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.

Difference between revisions of "Agenda-13-12-2011"

From EGIWiki
Jump to navigation Jump to search
 
(20 intermediate revisions by 4 users not shown)
Line 1: Line 1:
{{Template:Op menubar}}  
{{Template:Op menubar}}  
[[Category:Grid Operations Meetings]]


= Detailed agenda: Grid Operations Meeting 13 December 2011 14h00 Amsterdam time  =
= Detailed agenda: Grid Operations Meeting 13 December 2011 14h00 Amsterdam time  =
Line 15: Line 16:


=== 1.1 EMI-1 release status  ===
=== 1.1 EMI-1 release status  ===
External wiki page provided by Cristina (EMI): [https://twiki.cern.ch/twiki/bin/view/EMI/EmiEgiGOM#Status_12_12_2011 link]


=== 1.2 Staged Rollout (Mario)  ===
=== 1.2 Staged Rollout (Mario)  ===


==== 1.2.1 gLite 3.2 ====
==== 1.2.1 UMD1 ====


==== 1.2.2 UMD1  ====
UMD 1.4 freeze yesterday, now preparing the release to be out next Monday 19 Dec.  


*emi.cream.sl5.x86_64-1.13.3 with the following updates
**emi.cemon.sl5.x86_64-1.13.3
**emi.apel_parsers.sl5.x86_64-1.0.1
**emi.blah.sl5.x86_64-1.16.3
*emi.ge-utils.sl5.x86_64-1.0.0
*emi.mpi.sl5.x86_64-1.1.0
*emi.apel.sl5.x86_64-3.2.8
*emi.storm.sl5.x86_64-1.8.0
*emi.lcg-util.sl5.x86_64-1.11.19
*Update to Globus 5.0.5 baseline provided by IGE, ensuring a continuing integration with EMI products in the EGI
**ige.globus-default-security.sl5.x86_64-5.0.5
**ige.globus-gridftp.sl5.x86_64-5.0.5
**ige.globus-rls.sl5.x86_64-5.0.5
**ige.globus-gsissh.sl5.x86_64-4.3.3


==== 1.2.3 UMD1.4 release schedule  ====
Products in the SW provisioning, candidates for UMD 1.5 by the 30th January 2012:
 
Products under staged rollout:
 
*IGE.globus-myproxy.sl5.x86_64-5.4.4
 
To be verified:
 
*EMI.unicore-uvos.sl5.x86_64-1.5.0
*EMI.lb.sl5.x86_64-3.1.0
*EMI.dpm.sl5.x86_64-1.8.2
*EMI.lfc_mysql.sl5.x86_64-1.8.2
*EMI.lfc_oracle.sl5.x86_64-1.8.2
*IGE.gridsam.sl5.x86_64-2.3.0
*IGE.gridway.sl5.x86_64-5.8.1
 
Also waiting for the EMI update 11 on the 15th of December, WMS in partiular is expected in this elease.


== 2. Operational Issues  ==
== 2. Operational Issues  ==
=== 2.1 GOCDB data scoping overview ===
=== 2.1 GOCDB data scoping overview ===
Presentation: [[File:GocdbScopingHighLevel.pdf]]
=== 2.2 Top BDII memory issue ===
A serious memory issue was reported in two GGUS tickets:
[ [https://ggus.eu/tech/ticket_show.php?ticket=76337 76337] ]: "bdii tmpfs fills up and silently fails"
[ [https://ggus.eu/tech/ticket_show.php?ticket=73840 73840] ]: "bdii memory leak?"
''The RAMdisk usage of the service increases until the service occupy all the available memory''. After that it needs to be restarted.
The BDII versions affected are:
*gLite 3.2.11
*EMI 1.0.0-1
No information about EMI 1.0.1 (Distributed with UMD 1.3), if it is affected or not.
'''First partial mitigation suggested by developer''':
* Configuration patch, comments the following directives:
*: #set_flags DB_LOG_INMEMORY
*: #set_flags DB_TXN_NOSYNC
*: This reduces the memory grow, but does not solve the problem
'''Mitigation''':
* The only workaround for the problem, so far, is to restart the Top-BDII service.
=== 2.3 Survey ongoing ===
Survey about the request for SLURM support in CREAM.
[[/EGI_Operations_Surveys#.28OPEN.29_SLURM_support_for_CREAM| Survey description]]
*Deadline is ''tomorrow'': Dec 14th.
*Many sites answered (69), probably most of the sites not interested in SLURM, and happy with the list of LRMS supported by CREAM, did not answer (reasonably).
=== 2.4 Middleware requirements gathering campaign ===
*Requirements for ARC/gLite/GLOBUS/UNICORE.
*The campaign is ongoing and it will be closed on January 15th, to let OMB discuss the requirements list.
*[[Mw-requirements.html |instructions]]


