Go to the Indico Home Page Oldest event Older event Up to category Newer event Newest event Printable version
  • Compact style
  • Indico style
  • Indico style - inline minutes
  • Indico style - numbered
  • Indico style - numbered + minutes
  • Indico Weeks View
    • Switch to the management area of this event
    Login

    EGEE-SA1-SWE Meeting

    → Europe/Zurich
    Island VRVS Virtual Room (http://www.vrvs.org)

    Island VRVS Virtual Room (http://www.vrvs.org)
    Pacheco, Andreu
    Description
    Present: Javi Fontan (CESGA), Adria Cajajus (UB), Kai Neuffer (INTA), Mario David (LIP), Manuel Sanchez (USC), Miguel Caballer (USC), Gonzalo Merino (PIC), Javi Lopez (CESGA), Dani Cano (IFCA), Pablo Rey (CESGA), Carlos Borrego (PIC), Ricardo Graciani (UB), Fco. Jose Bernabe (CESGA), Alvaro fernandez (IFIC), Miguel Caballer (UPV-GRyCAP), Fco. Javier Rodriguez Calonge (CIEMAT), Marc Rodriguez (PIC), Santiago Gonzalez de la Hoz (IFIC), Celso Martinez (IFCA), Carlos Fernandez (CESGA)
      • 12:00 → 12:15
        Briefing of activities 15m VRVS

        VRVS
        Speaker: Pacheco, Andreu (IFAE/PIC)
        more information
      • 12:05 → 12:20
        Operational status from sites 15m VRVS

