LHConCray Meeting on 2016-11-02

• Date and time: Wednesday 2nd of November from 14:00 to 15:00
• Place : Vidyo (room: Swiss_Grid_Operations_Meeting, extension: 10537598)
• External link: https://vidyoportal.cern.ch/flex.html?roomdirect.html&key=FAEn4zjAba7BqoQ11TGZu66VSDE
• Phone gate: From Switzerland: 0227671400 (portal) + 10537598 (extension) + # (pound sign)
• Vidyo SIP IP: 137.138.248.204

Agenda

• Ongoing tests status report

Attendants

• Gianfranco (ATLAS)
• Fabio, Christoph (CMS)
• Dino, Dario, Stefano, Nick, Miguel, Pablo (CSCS)

Minutes

• We are in a time-critical situation, where help from everyone is needed, since the testing machine will only be available until the end of the month (date not confirmed yet).
• For ATLAS running all HC jobs is a first priority. Currently only 3/6 of the tests run.
• For CMS there are other problems at Central Operations that need to be solved (that currently impact Phoenix). Jobs are running in Brisi since the weekend and results are promising.
• No input from LHCb (Wednesdays not good for Roland)

Action items

• Miguel to check on Gianfranco's input to try to understand where the problem is
• Miguel to lower the minimum memory requirements of jobs from 3 to 2 GB/core, now that the Shifter memory leak seems to be solved.
• Gianfranco to help Miguel debugging and check with the ATLAS HPC Working Group to see if there is some sort of technical forum where we can get help
• Fabio/Joosep to continue debugging problems with CMS
• Pablo to send a status/progress report in a week's time and check if another meeting is needed after that.

• 20161102_LHConCRAY_system_status_update.pptx: 20161102_LHConCRAY_system_status_update.pptx

• cms_hpc_Nov02.pdf: CMS_report