VERITAS Volume Manager 3.5 User's Guide - VERITAS Enterprise Administrator (September 2004)

Troubleshooting
Volume Troubleshooting
Chapter 7 199
• Make sure that the physical disks are turned on, plugged in, and
attached to the computer. Repair any disk, controller, or cable
problems. Scan the disks on the system (Rescan) to make sure that
VxVM recognizes any newly attached hardware.
To return the RAID-5 volume to a healthy state, recover the volume.
• If there has been a partial or complete disk failure, either replace the
disk or move the affected subdisks to another disk.
Volume State: Unusable
The RAID-5 volume is unusable. This usually occurs when there is a
double disk failure (that is, when two subdisks in the same stripe or the
parity and one subdisk in a stripe are damaged or inaccessible). The
underlying disks may have failed or become inaccessible.
• Make sure that the physical disks are turned on, plugged in, and
attached to the computer. Repair any disk, controller, or cable
problems. Scan the disks on the system (Rescan) to ensure that
VxVM recognizes any newly attached hardware.
Run the following command to reattach the disks to their disk group:
# /etc/vx/bin/vxreattach [device_name]
To return the RAID-5 volume to a healthy state, recover the volume.
• If two of the volume disks have failed, the RAID-5 volume data is
unusable. Replace the failed disks and then try to restart the volume
by running the following command:
# vxvol -f start volume_name
Restore the volume from backup.
To prevent this problem in the future, enable logging for RAID-5
volumes.