        VRVS
        * From February 1st 2005 all sites must report by mail their operational status each friday if they want Gonzalo Merino to report with accuracy their status to EGEE Operations Meeting * Since the people from CERN is asking the ROCs to write the monday's 11 report following a sort of template, I think it will be useful that the SWE Site Reports that site managers are sending on friday to this list, try to follow a similar template as well. I suggest then that these reports are produced following the template available at: http://services2.pic.org.es/devel/download/txt/SR-template.txt
        Speaker: Merino, Gonzalo (PIC)
        • CESGA-EGEE 15m
          Javi Fontan reporting. Minor RGMA problem. Global status at START of the week: UP Global status at END of the week: UP Major operational issues encountered during the week: No problems detected Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? Already done By when will the WNs in the site be upgraded? Already done Other Comments Configuring SWE-TEST VO (problems with jom submision using this VO) until this morning (21st Feb).
          Speaker: Lopez, Javier (CESGA)
        • CIEMAT-LCG2 15m
          Javi Calonge reporting. RGMA problem fixed. Global status at START of the week: Up Global status at END of the week: Up Major operational issues encountered during the week: [Describe the main problems encountered during the week. If none, say "none"] None Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes If not, when will the upgrade be completed? Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? By when will the WNs in the site be upgraded? Although I said in my past report that we could move our cluster to SL3 at the end of this month (February), due to a cms production request based on RedHat 7.3 we have to postpone WN upgrade to SL3 at least a couple of weeks. During this time we will try to install our new storage element on SL3 and we will try to upgrade to SL3 other service nodes (UI, CE). As soon as this request finish we upgrade to SL3 all our WN. Other Comments [Enter other comments, clarifications, etc. here] In the LCG2 GIIS Monitor page, CIEMAT site is reported as "No RGMA Services Found", solved Monday, Feb 21st morning
          Speaker: Calonge, Javier (CIEMAT)
        • CNB-LCG2 15m
          Reported by mail. Global status at START of the week: Site was up Global status at END of the week: Site is up Major operational issues encountered during the week: The CNB-RGMA continues down.. I don´t know why, because my tests seem that all runs ok, but in the GIIS monitor webpage, error appears in this field. Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes If not, when will the upgrade be completed? Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? In the next month expected By when will the WNs in the site be upgraded? In the next mouth expected Other Comments The PCM (Parque Cientifico de Madrid) is cooperating in EGEE with 2 new bi-processor machine. A RB+BDII is being installed in one of them. It´s expected that the next week will be up and runnig.
          Speaker: Merino, Angel (CNB)
        • IFCA-LCG2 15m
          Dani Cano reporting. Gonzalo sees job submision failures with extrange authentication messages. Being investigated. Mario David requests that when problems are seen, tickets should be open in savannah in order to be tracked. Global status at START of the week: UP Global status at END of the week: UP Major operational issues encountered during the week: Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes If not, when will the upgrade be completed? Upgrade to Scientific Linux 3 (or equivalent): Already done.
          Speaker: Cano, Dani (IFCA)
        • IFIC-LCG2 15m
          Santi Gonzalez reporting. Main problems with router anti-p2p system blocking Atlas Data Challenge. Global status at START of the week: UP Global status at END of the week: UP Major operational issues encountered during the week: University Network Managers activated on Friday 11th a traffic control software on the router called CISCO NBAR to control peer to peer traffic. It wasn't notified to us. On Monday we discovered problems on access to RLS. Traffic on port 7777 were blocked in such a way that connection was done but, no answer was seen from the outside (port 7777 is used by edonkey). Globus-url-copy transfers were also considered as p2p, so even once the 7777 port restriction was removed, we had serious performance problems. As of Wednesday 16th, midday, the traffic control software was switch off and all became normal again. Quick response from the RLS and good bandwidth on globus-url-copy transfers. Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes If not, when will the upgrade be completed? - Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? 15-apr-2005 By when will the WNs in the site be upgraded? 15-apr-2005 Other Comments: New queues have been implemented: ificL dteamL atlasL swetestL All with 60 hours CPU limit at 72 Wall time limit. Dynamic information provider perl script has been updated accordingly
          Speaker: Sanchez, Javier (IFIC)
        • INTA-CAB 15m
          Kai Neuffer reporting. Global status at START of the week: Site was down Global status at END of the week: Site was up Major operational issues encountered during the week: [Describe the main problems encountered during the week. If none, say "none"] We have solved a hardware problem in the CE by changing it by one of the WNs. We have also changed the SE to another WN, so only 6 WNs are left. We have found several bugs in the YAIM configuration scripts (reported in savannah). At the time of writing this report there was a failure in the CERN BDII that made the monitoring jobs fail. Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes If not, when will the upgrade be completed? Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? Done on 17/11/05 By when will the WNs in the site be upgraded? Done on 17/11/05 Other Comments [Enter other comments, clarifications, etc. here] We are committing resources for the pre-production service. Points to raise at the Monday SWE Meeting [If you wish to raise any points for discussion at the weekly operations meeting, please put them here.] Status of the pre-production service from the participating sites (specially, from LIP).
          Speaker: Huedo, Eduardo (INTA)
        • LIP-LCG2 15m
          Mario David reporting. Setting up RLS for swetest VO. In production in one or two days. Suffering from BDII problem at CERN. Global status at START of the week: UP Global status at END of the week: UP Major operational issues encountered during the week: Friday 18-02-05 edg-jobs are in waiting or failure status when submited from the testzone RB at cern, seems to be a RB problem, globus-job-run work ok edg-job-submit from the lip rb02 was sucessfull Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? YES If not, when will the upgrade be completed? -- Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? All at SLC 3.0.3 By when will the WNs in the site be upgraded? All at SLC 3.0.3
          Speaker: David, Mario (LIP)
        • PIC-LCG2 15m
          Carlos Borrego reporting. Global status at START of the week: site was up Global status at END of the week: site was up Major operational issues encountered during the week: none Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? yes If not, when will the upgrade be completed? - Upgrade to Scientific Linux 3 (or equivalent): not yet, but we have already a development cluster up and running, running Scientific Linux 3. By when will all service nodes in the site be upgraded? this week By when will the WNs in the site be upgraded? this week
          Speaker: Borrego, Carlos (PIC)
        • UAM-LCG2 15m
          Reported by mail. Problem contacting the gatekeeper. A new ticket in savannah will be open. Global status at START of the week: UP Global status at END of the week: UP Major operational issues encountered during the week: Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? YES If not, when will the upgrade be completed? -- Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? ~ 1 Month By when will the WNs in the site be upgraded? ~ 1 Month Other Comments 5.1.17.7 CIEMAT, CNB, IFIC, INTA, PIC, UAM and USC to upgrade worker nodes to Scientific Linux 3.0.2 we will install the site to Scientific Linux 3.0.2 with Quattor (without yaim) (1 Month) + we have been in production for the Zeus + We are installing dcache
          Speaker: Pardo Navarro, Juan Jose (UAM)
        • UB-LCG2 15m
          Adria Casajus reporting. Global status at START of the week: down Global status at END of the week: up Major operational issues encountered during the week: Still problems when running "edg-rgma-run-exmaples", no other problem observed in R-GMA functionality, accounting uploaded since april 2004. Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? March By when will the WNs in the site be upgraded? March Other Comments Manual Installation with yaim, comments will be sent with experience.
          Speaker: Graciani, Ricardo (UB)
        • UPV-GRyCAP 15m
          Miquel Caballer reporting. Going to reinstall the CE machine with Scientific Linux in order to check if some problems encountered. Planning to change all machines to SCL3. Global status at START of the week: Site was up and running. Global status at END of the week: Site is up and running Major operational issues encountered during the week: Hardware Failure in Wednesday night. Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? Yes Upgrade to Scientific Linux 3 (or equivalent): Fedora Core 2 By when will all service nodes in the site be upgraded? Already done (FC2) By when will the WNs in the site be upgraded? Already done (FC2) Other Comments none PD: Todas las maquinas de UPV corren FC2. task #1529 to solve (extension to Glue schema not supported)
          Speaker: Blanquer, Ignacio (UPV-GRyCAP)
        • USC-LCG2 15m
          Manuel Sanchez reporting. CERN Monitoring problems observed. Global status at START of the week: up Global status at END of the week: job submission through the Test Zone RB failing Major operational issues encountered during the week: Since friday job submission to the site from the CERN TestZone resource broker is failing. Submission through the PIC RB works fine though. LHCb reports jobs failing at the site. This is being investigated. Upgrade to LCG-2.3.0: Has the site upgraded to LCG-2.3.0? YES Upgrade to Scientific Linux 3 (or equivalent): By when will all service nodes in the site be upgraded? March By when will the WNs in the site be upgraded? March Other Comments New nodes installed with SLC3. LCG 2.3.0 configuration with Quattor. CE upgraded to SLC3 and Torque.
          Speaker: Saborido, Juan Jose (USC)
        • Savannah EGEE SA1 SWE open tickets for sites 15m
          Mario David reporting. * Two open tasks. * The first task is related to UPV and reinstallation is ongoing. * The second problem is a problem at CIEMAT that now is solved.
          Speaker: David, Mario (LIP)
      • 12:10 → 12:25
        Presentation of the new Alert Monitoring System 15m VRVS

