data disks on RCF

From: Flemming Videbaek (videbaek@sgs1.hirg.bnl.gov)
Date: Thu Jan 25 2001 - 15:27:06 EST

  • Next message: Betty McBreen: "brahms AFS problem!"

    When the additional datadisk data04-08 were released it was under the assumption, which most of you probably do not remember , that they migt disappear with short notice because the vendor MTI did not
    deliver a system with proper specs. 
    THIS WILL NOW HAPPEN ON MONDAY.
    
    I enclosed two messages one from todays weekly RCF meeting
    - minutes thanks to Betty- and and second describing the actions to be taken
    
    // Minutes
    
    2)MTI disk deal cancelled.
    ZCYZX Co. on board, having configured a system that has passed RCF's
    specs.
    After RCF obtains positive references, the system could be
    delivered ~4-6 weeks.
    The old disks have to be removed so some reboots will occur as
    controllers are
    swapped etc. Announcements will tell when.
    
    
    
    // from Maurice A
    
    1) the data01 and data03 filesystems are on datadirect disk and are
    safe.
    2) the data02 filesystem needs to be moved from vivant to gladiator disk
    3) the data04-08 will go away.
    
    the data02 filesystem needs to be moved. i'm doing some practice runs
    now
    to get the process down. it will require:
    1) dropping the vivant filesystems that we're not using to free up one
    of the two hba's that the vivants were using.
    2) hooking up the gladiator to the free hba.
    3) reboot to find the gladiator disk.
    4) put the gladiator disk into veritas and move the filesystem
    transparently. (tests appear that it takes about 20hrs to move the
    data in the background with little performance loss)
     
    so i might have to reboot once or twice to configure the gladiator
    disk raid on rmine001 but that should be the extent of interruption.
    /////
    
    This implies that we will loose for a while the data04-data08 disks. I do not think all data from data04-data08
    can be moved nor do I think it is needed.
    --  I will make sure the data04/FFSel directory created for the
    FS analysis are moved
    
    -- All other users (ejkim, beam, jhlee, rami, ourdane, ekman, nbi, sanders,hiro ) are requested to check if all of the many raw datafiles
    present on data01-data03 are really needed, or if reduced files are ok for this purpose.Please clean-up.
    Only you can do it due to protection.
    
    - Please beware you will only have until Monday noon to giveup/save/archive files form data04-data08.You do not have to remove any files on data04-data08 thye will disappear.
    
    
    
    
    - current usage
    
    data01 91% full  of (390G)
    data02 82% full  of (310G)
    data03 96% full of  (400G)
    
    ------------------------------------------------------
    Flemming Videbaek
    Physics Department
    Brookhaven National Laboratory
    
    tlf: 631-344-4106
    fax 631-344-1334
    e-mail: videbaek@bnl.gov
    



    This archive was generated by hypermail 2b29 : Thu Jan 25 2001 - 15:21:41 EST