for RunLog can now exceed 4 GB without needing manual rollover
• TPC Anode Voltges:
  
  • For now, continue with methods of collection from last year
  • Dmitry tested a supposedly-optimized "bulk reading" methods and found them to be no better than channel-by-channel (actually worse as it appears to be the same + overhead: 13-15 seconds to collect all values versus less than 1 second); dead end.
  • Possible new methods of collection under development (Tonko & the TPC group)
  • Feedback to the TPC group that wider "dead band zones" could reduce traffic and help alleviate problems due to load

FastOffline [Jerome]:

• Same as last week: many files coming into the system, but all are tests, pulsers, etc., so not being processed
• Gene notes some possible tweaks to the chain may be needed for TPC calibration status (awaiting TPC feedback)
• Dmitry need to verify with Lidia some exported info from from onldb1

QA [Gene]:

• Offline QA documentation updated (Lanny & Gene)
• Jerome notes the need to monitor and document and issues with QA Shift
• Pre-combining of FastOffline files still needs tested and configured to be the default