Regional Operations Meeting DECH

Friday 23 Feb 2007, 15:00 → 16:00 Europe/Zurich

FZK

Sven Hermann

Telephone conference. Details for connecting to the conference were distributed through the regional mailing list.

Minutes for Regional Operations Meeting DECH (February 23th, 2007) Attendance: Sven Hermann <chair>, Clemens Koerdt, Torsten Antoni (FZK) Hans-Gunter Borrmann (Uni Freiburg) Alessandro Usai (CSCS) Uwe Ensslin, Christoph Wissing, Andreas Gellrich (DESY-HH) Jisamma, Horst Schwichtenberg (SCAI) Andreas Haupt, Kai Leffhalm (DESY-ZN) Apologies: (Uni Wuppertal) (ITWM) Missing: (MPPMU) (Uni Karlsruhe) (Uni Dortmund) (RWTH-Aachen) (GSI) 1. Introduction Last meetings minutes: no comments Announcements: Update 14 released (see link on agenda) PPS Update 20 released (see link on the agenda page) SL4 status was requested in last operations meeting (see link on agenda page for details) SA3: Highest priority is on SL4 32bit. There was one very first successful run to built the native SL4 32bit WN package. Will probably take another month for testing. Sites are encouraged to use tarballs in the meantime See also link on Markus talk (on the agenda page) AG: Wonder why experiments not increasing pressure on people? SH: Think pressure is already very high on SA3 people and its their highest priority. There is currently not much more that can be done from our side. AG: Agree. Action Items: PPS certification (start monitoring: #! ITWM, get infosystem and SAM tests working: #! GSI, ITWM) -> no news from GSI, ITWM delayed. Dech VO: Supported by all sites in region? -> ITWM now running. Karlsruhe may still have problems as well as MPPMU. Regional SFT-Server: -> in progress. will need some more time though (!# CK) SL4 issue (ask for timeline) was forwarded to operations meeting -> Answers was that it's their highest priority, but they cannot provide a timeline. Closed. DECH ticket on WMS troubles was forwarded also. -> give supporters some time. CW: its now become urgent, I think we need a quick fix otherwise we'll have to cease this service. Priority was increased to very urgent. SH: will be raised again at next ops (!#SH, done at time of writing) Action on sites to give feedback to TCG issues (discussed in topic 3 of this meeting). Closed. Action on GGUS to provide statistics link -> done. (see link on agenda page) Action on MPPMU (no response on some tickets) -> reminder was added to ticket history. Give them a phone call by next meeting. (!#CK). Action on Torsten: TPM training -> Torsten: GSI and CSCS did not receive training because they said they did not need it. Easiest is to join the shift with CSCS and GSI and I think it will not be that difficult. Will send you a link where to find the TPM schedule. Maybe we could have a short phone conference before next TPM meeting. Shift will probably start next Monday. Phone conf Monday afternoon?! TA will send an invitation (!#) 2. Round the Sites CSCS Trying to have dCache in production (probably by end of next week) DPM will stay for some time though will have to take downtime (for a week probably) to integrate remaining WNs to the system DESY PPS is OK, 4wns running, working on integration of SL4 worker nodes Production running much annoyed by WMS troubles though (see ticket 1581) DESY-ZN nothing new. service as usual progress on APEL ticket but not solved yet CW: It might help if you have a look at our YAIM configuration problem experienced at Uni-Dortmund AH: Have had this issue also before, but was fixed. Current issue is: data is not taken into account that is already in the database! FZK Soft- and Hardware maintenance completed successfully business as usual since short top BDII outage short dCache problems with high load on the PNFS server, fixed installed SL4 worker nodes on one rack to be integrated to PPS or directly into production environment with special batch queue planning to upgrade PPS SE to dCache 1.7 problems with publishing of accounting data. Update to very latest APEL RPMs solved issue GSI (no information received) ITWM (received report by Christian Peter before meeting) up and running reconfiguration of APEL solved problems with Accounting now supporting DECH VO PPS still some delay: currently integration of new worker nodes MPPMU (no information received) LRZ newly certified site working closely together with MPPMU RWTH (no information received) SCAI also problems with accounting data: MON reconfiguration has been done suggestion from rollout list with issue - Problem with proxies expiring after 24 hours - solved the problem. Extending to 72h for Dech VO worked. Biomed will be contacted (to do the same for this VO). Dortmund business as usual there is now a second site in Dortmund in context with D-grid project it's not physics but computing engineering department There is D-grid review on 1.3. after which one could think about how to integrate site into the EGEE/WLCG infrastructure. Freiburg planning to remake dCache node from scratch, data migration might be difficult however will get help from Patrick Fuhrmann will take the next weeks SE will not be available during that time Karlsruhe (some feedback received from Yves Kemp who is now with DESY) experiencing troubles with cluster, details are unknown however downtime will probably be extended Wuppertal (no information received) 3. Feedback to the TCG There has been an action on all sites to come back with suggestions from their users. But no feedback was received. HS proposed to prepare a questionnaire to be forwarded to the users by each site. Everyone supported this idea. Specific suggestions that were proposed by HS in the meeting: new applications: what can we do to support them? interactive access, user needs with the applications? Dech VO is good for new applications and new users how can we propagate this ideas to potential users? what kind of support do the users need? is it of interest to have workshop or training? what about a gLite installation training? HB: could be a good idea. we already have a workshop on grid computing (but no more registration possible). We could have another one sometime this year?! At Freiburg we mainly have local people using the local cluster so it could be of interest for them to learn how to use the grid, e.g. via the Dech VO. 4. COD Next shift beginning of March There is a new test: VOBOX There are prototypes for a new accounting test which would then replace the current APEL test. 5. ROC-On-Duty Handover FZK to SCAI FZK: nothing special to mention. Some old tickets still in the system. SCAI: no questions. GG: Statistics are now available (see link on agenda page) SH: Its a good feature. In summary 42 tickets were created and 36 tickets solved during the last two weeks. SH: Any problematic tickets to discuss? FZK: none. 6. AOB ------ none

15:00 → 15:05 Introduction

Announcements - Comments on last meetings minutes - Production update 14 released. update - PPS update 20 (mail "Release of new update to PPS: PPS-UPDATE 20" to project-eu-egee-pre-production-service@cern.ch ) - SL4 news o Operations Meeting -> "2. SL4 migration schedule urgently needed..." o live status o status talk by SA3 (Feb 7th, 2007) Action Items

Speaker: Sven Hermann

15:05 → 15:40 Round the sites

Speaker: ALL

15:40 → 15:45 Feedback to the TCG

Speaker: Horst Schwichtenberg

15:45 → 15:50 Status COD DECH

- Organisation - Report about shift (if applicable)

Speaker: Clemens Koerdt

15:50 → 15:55 Status ROC-DECH-On-Duty-Support

- Handover - Ticket Statistics

Speaker: FZK -> SCAI

15:55 → 16:00 AOB