MinutesJun6

Introduction

• Last meeting: First in series.
• Previous meetings: IntegrationProgram
• Coordinates: 1:00pm Eastern, 1-510-665-5437 #24743

Attending

• Meeting attendees: Michael, Rob, Joe, Charles, Fred, Yuri, Kaushik, Horst, Xin, Rich Carlson, Wensheng, Jay Packard, Bob, Wei, Kristy, Dantong, Karthik, ...
• Apologies: none

Integration Program: Review of Phase 1 Milestones

• See IntegrationProgram
• SiteCertificationP1
• Michael: Looking for a more homogeneous approach as we move towards common goals, in this case towards the full dress rehearsal.

SRM issue

• Dantong: Is it a requirement? Wei: Storm implementation from INFN. Michael notes that this does include Storm - being worked on in the WLCG - its in their task interface, including SRM v2.
• Dantong will make a propoals for capabilities as a function of time.

OSG 0.6 Deployment

• OSGservicesP1
• BU - no report
• BNL - Xin has upgraded _1, plans to do the other one shortly.
• MWT2_IU, CE upgraded; _UC, this week.
• UC_ATLAS_MWT2 - done
• UC_Teraport - in progress
• OU - upgraded OSCER; holding off on the _SWT2 cluster: July 5.
• SLACXRD - before T2 meeting
• UTA - need update.

Logging: Syslog-ng upgrade

• LoggingServicesP1 - describes VDT-based syslog-ng install.
• Need to add FTS logfile from Hiro

DQ2 0.3 deployment planning

• Alexei - discussion is just starting as to the deployment. The migration scenario is the main question. A major stoppage for two days. According to Miguel - no problem for LCG sites. By end of the week, we'll know the situation.
• There will be a testbed between BNL and UTA. Wensheng has some information from Mario about the client - it uses YUM.
• Hiro will be going to CERN to work with Miguel on re-packaging. Will be packaged as rpms.

ATLAS releases via Panda

• Report from Torre via Fred: Tadashi is working on this and it is well along.
• Fred will also have information at the Tier2/3 workshop on how to validate your local release.

Monitoring: Mysql

• MysqlMonitoringServicesP1
• Yuri - a service at BNL that collects information about LRC's at sites - threads/queueries, etc. Tool was in use at BNL to measure server performance, also at CERN - the central DQ2 server. Want to add all Tier2 services.

Monitoring: Nagios

• Overview of currently monitored services, and integration with RT
• Will provide a simple table describing what/frequency/action.
• Need to make Nagios monitoring pages visible by all Facility participants via authentication - Tomasz will consult local experts.
• LRC monitoring using a standard plugin for databases. 16 LRCs are being monitored at the moment.
• Plugin sharing - create a plugin repository.

Tier3/Tier2 workshop

• http://indico.cern.ch/conferenceDisplay.py?confId=15523
• Note change of Tier3 / Tier2 days.

AOB

• Production - there is a lack of jobs generally due to data slowness out of BNL. Initial indication is that files needed are not on disk.
• AOD replication question - at SLAC, archive bit is not being set. Will consult Hiro.

Summary of Actions

• Tomasz will consult local experts for off-site Nagios console access
• Tomasz will take first steps towards creating a Nagios plugins respository
• All: sites to upgrade syslog-ng installation
• Develop load tests requirements and agenda (for LoadTestsP1) - Michael, Rob, ...
• Develop DQ2 0.3 deployment plan for US ATLAS facilities - Michael, Rob, Alexei, ...
• All sites: continue with OSG 0.6 installations as time permits
• All sites: check off actions taken in SiteCertificationP1
• Follow-up on AOD replication question about archival bit not being set - Hiro, Michael

-- RobertGardner - 05 Jun 2007