Difference between revisions of "Agenda-14-03-2011"

From EGIWiki
Jump to: navigation, search
(3 - Operational Issues (all))
(3.2 NGI wide WMS and top-level BDII used by Nagios)
Line 85: Line 85:
 
Is it a good practice to set up a dedicated WMS for nagios tests with BDII limited to NGI resources?
 
Is it a good practice to set up a dedicated WMS for nagios tests with BDII limited to NGI resources?
 
My concern comes from the fact that our CEs are sometimes marked as having problem because of non-responsive BDII/WMS.
 
My concern comes from the fact that our CEs are sometimes marked as having problem because of non-responsive BDII/WMS.
See ticket https://gus.fzk.de/dmsu/dmsu_ticket.php?ticket=67631
+
See ticket https://gus.fzk.de/dmsu/dmsu_ticket.php?ticket=67631 (-- Tomas Kouba (NGI_CZ))
 
 
-- Tomas Kouba (NGI_CZ)
 
  
 
==4 - AOB==
 
==4 - AOB==

Revision as of 14:45, 14 March 2011

Detailed Agenda

1 - Information (Mario)

1.1 - Update on status of the EMI 1.0 release

C. Aiftimiei (EMI release manager) will update the operations community about the current status of the EMI 1.0 release.

1.2 - Transition from lcg-CE to CREAM-CE

Mail from Gergely Sipos March 1 (snippet):

LCG-CE hosts are gradually being replaced by CREAM-CEs within EGI the gLite developer consortium has decided to focus its effort on the CREAM-CE and terminate support for the LCG-CE software. The exact schedule is still under discussion, but the LCG-CE software will likely reach its end of support during the summer of 2011.

In order to help EGI user communities prepare to work with CREAM-CEs, the Operations and User Support teams of EGI.eu have prepared an information page about the CREAM-CE software and about the transfer of applications from LCG-CE to CREAM-CE:

https://wiki.egi.eu/wiki/FAQ:_lcg-ce_to_cream-ce, Support team: ucst@egi.eu

1.3 - gLite 3.1 CREAM reached end of life and DPM

Mail from Tiziana March 1 for the CREAM in glite 3.1

For DPM in glite 3.1, the following NGI's showed interest: Italy, UK, Greece, Geórgia. In order for us to make any request to the developers to produce new rpm with the dependencies in voms-api version that have the mem leak fixed, we need to have early adopter sites for the test. Currently there is only one EA from the UK.

1.4 - phasing out of EGEE Grid name

Mail from Tiziana March 3

Instructions are available at: https://wiki.egi.eu/wiki/MAN1_How_to_publish_Site_Information

As of March 10: 168 sites still publish EGEE (http://gstat.egi.eu/gstat/summary/GRID/EGEE/)

Every NGI is requested to follow-up this issue internally with the respective sites.

1.5 - Decomissioning of the old SAM infrastructure

EGI Broadcast March 7

Decommission the old SAM infrastructure by the end of June 2011.

Document which describes the new programmatic interface:

https://tomtools.cern.ch/confluence/display/SAM/Migration+to+the+new+SAM+portal+and+programmatic+interface

List of machines which are contacting old SAM (Nagioses excluded):

https://tomtools.cern.ch/confluence/display/SAM/List+of+hosts+using+the+old+system

Every NGI is requested to follow-up this issue internally with the respective sites.


2 - Staged Rollout (Mario)

2.1 - gLite 3.1

  • FTS/FTM/FTA 2.2.5: are in certification.
  • Torque_Utils: in staged rollout

2.2 - gLite 3.2

  • Ready for Certification, in Certification - Upcoming in the next few days to staged rollout:
    • L&B 2.1.11
    • Torque_utils: re certification because of configuration issues
    • FTS/A/M 2.2.5
  • In staged rollout
    • TopBDII 5.1.22: under test - Contains the updated GOCDB url.
    • SiteBDII 5.1.22: under test
    • VOBOX 3.2.11: in staged rollout
    • UI 3.2.9: in staged rollout
    • Torque clients/server: The Torque utils has configuration issues, and are being solved by the developer.

2.3 - Operational Tools

3 - Operational Issues (all)

3.1 Jobs work directory and temportary directory

We would like to discuss and finalize a long-standing issue which affects the LRMS administrators, concerning the management of the job work directory and the usage of temporary directories. A proposal is presented for discussion.

3.2 NGI wide WMS and top-level BDII used by Nagios

I would like to discuss setup of NGI Nagios services in terms of WMS/BDII used. Is there a recommended way what WMS should be used for good performance and reliability of nagios tests? Is it a good practice to set up a dedicated WMS for nagios tests with BDII limited to NGI resources? My concern comes from the fact that our CEs are sometimes marked as having problem because of non-responsive BDII/WMS. See ticket https://gus.fzk.de/dmsu/dmsu_ticket.php?ticket=67631 (-- Tomas Kouba (NGI_CZ))

4 - AOB

Next Grid Operations Meeting:

28 March 2011, 14h00 Amsterdam time