NERSC File System Outage Wednesday, May 30

From: David Turner (dpturner_at_lbl_dot_gov)
Date: 05/23/2007

  • Next message: Katie Antypas: "Fast turn around for short large node jobs on Seaborg"
    Greetings NERSC User,
    
    There will an outage next week that will affect all NERSC compute systems.
    This note provides a general overview of the outage; subsequent messages
    will provide additional details.
    
    On Wednesday, May 30, at 10:00 AM PDT, the NERSC Global Filesystem (NGF)
    will be taken down for a 3-hour maintenance.  NGF provides project directories 
    (/project/projectdirs) to all the compute platforms.  In order to take NGF
    down "cleanly", we will first need to drain the compute systems of all
    running batch jobs.  Any interactive processes that are accessing NGF will
    then be terminated.  After removing NGF from the compute platforms, we will
    allow each system to resume scheduling batch jobs that do not require NGF.
    
    Identifying which jobs can safely be started during the NGF outage will probably
    require some action on the part of the users.  We are still investigating the
    best way to accomplish this, and we will notify NERSC users of the procedure
    well in advance of the outage.
    
    -- 
    Best regards,
    
    David Turner
    User Services Group        email: dpturner_at_lbl_dot_gov
    NERSC Division             phone: (510) 486-4027
    Lawrence Berkeley Lab        fax: (510) 486-4316
    

  • Next message: Katie Antypas: "Fast turn around for short large node jobs on Seaborg"

    This archive was generated by hypermail 2.1.6 : 05/23/2007 PDT