COD Pole1 - before COD-20

Europe/Zurich
1/1-025 (CERN)

1/1-025

CERN

20
Show room on map
Marcin Radecki (Unknown)
Attendance: Alessandro - IT, Diana - CERN, Vera, Helene, Luuk, Marcin, Malgorzata, Victor Edneral 1. Feedback from CERN/IT/FR after 1 month operating new model Alessandro: troubles using dashboard - turning alarms off takes a lot of time, a button for closing all OK alarms for entire region could be useful. CIC portal is slow, have to wait 1,2 minutes sometimes. Diana: similar as IT experience, CIC portal is slow. Vera: this slowness was a major point from a long time. Alessandro: Spend a lot of time closing OK alarms (CERN network failure). Maybe it would be useful to turn off alarms automatically after several OKs. Marcin: There was similar request some time ago IIRC but it was dropped for some reason, probably due to possibility of having sites switching between OK and intermittent failures. Diana: it could be useful to have a kind of "Quick Start Guide" for example how to turn off all alarms quickly. Action 1 on David to make sure such a section is included in new Dashboard Guide. 2. Feedback on dashboard improvements (All) MR: New dashboard version was released on May 13th, so we had one week to play with it. Malgorzata: 1) notepad improvements very helpful: one comment it would be great if in CC (and replay-to) for the e-mail send from notepad will be also 1st line support. They contact with the site directly - not the ROD. Will add comments to #106737 2) view pretty new look 3) happy with annotation that node is in SD in alarms' row 4) selection of range for alarms ([0,24h][24,72h][+72h]) as it it now in Alarms section would be helpful also in Dashboard section with memory of settings Vera: 1) it happened that a lot of not expired alarms were appeared in the CCOD dashboard. 2) it is not very effective to send an e-mail to all RODs. Need to clear the way how CCOD communicated with all RODs. Never got a response. Maybe could be useful to have a log in dashboard? 2.1. Comments on new version of ROD metrics metrics link MR: Please keep an eye on "alarms closed with status <> OK" it should not happen. More verbose report in Helsinki. MR: I think alarms still do age on weekends. Action 1a on Marcin to send info about ageing on weekends to Cyril. 3. Emulation of CCOD role after COD-20 - discussion - report from USAG meeting Vera: COD was not discussed during that meeting. Helene: David joined and there was not so much about similarities between COD and TPM. Helene: Related to action 9 from last meeting: I sent an e-mail elaborating duties out of the region scope. Action 2 on all people to read the e-mail about duties out of regional scope and provide feedback for the discussion triggered by Helene's mail. Helene: I'd need feedback for SA1 coordination meeting which is 1 week before Helsinki! 4. Knowledge sharing MR: related to action 4 from last meeting: I send an e-mail elaborating position of CE region wrt. web forum, trying to assess needs and possibilities, but got no answer. Shu-Ting: action 5: I'm waiting on reply from Torsten about possibilites to integrate web forum with GGUS search engine. It looks like there shall be no problem if there is no local language used. Diana: For USAG it is important how to build knowledge from tickets, GGUS already have a primitive interface, the manpower however is low so the issue is rather long term. Action 3 on Marcin to prepare contents of KB session in Helsinki, send it then to Shu-Ting, Diana, Victor, Helene. The plan is to trigger debate there, show was has been doen, what are the problems etc. 5. Report on CCOD ticket watching procedure - initial evaluation MR: initial report shows that 16-37% of tickets would beed completion with additional info to make is valuable for knowledge search GGUS' engine. Still we have an issue with missing original error message in the ticket. Helene: the procedure looks to me as it falls into knowledge sharing category. Marcin: yes, it is. In KS category we have two: implicit and explicit. Implicit builds the knowledge as a side-effect, explicit means we need a dedicated effort. The reports falls into implicit category. Action 4 on Marcin to trigger discussion on OAT forum about error messages from monitoring system. 6. Things we would like to cover during COD-20 HC: feedback from the CERN/IT/FR, early feedback presentations from last 4 federations. Vera: CCOD role and duties. Marcin: model evaluation - based on metrics. Review of Actions: Most of them were covered in the discussion above and most of them are done, some minor things left, people to have a look at the last meeting minutes.
There are minutes attached to this event. Show them.
    • 11:00 13:00
      COD Pole1 - before COD-20 EVO

      EVO

      CERN

      Connecting details for EVO client:
      Title: COD Pole1
      Community: Universe
      *Password: pole1
      Booked from 10:30 (advance 30' for testing)

      Phone call to EVO:
      Phone Bridge ID: 1009996
      Password: 9112

      Phone numbers:
      - Switzerland (CERN, Geneva)
      +41 22 76 71400
      - Slovakia (UPJS, Kosice)
      +421 55 234 2420
      - Italy (INFN, several cities)
      http://server10.infn.it/video/index.php?page=telephone_numbers
      Enter '4000' to access the EVO bridge
      - Germany (DESY, Hamburg)
      +49 40 8998 1340
      - United Kingdom (University of Manchester)
      +44 161 306 6802
      - Netherlands (Nikhef, Amsterdam)
      +31 20 7165293
      Dial '2' at the prompt

      Agenda

      1. Feedback from CERN/IT/FR after 1 month operating new model
      2. Feedback on dashboard improvements (All)
        2.1. Comments on new version of ROD metrics metrics link
      3. Emulation of CCOD role after COD-20 - discussion
      4. report from USAG meeting
      5. Knowledge sharing
      6. Report on CCOD ticket watching procedure - initial evaluation
      7. Things we would like to cover during COD-20

      Review of Actions:
      They are in the minutes from last meeting