CIC PORTAL

jeudi 4 févr. 2010, 14:00 → 16:00 Europe/Paris

301-- (IN2P3)

Cordier Hélène

secretaire : Hélène

14:00 → 14:05 Nouveau Format de réunion

14:05 → 14:25 Task list

1- Recap. des taches/issues/bugs à faire : cf FORGE --20'

Bug #403: metrics for C-COD (C-COD metrics) 02/02/2010 12:42 C-COD metrics should be identical as it was in old Dashboard - according to description in document https://edms.cern.ch/file/1020727/1/CCOD-assessment-v1.1.pdf Additionally, Very important: 'Quality Metric' Total cell should not be average of the whole Quality Metric column but calculated from Total cells 'A Closed' and 'A Closed OK'. => responsable : Pierre dead line : 09 février Feature #397: "AT RISK" or "OUTAGE" in announcement subject 01/02/2010 15:48 Roberto Santinelli: I have also a second request: would it be possible to put in the subject (or early in the body of the message) the nature of the intervention/its severity? I mean the magic words, "AT RISK" or "OUTAGE" in the subject; for us the severity plays a big difference and would steer immediately our attention to the announcement while as it is now we have to drill down in the message and often we overlook really important announcements. => responsable : Sinnikka - 08 février Bug #395: Close Alarms - error message : windows sys msg is already registered in the DOM... 01/02/2010 10:54 Reported by H. Cordier N. & LAJILI N. On close Alarms sometimes this message appears : windows sys msg is already registered in the DOM Make sure you use set Destroy on CLose or destro on close : true in the constructor => responsable : Olivier , fin février Feature #391: Integrate xml schema to check GOC DB data 26/01/2010 16:45 Integrate xml schema to check GOC DB data in Lavoisier. => responsable Cyril , à faire avec Lavoisier 2 Feature #387: Nice Overview of Downtimes 25/01/2010 10:57 Nice Overview of Downtimes => by using the fusion Chart Library => responsable : Pierre , 15 février Feature #386: Overview of the ROC via a nice site map 25/01/2010 09:52 Overview of the ROC via a nice clickable site map => responsable : Pierre , fin février Feature #383: Pre-opened Site boxes 20/01/2010 14:39 Feedback from Melchio Jurczyk (melchior.j@gmail.com) : opening a site-box with pre-opened alarms and ticket boxes ( if there any exists ) => responsable : Olivier , fin février Feature #382: Escalation steps in the ticketing procedure 20/01/2010 14:38 In our opinion it has sence to expand the list of possible Escalations (new action taken) by position " unstable". It is not enough positions "OK" and "Unsolvable" for discribing a situation in many cases. => responsable : Cyril doit faire un document décrivant la situation pour le donner à Olivier avant l'implémentation Feature #381: Filtering alarms by age 20/01/2010 14:37 Feedback from Melchio Jurczyk (melchior.j@gmail.com) : filtering site-alarms by alarm age like before ( 24h-72h and 72h< ) => responsable : Olivier , 12 février Bug #380: Nicer display of ticket history 20/01/2010 14:35 Feedback from Melchio Jurczyk (melchior.j@gmail.com) : if you update a ticket, there is a small box showing the "history" from the GGUS, but unforunately the box is very small and much too small to use,3 lines and a scroll bar would be very helpful => responsable : Olivier, fin février Feature #379: Improve the notifications // VO Changes 20/01/2010 14:00 Currently there is an interface to see the changes into VO ID card : https://cic.gridops.org/index.php?section=vo&page=logreport An email is also sent every monday to cic-information which is the summary of changes for the last week . One improvement will be to generalize this to every users registered and also with RSS feeds => responsable : Cyril , voir avec les nouvelles VO Id Card Feature #378: Broadcast improvements // predefined options 20/01/2010 13:53 Hello, I sometimes need to use the broadcast tool from the CIC portal. I'd like to request a few enhancements to this tool, if possible. 1) As I'm always contacting the same users/VOs, would it be possible to have them defined by default? i.e. with my DN, I would always get the ones I'm interested in selected. This is to avoid wasting the same amount of time clicking on always the same boxes, sometimes forgetting one. 2) At the moment, I see only the very latest broadcasts, and I when I click "Read more news", the more recent news I see is from 2008. Can you fix that please? 3) It would be handy to be able to reply to an earlier broadcast, as with an email. Typically: "this change announced earlier has been done". Cheers, Sophie. => Responsable : Cyril , à refaire avec le nouvel outil Feature #377: Ticket automatique d'aide à la création de la VO 20/01/2010 13:47 CIC Portal Task #2779 : Pour la procedure d'enregistrement d'une VO, ajouter une création automatique de ticket automatique si le VO manager demande de l'aide pour le set-up d'un VO Server Feature #376: tickets against operational tools => Responsable : Cyril , à refaire avec le nouvel outil 20/01/2010 13:25 It will be interesting to open tickets against operational tools : SAM , GGUS , CIC Portal , GOC Db A faire en même temps que les tickets pour le middleware ... => responsable Cyril , fin mars Feature #375: Ticket template 20/01/2010 13:24 It would be good to give RODs possibility to put own part of the ticket template. It should be defined once (with possibility to change it). => déja fait pour les packages =>responsable Olivier , début avril Feature #374: 1st line support view 20/01/2010 13:22 We need to have different views (and accessible tasks/functions) for 1st line support and ROD roles in the dashboard, however, some federations combine the two roles, and for convenience, it would be good to have a separate tab in the ROD role dashboard that shows young alarms (<24 hours) which only 1st line support would normally see. This way ROD operators will not have to switch between roles. => Olivier , 12 février Bug #410: Downtime annoucement late (1 hour). 03/02/2010 17:21 Reported by Rolf Rumler : Original Content Salut, lors de la téléconf journalière WLCG d'aujourd'hui, quelqu'un de LHCb s'est plaint que le message annonçant l'extension du downtime du site est arrivé avec un retard de presque une heure. En vérifiant je suis tombé sur le message cité plus loin, confirmant ce délai. Le downtime associé a été déclaré à 11h01 UTC (12h01 Lyon). Est-ce que cela est le comportement attendu du mécanisme de notification dans une situation pareille ou y-a-t-il un problème quelque part? A+ Downtime annoucement late (1 hour). Rolf => Sinikka , 5 février

14:25 → 14:35 Background tasks

Point sur la GOC 4 ----------------------------- Point sur le dashboard : - Urls à intégrer - tests à faire de façon approfondie - tables à vérifier => Pierre Point sur le CIC : - downtimes => rss files , xml files . - changements faits dans CVS pour le test - pour la base de données => 5 tables - updater tous les liens des downtimes 4 tables changées - 1 à vérifier => alimentation automatique à vérifier Migration Lavoisier ----------------------------- => grid16 utilisée => comment tester les adapteurs => 1h de travail par jour avec Pierre sur la thématique Régionalisation ------------------------ - refaire le package - tester l'interaction des informations - vérifier les infos qu'il faut pousser à partir de la région : metrics, notepad ,handover , nagios

14:35 → 14:45 Middle term plan

VO ID CARD ------------------ Brodcast tool ---------------------- => A rediscuter fin mars.

14:45 → 14:50 AOB