Weekly Operations' Meeting Action List

Agenda: http://agenda.cern.ch/displayLevel.php?fid=258

 Status of 2005-05-09

Number

Description

Assigned To:

Status

2004-12-06--1

Sites should accelerate migration away from RH7.3 (and other non-secure OSs), at least on the service nodes due to security considerations. Deadline: End of April 2005.

Progress:
Frederic: French sites will be done by the end of May.
Gonzalo: 3 sites remains in SWE.

ROC mgrs

OPEN

2005-02-07--4

John Gordon volunteered, using some information circulated by Ian Bird, to define metrics on site performance.
Progress:
Min working on this now.

Nick

OPEN

2005-02-14--5

Merged with action 2005-02-14--4: Provide a page, probably hunging off the Regions' page) with a flag indicating whether a site had been down over a given period and foresee use of this info in the reporting template. Details in the action list of the 2005-02-28 meeting notes.

Piotr & Min

OPEN

2005-02-28--1

Sites to set the value of max_running_jobs to the number of available CPUs instead of 9999 for those cases when there is no limit. Progress will be monitored via the ROC managers' meeting. This is an ATLAS request.
Progress:
Read the notes of the Feb.28th meeting for the reasons of this request. Value 9999 is set to zero. Nick will check with Simone Campana whether Atlas is happy.

On 2005-05-02 meeting Simone reported that no more 9999 value is present but far too mamy zeroes, which means no jobs are accepted in that site. Gonzalo this is the default Info provider value. He suggested obliging the site to put a value. Simone will discuss it with the customers and the deployers.

Simone to Oliver??

OPEN

2005-03-14--1

Analyse the information submitted by the ROCs in the weekly reports over a number of weeks, to see what can be learnt from it.

Progress:
tailoring, filtering in the new form by Osman.

Nick

OPEN

2005-03-21--1

Resolve conflict of interests between the ROCs which make RCs available for VO support and VOs which desire to black/white list sites based on the Site Functional Test (SFT) results.
Progress:
Sven Hermann reminded the ROC managers list this should be discussed in their meeting of 2005-05-03.

ROC mgrs meeting & EIS-experiment meetings.

OPEN

2005-03-21--2

Document in the Operations Manual that, after initial contact with the site and the ROC, escalation will go to the ROC only (and no more to the site) by the CIC-on-duty.

Piotr

OPEN

2005-04-11--1

Lack of possibility for CICs to run the cron job that produces that Functional Site Reports (FST) daily is a problem.
Progress:
Now Frederic,Piotr and Judit have permissions for cvs updates, the cron changes automatically to incorporate recent changes committed into cvs.

Piotr & CICs

DONE Close at next meeting

2005-05-02--1

Change http://egee-docs.web.cern.ch/egee-docs/list.php?dir=.\operational_tools\& to point to https://cic.in2p3.fr/index.php?id=roc&roc_page=1

Nick

OPEN

2005-05-02--2

CMS requires a normalisation of OS used by the sites. Define agreed list of names and publish it with the next Release Notes.

Markus

OPEN

2005-05-02--3

Clarify the policy concerning VO data lost on a SE.

Nick

OPEN

2005-05-02--4

Make SFT easily configurable for VOs. Obtain critical SFTs per VO.

Piotr & EIS

OPEN

2005-05-09--1

Discuss site suspension and re-integration procedures.

Bologna LCG Workshop participants 2005-5-23

OPEN

Maria Dimou, IT/GD, Grid Infrastructure Services