×
Create a new article
Write your page title here:
We currently have 3,189 articles on s23. Type your article name above or create one of the articles listed here!



    s23
    3,189Articles

    Replacing A Meta Disk (Solaris disk suite)

    Revision as of 19:02, 14 April 2008 by imported>mutante (Category:Solaris)
    (diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

    A note[edit]

    Just a quick note to say that if you are replacing a failed disk with Sol9/10 and disk suite, you should detach and clear any submirrors on the disk that are not in 'needs maintenance state', otherwise disksuite will get in a mess wrt the serial numbers on the disks at some point in the future.

    Replace A disk[edit]

    If a disk has gone down and been replaced with a new one...

    the broken disk status would like this:

    # metastat d90
     
    d90: Mirror
       Submirror 0: d91
         State: Needs maintenance 
       Submirror 1: d92
         State: Okay         
       Pass: 1
       Read option: roundrobin (default)
       Write option: parallel (default)
       Size: 430007232 blocks
     
    d91: Submirror of d90
       State: Needs maintenance 
       Invoke: metareplace d90 c3t10d0s0 <new device>
       Size: 430007232 blocks
       Stripe 0: (interlace: 2048 blocks)
           Device      Start Block  Dbase State        Hot Spare
           c3t8d0s0           0     No    Okay         
           c3t9d0s0           0     No    Okay         
           c3t10d0s0          0     No    Maintenance  
     
    d92: Submirror of d90
        State: Okay         
        Size: 430007232 blocks
        Stripe 0: (interlace: 2048 blocks)
            Device      Start Block  Dbase State        Hot Spare
            c4t8d0s0           0     No    Okay         
            c4t9d0s0           0     No    Okay         
            c4t10d0s0          0     No    Okay   
    

    You should also check the metadb placement, eg:

    # usr/sbin/metadb
           flags           first blk       block count
        a m  p  luo        16              8192            /dev/dsk/c1t0d0s7
        a    p  luo        8208            8192            /dev/dsk/c1t0d0s7
        a    p  luo        16400           8192            /dev/dsk/c1t0d0s7
        a    p  luo        16              8192            /dev/dsk/c1t1d0s7
        a    p  luo        8208            8192            /dev/dsk/c1t1d0s7
        a    p  luo        16400           8192            /dev/dsk/c1t1d0s7
    

    There should be 3 copies of the metadb on each disk ideally, the flags will be different on a failing disk (if the disk is totally failing), use 'metadb -i' to see descriptions.

    Remove metadb's on the faulty disk before removing the disk, eg.

    # metadb -d /dev/dsk/c1t1d0s7
    


    Installing the new Disk[edit]

    label the disk[edit]

    use format to label the disk

    partition the disk[edit]
    # prtvtoc /dev/rdsk/c3t10d0s2
    

    this show the current partitions

    you can copy the structure form a similar disk ie:

    # prtvtoc /dev/rdsk/c4t10d0s2 | fmthard -s - /dev/rdsk/c3t10d0s2
    

    check the new structure

    # prtvtoc /dev/rdsk/c3t10d0s2
    * /dev/rdsk/c3t10d0s2 partition map
    *
    * Dimensions:
    *     512 bytes/sector
    *     424 sectors/track
    *      24 tracks/cylinder
    *   10176 sectors/cylinder
    *   14089 cylinders
    *   14087 accessible cylinders
    *
    * Flags:
    *   1: unmountable
    *  10: read-only
    *
    *                           First     Sector    Last
    * Partition  Tag  Flags    Sector     Count    Sector  Mount Directory
           0      0    00      10176 143339136 143349311
           2      5    01          0 143349312 143349311
           7      0    00          0     10176     10175
    

    Enable the new disk[edit]

    # metareplace -e d90 c3t10d0s0     
    d90: device c3t10d0s0 is enabled
    

    If metadb replicas already existed on the disk then replace them eg.

    # metadb -a -c 3 /dev/dsk/c1t1d0s0
    

    check the status[edit]

    the rebuild could take some time

    # metastat d90
    d90: Mirror
        Submirror 0: d91
          State: Resyncing    
        Submirror 1: d92
          State: Okay         
        Resync in progress: 0 % done
        Pass: 1
        Read option: roundrobin (default)
        Write option: parallel (default)
        Size: 430007232 blocks
    
    d91: Submirror of d90
        State: Resyncing    
        Size: 430007232 blocks
        Stripe 0: (interlace: 2048 blocks)
            Device      Start Block  Dbase State        Hot Spare
            c3t8d0s0           0     No    Okay         
            c3t9d0s0           0     No    Okay         
            c3t10d0s0          0     No    Resyncing    
    
    
    d92: Submirror of d90
        State: Okay         
        Size: 430007232 blocks
        Stripe 0: (interlace: 2048 blocks)
            Device      Start Block  Dbase State        Hot Spare
            c4t8d0s0           0     No    Okay         
            c4t9d0s0           0     No    Okay         
            c4t10d0s0          0     No    Okay
    
    Cookies help us deliver our services. By using our services, you agree to our use of cookies.

    Recent changes

  • BruderSpektrum • 3 months ago
  • BruderSpektrum • 3 months ago
  • BruderSpektrum • 3 months ago
  • BruderSpektrum • 3 months ago
  • Cookies help us deliver our services. By using our services, you agree to our use of cookies.