VERITAS Volume Manager 4.1 Release Notes (5900-0591, March 2010)

VERITAS Volume Manager 4.1 Release Notes
Known Problems and Workarounds
Chapter 128
If a disk fails while configuring VxVM with the vxconfigd daemon, the following problems
occur:
• The VxVM administrative commands cannot be run on the node on which the disk has
failed.
• Service groups timeout and fail in a clustered or HA environment as the VERITAS
Cluster Server agents cannot communicate with vxconfigd daemon to determine the
health of VxVM components.
Also, there is no way to communicate with the vxconfigd daemon through signals or native
interprocess communication.
DMP Issues
Action Required After Swapping Fibre Channel Paths to an Array
If the Fibre Channel paths to an array become swapped, either by accident or design, this
changes the nportid setting for the disks in that array. You must use the fcmsutil command
(see the fcmsutil(1M) manual page) to reconfigure the disks before attempting to rediscover
the VxVM configuration.
If the disks are not reconfigured using the fcmsutil command, the ioscan, insf-e and
vxdctlenable commands may either take a very long time to complete, or they may hang.
Messages should also be visible in /var/adm/syslog/syslog.log to indicate that the
fcmsutil command should be used to correct the problem.
Take the following steps if the paths have become swapped:
1. Use the following command to obtain the N_Port_id settings for the affected paths:
# fcmsutil TD_device_file get remote all | grep N_Port_id
This is illustrated by the following example:
# fcmsutil /dev/td0 get remote all | grep N_Port_id
Target N_Port_id is = 0x00006b
Target N_Port_id is = 0x00006c
Target N_Port_id is = 0x00006d
Target N_Port_id is = 0x00006e
Target N_Port_id is = 0x000071
Target N_Port_id is = 0x000072
Target N_Port_id is = 0x000073
Target N_Port_id is = 0x000074