When dmp failover does not occur, or seems to hang, check for the following error in the messages file:
"rcm_daemon[13297]: [ID 546518 daemon.error] rcm script es_rcm.pl: VxVM vxdmpadm ERROR V-5-1-2395 Invalid arguments"
Check the "dmpnode" section in the dmppolicy.info file for "enable_all 1" as this setting disables dmp path failover.
(/etc/vx) # cat /etc/vx/dmppolicy.info
arraytype
#
arrayname
#
enclosure
#
naming
scheme=ebn persistence=yes mode=- lowercase=yes use_avid=yes
#
dmpnode
enable_all 1
#
IMPORTANT: If running VCS, you must freeze all service groups first before restarting vxconfigd to prevent possible failover.
Move the dmppolicy.info file aside, and restart vxconfigd.
# mv /etc/vx/dmppolicy.info /etc/vx/dmppolicy.info.orig
# vxconfigd -k
Once vxconfigd has been restarted, you may change the namescheme to OSN or EBN and recreate the dmppolicy.info file as shown below.
To have OS native names in "vxdisk list" run the following command to change to OSN:
# vxddladm set namingscheme=osn persistence=yes
or
To have Enclosure Based Names in "vxdisk list" run the following command to change to EBN:
# vxddladm set namingscheme=ebn persistence=yes