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:Switzerland-QR12

From EGIWiki
Revision as of 12:41, 3 May 2013 by Shaug (talk | contribs)
Jump to navigation Jump to search
Main EGI.eu operations services Support Documentation Tools Activities Performance Technology Catch-all Services Resource Allocation Security


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 12 NG-CH Switzerland Sigve Haug


1. MEETINGS AND DISSEMINATION

Note: Complete the tables below by adding as many rows as needed.

1.1. CONFERENCES/WORKSHOPS ORGANISED

Date Location Title Participants Outcome (Short report & Indico URL)
2013-03-21 Lugano (CSCS) CHIPP-CSCS/NGI-CH ~20 -
2013-03-12/13 UZH Python training organized at UZH. 15 http://www.gc3.uzh.ch/teaching/2013/python-march/
2013-02-26 UNIBE ARC Tutorial ~20 -

1.2. OTHER CONFERENCES/WORKSHOPS ATTENDED

Date Location Title Participants Outcome (Short report & Indico URL)
2013-04 Manchester EGI CF, Council, etc 3 from NGI-CH


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

PSI

1 Progress Summary

  - Good experiences with running all dcache services (including head node and postgresql) on our
    VMware/Netapp based virtualization environment.
  - Middleware upgrades
    - Upgraded WNs middleware to version: emi-wn-2.6.0-1_v1
    - Reinstalled 10 SL5 UIs with middleware version: emi-ui-2.0.1-1.el5
  - Storage element (dcache) upgrades
    - PostgreSQL upgraded from 9.2.2 to 9.2.3
    - dCache 1.9.12 to new golden release 2.2
    - upgrade to upcoming new golden release 2.6 planned for Q1/2014
    - gave users NFSv3 read only access to the file meta data
    - scripts for generating local dcache configuration (gplazma2)
      from LDAP user information
    - used space calculations for users, groups, directories implemented
      inside of postgresql
    - currently implementing soft quotas (i.e. notificatons) based on the
  space calculations

2.2. Main Achievements

  - dCache upgraded to the current golden release 2.2, full virtualization
  of the service


UZH Nothing to report

UNIBE-ID

 2.1. Progress Summary
 Stable operations with some short-term glitches reported below
 - M. Rolli started at 1st of March as the new man in charge for UNIBE-ID after D. Gurtner left at end of January
 - replaced some defective hardware (switch, memory modules)
 2.2. Main Achievements
 - Familiarisation with UNIBE-ID and improvements in its documentation
 - Started planning and scheduling of relocation of UNIBE-ID to a new, bigger server room (QR14)
 - Started planning further expansion of UNIBE-ID (after relocation and next year)
 2.3 Issues and mitigation
 - One weekend downtime after emergency shutdown due to cooling outage of external supplier
 - 2h outage due to power loss caused by defective high-voltage cable which affected a whole district in Bern
 - Short breakage of atlas jobs due to cvmfs proxy outage; proxy settings now setup redundant

UNIGE-DPNC

 2.1. Progress Summary
 Stable operations, with some short-term glitches reported below.
 - Fixed stability problems of new machines under SLC5
 - Replaced CPU in one new machine, replaced two disks in different disk servers
 2.2. Main Achievements
 - Common computing plans for 2013-2014 discussed between Bern and Geneva groups, priorities established
 - Understood performance issues of the 'mclimits' application
 - Studied performance scaling of ROOT benchmarks on 16 and 32-core AMD machines
 - Started preparing worker node config with SLC6
 - Started upgrading ganglia monitoring 
 2.3. Issues and mitigation
 - One day of cooling problems, an emergency shutdown.
 - Crashes of processes in the DPM SE on two occasions. Running with core dumps enabled now.
 - The issue of Solaris disk servers is still there, as reported before. 

UNIBE-LHEP Nothing reported