        VRVS
        * Pablo will introduce a new alert system being developped at CESGA for the SWE federation. * You can know if a site is failing by receiving mails alerting problems. * May start sending messages by external problems such as BDII. * Mario David proposes that the GOC tests would be moved to the federation level in order to control error messages and fill the central GOC. New task for Andreu * In order to receive an alert the central monitoring tests and the local CESGA tests will have to fail. * The URL of the new alert system will be: http://www.egee.cesga.es/EGEE-SA1-SWE/monitoring/notifications/
        Speaker: Rey, Pablo (CESGA)
        more information
      • 12:15 → 12:30
        Action lists 15m VRVS

        VRVS
        Speaker: Pacheco, Andreu (IFAE)
        • 5.1.17.4 Andreu Pacheco to update the SWE Execution Plan (PENDING) 15m
          * The execution plan will be reviewed in March. This action is postponed until then.
          Speaker: Pacheco, Andres (PIC)
        • 5.1.17.5 All sites except LIP and IFIC to support SWETEST Virtual Organization (PENDING) 15m
          Speaker: Pacheco, Andres (IFAE)
        • 5.1.17.6 UB to upgrade to LCG 2.3 (CLOSED) 15m
          * Done with some problems reported in the operations report.
          Speaker: Merino, Gonzalo (PIC)
        • 5.1.17.7 CIEMAT, CNB, IFIC, PIC, UAM and USC to upgrade worker nodes to Scientific Linux 3.0.2 (PENDING) 15m
          Speaker: Merino, Gonzalo (PIC)
        • 5.1.17.8 CNB to upgrade Apel version 3.4.39-1 (PENDING) 15m
          Ok for UB.
          Speaker: Lopez Cacheiro, Javier (CESGA)
        • 5.2.7.4 Implement per-user accounting in a restricted web area (CLOSED) 15m
          * Done! * See the accounting web page. Left menu down. Access controlled by password. * In future the access will be granted by certificate using gridsite. * Accounting page http://www.egee.cesga.es/EGEE-SA1-SWE/accounting/reports/
          Speaker: Rey, Pablo (CESGA)
          more information
        • 5.2.7.5 Check security issue implementing per-user accounting web (CLOSED) 15m
          * ANSWER: Up to the SWE federation to decide if publish or not accunting per user. * No privacy violated if only name, vo and institution are shown. Do not use biological data such a date of birth or country ID numbers. * Open for consultation.
          Speaker: Pacheco, Andreu (PIC)
        • 5.2.14.1 Andreu Pacheco will consult ROC Managers about normalization in the CPU accounting (CLOSED) 15m
          * The issue is that the CPU time reported to the accounting is not normalized to Spec2000 units. This means that the type and speed of the processor is not taken into account. * The proposal to raise is the agreement to normalize all accounting by Spec2000 CPU units in all EGGE sites. ANSWER from J. Coles (GOC-UK): Publish spec2000 normalized values for CPU processor time using the procedure: http://goc.grid-support.ac.uk/gridsite/accounting/normalisation_procedure.html
          Speaker: Pacheco, Andreu (PIC)
          more information
        • 5.2.14.2 Implementation of a common list of Spec2000 rates for SWE federation worker nodes (CLOSED) 15m
          * The goal is that all sites will normalize the same CPU power using the same list. This list is referenced by the www.spec.org maintained list, the cpu clock rate or the manual execution of the specint tests. * Document released on the issue of the CPU normalization: http://www.egee.cesga.es/EGEE-SA1-SWE/accounting/guides/NormalizationCriterion/index.html * Document released explaining how to normalize CPU times in heterogeneous sites using PBS mom cputmult factor: http://www.egee.cesga.es/EGEE-SA1-SWE/accounting/guides/CPUnormalizationWithPBS/index.html We hope these documents will help to clarify the current situation and to start a discussion about the Normalization procedure between all SWE members.
          Speaker: Fernandez, Carlos (CESGA)
          more information
        • 5.2.21.1 Create new mailing list for SWE site admins (NEW) 15m
          * Requested from Alvaro to redirect new additions to the swetest VO list.
          Speaker: Pacheco, Andreu (PIC)
        • 5.2.21.1 Andreu Pacheco to propose in the ROC Managers meeting to decentralize monitoring tests (NEW) 15m
          Speaker: Pacheco, Andreu (PIC)
        • 5.2.21.3 Mario to send information on network monitoring tools developped at Crossgrid (NEW) 15m
          Speaker: David, Mario (LIP)