== 3. AOB  ==
== 3. AOB  ==
Line 32: Line 99:


=== 3.2 Next meetings  ===
=== 3.2 Next meetings  ===
TBA
January 9th 2012
 
[[Category:GridOpsMeeting]]

Latest revision as of 16:17, 29 November 2012

Main EGI.eu operations services Support Documentation Tools Activities Performance Technology Catch-all Services Resource Allocation Security

Detailed agenda: Grid Operations Meeting 13 December 2011 14h00 Amsterdam time

EVO direct link Pwd: gridops
EVO details Indico page

1. Middleware releases and staged rollout

1.1 EMI-1 release status

External wiki page provided by Cristina (EMI): link

1.2 Staged Rollout (Mario)

1.2.1 UMD1

UMD 1.4 freeze yesterday, now preparing the release to be out next Monday 19 Dec.

  • emi.cream.sl5.x86_64-1.13.3 with the following updates
    • emi.cemon.sl5.x86_64-1.13.3
    • emi.apel_parsers.sl5.x86_64-1.0.1
    • emi.blah.sl5.x86_64-1.16.3
  • emi.ge-utils.sl5.x86_64-1.0.0
  • emi.mpi.sl5.x86_64-1.1.0
  • emi.apel.sl5.x86_64-3.2.8
  • emi.storm.sl5.x86_64-1.8.0
  • emi.lcg-util.sl5.x86_64-1.11.19
  • Update to Globus 5.0.5 baseline provided by IGE, ensuring a continuing integration with EMI products in the EGI
    • ige.globus-default-security.sl5.x86_64-5.0.5
    • ige.globus-gridftp.sl5.x86_64-5.0.5
    • ige.globus-rls.sl5.x86_64-5.0.5
    • ige.globus-gsissh.sl5.x86_64-4.3.3

Products in the SW provisioning, candidates for UMD 1.5 by the 30th January 2012:

Products under staged rollout:

  • IGE.globus-myproxy.sl5.x86_64-5.4.4

To be verified:

  • EMI.unicore-uvos.sl5.x86_64-1.5.0
  • EMI.lb.sl5.x86_64-3.1.0
  • EMI.dpm.sl5.x86_64-1.8.2
  • EMI.lfc_mysql.sl5.x86_64-1.8.2
  • EMI.lfc_oracle.sl5.x86_64-1.8.2
  • IGE.gridsam.sl5.x86_64-2.3.0
  • IGE.gridway.sl5.x86_64-5.8.1

Also waiting for the EMI update 11 on the 15th of December, WMS in partiular is expected in this elease.

2. Operational Issues

2.1 GOCDB data scoping overview

Presentation: File:GocdbScopingHighLevel.pdf

2.2 Top BDII memory issue

A serious memory issue was reported in two GGUS tickets:

[ 76337 ]: "bdii tmpfs fills up and silently fails"

[ 73840 ]: "bdii memory leak?"

The RAMdisk usage of the service increases until the service occupy all the available memory. After that it needs to be restarted.

The BDII versions affected are:

  • gLite 3.2.11
  • EMI 1.0.0-1

No information about EMI 1.0.1 (Distributed with UMD 1.3), if it is affected or not.

First partial mitigation suggested by developer:

  • Configuration patch, comments the following directives:
    #set_flags DB_LOG_INMEMORY
    #set_flags DB_TXN_NOSYNC
    This reduces the memory grow, but does not solve the problem

Mitigation:

  • The only workaround for the problem, so far, is to restart the Top-BDII service.

2.3 Survey ongoing

Survey about the request for SLURM support in CREAM. Survey description

  • Deadline is tomorrow: Dec 14th.
  • Many sites answered (69), probably most of the sites not interested in SLURM, and happy with the list of LRMS supported by CREAM, did not answer (reasonably).

2.4 Middleware requirements gathering campaign

  • Requirements for ARC/gLite/GLOBUS/UNICORE.
  • The campaign is ongoing and it will be closed on January 15th, to let OMB discuss the requirements list.
  • instructions

3. AOB

3.2 Next meetings

January 9th 2012