Castor external operations Face-to-Face Meeting

chaired by Ignacio Reguero (CERN, IT DEPARTMENT)
from to (Europe/Zurich)
at CERN ( 1-1-025 )
Material:
Go to day
  • Wednesday, 18 February 2009
    • 10:00 - 13:00 Site and development reports
      Reports from the four Castor sites, with a focus on plans for 2009 (and beyond).
      Points to be covered:
          - scaling up the deployments
            - number of instances, size of diskcaches, tape infrastructure
          - s/w upgrades
            - Castor s/w version
            - migration to SLC5, usage of 32bit architecture
      
      The report of the Development team will also focus on plans for 2009
      (and beyond). Items to cover:
          - planned releases
            - feature changes
            - time lines
          - support for the different software versions
      
      
      
      
      • 10:00 RAL Site report 30'
        Material: Slides powerpoint file pdf file
      • 10:30 CNAF Site report 30'
        Material: Slides pdf file
      • 11:00 CERN site report 30'
        Speaker: Olof Barring (CERN)
        Material: Slides powerpoint file pdf file
      • 11:30 ASGC Site report 30'
        Material: Slides pdf file
      • 12:00 Development report 30'
                 The report of the Development team will also focus on plans for 2009 (and beyond).
                 Points to be covered:
                  - planned releases
                    - feature changes
                    - time lines
                  - support for the different versions
        
        Material: Slides pdf file
      • 12:30 Discussion 30'
    • 13:00 - 14:00 Lunch break
    • 14:00 - 17:15 Castor monitoring
      The purpose of this session is to review how we can best
      learn from each other, given that we use different monitoring systems.
      Particularly when you move to a new release what monitoring changes -
      and why. Is it useful to inform each other of this, or do we all just
      find out by trial and error?
      • 14:00 Castor monitoring at RAL 30'
        Material: Slides powerpoint filedown arrow pdf filedown arrow
      • 14:30 Castor monitoring at CERN 30'
        Speaker: Miguel Marques Coelho Dos Santos (CERN)
        Material: Slides powerpoint file pdf filedown arrow
      • 15:00 Castor monitoring at CNAF 15'
      • 15:15 Castor monitoring at ASGC 15'
      • 15:30 Break 30'
      • 16:00 Castor monitoring developments 30'
        Speaker: Mr. Dennis Waldron (CERN)
        Material: Slides powerpoint file pdf file
      • 16:30 RAL monitoring developments 15'
         Under Castor monitoring developments (Feb 18 16:00), we'd like to
        include James Jackson's /      Brian Davies work on LSF queue
        monitoring.
        
        Speaker: Brian Davies (RAL Tier1)
      • 16:45 Discussion 30'
  • Thursday, 19 February 2009
    • 09:00 - 11:00 Tape session
      • 09:00 Tape operations at ASGC 20'
        Material: pdf pdf file
      • 09:20 Tape operations at CNAF 20'
      • 09:40 Tape operations at RAL 20'
        Material: Slides pdf file
      • 10:00 Tape operations at CERN 30'
        Speaker: Tim Bell (CERN)
        Material: Slides powerpoint file pdf file
      • 10:30 Developments in the tape area 30'
        Speaker: German Cancio Melia (CERN)
        Material: Slides powerpoint file pdf file
    • 11:00 - 11:30 break
    • 11:30 - 13:00 Database operations
      Gordon suggests:
      
      For the database session I think we need to look at the status and plans
      of each installation, discuss the current oracle issues and discuss more
      widely the way we monitor, tune and test. Specifically:
      
       * status of installations at each site (include CASTOR DB version,
      oracle configuration, database hardware, oracle patching
      status/schedule)
       * current oracle issues (and how we work with Oracle Corp to resolve
      them)
       * provision for test databases (and recreation of bugs/issues)
       * tuning and monitoring: how each site performs this
       * database capacity planning
       * discussion on how we can exchange knowledge
      
      I appreciate this is quite a lot for the time slots, but a quick summary
      of each site could lead to some time for a more open discussion on the
      above points. The ultimate goal for this session is that we go away
      knowing what we need to do help each other and work with Oracle to solve
      any issues.
      
      
      • 11:30 Database operations at CERN 30'
        Speaker: Nilo Segura Chinchilla (CERN)
        Material: Paper pdf file Slides powerpoint file
      • 12:00 Database operations at RAL 30'
        Material: Slides powerpoint file pdf file
      • 12:30 Database operations at ASGC 15'
      • 12:45 Database operations at CNAF 15'
        Material: Slides powerpoint file pdf file
    • 13:00 - 14:00 Lunch break
    • 14:00 - 15:30 Diskserver deployment
      The purpose of the disk server deployment session is to understand how
      and why each site does what it does, and to share understanding to
      increase our ability to move towards fully automated deployment.
      
      • 14:00 Diskserver deployment at CERN 30'
        Speaker: Mr. Ignacio Reguero (CERN, IT DEPARTMENT)
        Material: Slides powerpoint file pdf filedown arrow
      • 14:30 Diskserver deployment at RAL 30'
        Material: Slides powerpoint file pdf file
      • 15:00 Diskserver deployment at ASGC 15'
      • 15:15 Diskserver deployment at CNAF 15'
    • 15:30 - 16:00 Load testing
      The purpose of Load testing session is to agree how to improve the test
      certification to include load as well as functional testing, what this
      will involve by way of effort and technology, whose responsibility it is
      and where it will happen ;-)
      
      • 15:30 Castor loadtesting at CERN 15'
        Speaker: Mr. Dennis Waldron (CERN)
        Material: Slides powerpoint file pdf file
      • 15:45 Discussion 15'