Alert.png The wiki is deprecated and due to be decommissioned by the end of September 2022.
The content is being migrated to other supports, new updates will be ignored and lost.
If needed you can get in touch with EGI SDIS team using operations @ egi.eu.

EGI-InSPIRE:Germany-QR10

From EGIWiki
(Redirected from Germany-QR10)
Jump to navigation Jump to search
EGI Inspire Main page


Inspire reports menu: Home SA1 weekly Reports SA1 Task QR Reports NGI QR Reports NGI QR User support Reports



Quarterly Report Number NGI Name Partner Name Author
QR 10 NGI_DE Germany Jie Tao


1. MEETINGS AND DISSEMINATION

1.1. CONFERENCES/WORKSHOPS ORGANISED

Date Location Title Participants Outcome (Short report & Indico URL)
August 27-31, 2012 Karlsruhe, Germany GridKa School 2012 140 participates from 18 countries Training in Grid, Cloud, and virtualization
August 27-31, 2012 Karlsruhe, Germany Workshop on Cluster security tournament - Hands-on incident response and forensics in a realistic environment (in cooperation with EGI-CSIRT) Together with GridKa School Training in Grid security. See: http://gridka-school.scc.kit.edu/2012/103.php
October 23-24, 2012 Karlsruhe, Germany Workshop of NGI-DE general operations “Grid in Germany” NGI-DE grid sites Prepare common and sustainable Operations Procedures for NGI-DE (EGI compatible) valid for 2013 and beyond.

1.2. OTHER CONFERENCES/WORKSHOPS ATTENDED

Date Location Title Participants Outcome (Short report & Indico URL)
September 3-5, 2012 Liverpool VERCE Training LRZ staff Training in VERSE platform: a service-oriented architecture and a data-intensive platform delivering services, workflow tools, and software as a service for the seismology community
September 17 Prague GlobusEurope LRZ staff Network and exchange knowledge with other European Globus users & Globus experts
September 18-20,2012 Prague EGI Technical Forum NGI-DE staff members (KIT, TWM, etc.) Training in Middleware, security operation, and ROD activities
September 18-21,2012 Prague EGI-CSIRT Security Training Hands-on KIT NGI-DE staff Provided Grid-infrastructure for the workshop


1.3. PUBLICATIONS

Publication title Journal / Proceedings title Journal references
Volume number
Issue

Pages from - to
Authors
1.
2.
3.
Et al?

2. ACTIVITY REPORT

2.1. Progress Summary

NGI_DE keeps smoothly operating Grid services in the region. A main contribution in the report period is the upgrading of the service software and the provision of new services. In addition, Hardware capacities (storage & CPU) was kept increasing. The ROD shift ran well.

2.2. Main Achievements

The following are concrete achievements of the NGI-DE sites:

  • DESY-HH increased storage (more than 700TB additional) and CPU (partly replacement of old worker nodes) capacities. Migration to the supported glite(EMI) releases. Stable operation with high outcome to the supported vo's.
  • FZJ contributed to UNICORE Integration Task Force; planed to solve site certification and OLA issues; decommissioned JUGGLE compute resource.
  • SCAI upgraded and moved storage element to EMI2 DPM: glite-se.scai.fraunhofer.de; upgraded LFC to EMI2: glite-lfc.scai.fraunhofer.de; upgraded VOMS to EMI2: glite-voms.scai.fraunhofer.de; upgraded all worker nodes to SL 5.8 and EMI2 WN with upgraded OFED and MPI stack; introduced additional EMI2 CREAM and APEL nodes to upgrade CE without interruption: glite-cream.scai.fraunhofer.de glite-apel.scai.fraunhofer.de (the old glite 3.2 CREAM is being phased out); upgraded WMS to EMI2; upgraded BDII to EMI2; upgraded Torque and Maui stack to most recent releases to fix some bugs; upgraded UI to EMI2 and allow GSISSH interactive login with the use of LCAS/LCMAPS; introduced VERCE.eu VO using SCAI VOMS and central LFC, support on all services including WMS.
  • LRZ developed Globus adaptors for GridSAFE and BDII; performed further development with D-MON (D-MON presentation on EGI TF 2012 in Prague);- prepared support procedure for D-Grid sites to migrate their Globus resources to EGI / NGI_DE; migrated Globus Nagios probes to IGE
  • TU-Dresden fixed some problems with the VO snoplus.
  • KIT upgraded the monitoring system with update 17 that includes sensors for Globus, Unicore, and EMI 2 WNs; updated WN to EMI 2.
  • Uni-Freiburg migrated EMI to EMI 2 and added some new WNS; CREAM 3 was in test phase.
  • ITWM participated in EMI staged rolled EMI WN, TORQUE client, EMI DPM; completed the migration of all services to EMI-2/UMD-2 and SL6 x86_64; and prepared for bidirectional data synchronization between primary and secondary GOCDB instances (RAL and ITWM).


2.3. Issues and mitigation

Issue Description Mitigation Description
FZJ reported: Resource Centre Certification and Resource Centre Operational Level Agreement procedures and requirements as currently established do not match the reality of sites offering UNICORE services. Discussions in the UNICORE Integration Task Force are ongoing. As EMI is developing a UNICORE interface to BDII, this could be the solution for publishing information to a BDII service.
SCAI detected a bug in EMI2 regarding proxy's of successful jobs. see ggus #87802.
SCAI also reported a bug in LCAS/LCMAPS regarding globus callout The bug has been fixed in LCAS 1.3.19 due to this report.
DESY-HH reported:Beeing large Tier2 we are now attraction point to the user jobs and currently we have situation that with 6000 jobs slots, we have constantly more than 10.000 queued jobs, which are mainly from one vo/group, for which we have not more than 1000 job slots. For sure many of the jobs from this queue have no possibility to run even in few days. Such long list of jobs creates enormous load on the cream-ce, which for every job propagates information from batch to wms/vo-central service. It is discussed about solutions.