 |
 |
Dear All Many thanks to all the system administrators whose hard work has made the LCG2 2.4.0 upgrade our most successful. It is certainly reassuring to see that we are making good progress in deployment and operations despite a number of still current problems. The status map for the last week has been quite healthy. Also the number of deployed resources has grown well over the last few weeks so the UKI region is delivering above expectations for this point in the EGEE project which is an excellent achievement. Moving forward there are a number of areas we need to tackle. For those of you who attended the HEPSYSMAN meeting at RAL two weeks ago (http://hepwww.rl.ac.uk/sysman/april2005/agenda.html) you will know that we are addressing issues around the UKI ticketing system. There will be a brief update on progress at tomorrow's UKI monthly operations meeting (TB-SUPPORT) (http://agenda.cern.ch/fullAgenda.php?ida=a052246). We also need to increase usage of our resources by closer monitoring of experiment software installations and problems, as well as exploring the support of wider VOs. With more usage we need to be sure of capturing usage via the accounting system but as you can see here http://goc.grid-support.ac.uk/gridsite/accounting/tree/country_view.php? Path=1.27 not all sites are yet publishing (another topic for tomorrow (also see note below)). It would be appreciated if ALL SITES (where their batch system is supported by APEL) could be publishing data before the next LCG Operations Workshop later this month (details here: http://infnforge.cnaf.infn.it/cdsagenda//fullAgenda.php?ida=a0517). Many thanks. Finally and as the second part of the mail subject suggests, I would like to bring a recent LCG-ROLLOUT mail to the attention of everyone concerned. There are still a number of sites that have R-GMA problems. Please could you be sure to raise any particular problems you need help with at tomorrow's meeting. Thanks once again to everyone for working hard to make UKI ROC and GridPP deployment and operations a success. Kind regards, Jeremy -----Original Message----- From: Laurence [mailto:[log in to unmask]] Sent: 10 May 2005 09:42 To: LHC Computer Grid - Rollout; [log in to unmask] Subject: Re: [LCG-ROLLOUT] R-GMA Status Here is todays R-GMA status. 138 Sites, 78 have R-GMA working, 29 have R-GMA Failures, 16 Have Job submission problems and 15 are in scheduled downtime. On request I have added more information to make the percentages more understandable. Next to each region is also the number of sites in the region as well as the percentage. Job Submission problems are also show as we can't test R-GMA if we can't submit jobs. For scheduled downtime I have looked for the last time that the tests worked for that site. If the test last worked over 1 week ago I also class this a failure. It is interesting to note that some sites have never passed the SFTs. UKI 20 Sites, 75% HP-Bristol (JS) QMUL-eScience (10) LivHEP-LCG2 (SD) Never worked since 08/12/2004 RHUL-LCG2 (1) ManHEP-LCG2 (SD) Last worked 28/04/2005 (JS) Job submittion problem (JL) Job list match problem (SD) Scheduled Down (1)Can connect to Tomcat Restart Tomcat an look in the log file. rm -f /var/tomcat4/logs/catalina.out /etc/rc.d/init.d/tomcat4 restart tail -f /var/tomcat4/logs/catalina.out (2)Consumer error on the test. (3)No C++ compiler found (4)Does not have a MON box, java problem on WNS (5)Memory problem, Try a reboot. (6)Upgrade problem http://goc.grid.sinica.edu.tw/gocwiki/Consumer__start_Timeinterval%3aHTM L_returned_instead_of_XML (7)Schema Access Problem (8)Firewall problem (9)Site not authorized in Registry, contact Steve Traylen. (10) RGMA_HOME is not set (11) R-GMA not installed