Sun Cluster 3.0 Hardware Guide

How to Replace a StorEdge A5x00 Disk Drive

Use this procedure to replace an StorEdge A5x00 disk array disk drive. "Example--Replacing a StorEdge A5x00 Disk Drive" shows you how to apply this procedure. Perform the steps in this procedure in conjunction with the procedures in Sun Cluster 3.0 System Administration Guide and your server hardware manual. Use the procedures in your server hardware manual to identify a failed disk drive.

For conceptual information on quorum, quorum devices, global devices, and device IDs, see Sun Cluster 3.0 Concepts.

  1. Identify the disk drive that needs replacement.

    If the disk error message reports the drive problem by device ID (DID), use the scdidadm -l command to determine the Solaris logical device name. If the disk error message reports the drive problem by the Solaris physical device name, use your Solaris documentation to map the Solaris physical device name to the Solaris logical device name. Use this Solaris logical device name and DID throughout this procedure.


    # scdidadm -l deviceID
    
  2. Determine if the disk drive you want to replace is a quorum device.


    # scstat -q
    

    If the disk drive you want to replace is a quorum device, remove the quorum device before you proceed. Otherwise, proceed to Step 3.

    Refer to Sun Cluster 3.0 System Administration Guide for procedures on replacing a quorum device and putting a quorum device into maintenance state.

  3. If possible, back up the metadevice or volume.

    For more information, see your Solstice DiskSuite or VERITAS Volume Manager documentation.

  4. Perform volume management administration to remove the disk drive from the configuration.

    For more information, see your Solstice DiskSuite or VERITAS Volume Manager documentation.

  5. Identify the failed disk drive's physical DID.

    Use this physical DID in Step 14 to verify that the failed disk drive has been replaced with a new disk drive. The DID and the World Wide Name (WWN) for the disk drive should be the same.


    # scdidadm -o diskid -l cNtXdY
    
  6. If you are using Solstice DiskSuite as your volume manager, save the disk partitioning for use when partitioning the new disk drive.

    If you are using VERITAS Volume Manager, proceed to Step 7.


    # prtvtoc /dev/rdsk/cNtXdYsZ > filename
    
  7. On any node connected to the disk array, run the luxadm remove command.


    # luxadm remove -F /dev/rdsk/cNtXdYsZ
    
  8. Replace the failed disk drive.

    For more information, see the documentation that shipped with your disk array.

  9. On any node connected to the disk array, run the luxadm insert command.


    # luxadm insert boxname,rslotnumber
    # luxadm insert boxname,fslotnumber
    

    If you want to insert a front disk drive, use the fslotnumber parameter. If you want to insert a rear disk drive, use the rslotnumber parameter.

  10. On all other nodes attached to the disk array, run the devfsadm(1M) command to probe all devices and to write the new disk drive to the /dev/rdsk directory.

    Depending on the number of devices connected to the node, the devfsadm command can take at least five minutes to complete.


    # devfsadm
    
  11. If you are using Solstice DiskSuite as your volume manager, on one node connected to the disk array, partition the new disk drive, using the partitioning you saved in Step 6.

    If you are using VERITAS Volume Manager, proceed to Step 12.


    # fmthard -s filename /dev/rdsk/cNtXdYsZ
    
  12. One at a time, shut down and reboot the nodes connected to the disk array.


    # scswitch -S -h nodename
    # shutdown -y -g 0 -i 6
    

    For more information, see Sun Cluster 3.0 System Administration Guide.

  13. On any of the nodes connected to the disk array, update the DID database.


    # scdidadm -R deviceID
    
  14. On any node, confirm that the failed disk drive has been replaced by comparing the following physical DID to the physical DID in Step 5.

    If the following physical DID is different from the physical DID in Step 5, you successfully replaced the failed disk drive with a new disk drive.


    # scdidadm -o diskid -l cNtXdY
    
  15. On all nodes connected to the disk array, upload the new information to the DID driver.

    If a volume management daemon such as vold is running on your node, and you have a CD-ROM drive connected to the node, a device busy error might be returned even if no disk is in the drive. This error is an expected behavior.


    # scdidadm -ui
    
  16. Perform volume management administration to add the disk drive back to its diskset or disk group.

    For more information, see your Solstice DiskSuite or VERITAS Volume Manager documentation.

  17. If you want this new disk drive to be a quorum device, add the quorum device.

    Refer to Sun Cluster 3.0 System Administration Guide for the procedure on adding a quorum device.

Example--Replacing a StorEdge A5x00 Disk Drive

The following example shows how to apply the procedure for replacing an StorEdge A5x00 disk array disk drive.


# scstat -q
# scdidadm -l d4
4        phys-schost-2:/dev/rdsk/c1t32d0 /dev/did/rdsk/d4
# scdidadm -o diskid -l c1t32d0
2000002037000edf
# prtvtoc /dev/rdsk/c1t32d0s2 > /usr/tmp/c1t32d0.vtoc 
# luxadm remove -F /dev/rdsk/c1t32d0s2
WARNING!!! Please ensure that no filesystems are mounted on these device(s). All data on these devices should have been backed up.

The list of devices that will be removed is:  1: Box Name "venus1" front slot 0

Please enter 'q' to Quit or <Return> to Continue: <Return>

stopping:  Drive in "venus1" front slot 0....Done
offlining: Drive in "venus1" front slot 0....Done

Hit <Return> after removing the device(s). <Return>

Drive in Box Name "venus1" front slot 0
Logical Nodes being removed under /dev/dsk/ and /dev/rdsk:
        c1t32d0s0
        c1t32d0s1
        c1t32d0s2
        c1t32d0s3
        c1t32d0s4
        c1t32d0s5
        c1t32d0s6
        c1t32d0s7

# devfsadm
# fmthard -s /usr/tmp/c1t32d0.vtoc /dev/rdsk/c1t32d0s2
# scswitch -S -h node1
# shutdown -y -g 0 -i 6
# scdidadm -R d4
# scdidadm -o diskid -l c1t32d0
20000020370bf955
# scdidadm -ui