This error message is only seen in Solaris10 U7 andlater release.

Why is this error message seen only in Solaris10 U7 andlater?
To protect VxVM device in Solaris, Sun added VxVM disk detectingfunction from Solaris 10U7, in case of some Solaris commands(dumpadm, etc...).If we try to specify VxVM device(swap volume in rootdg, etc..) as a dump device,this error message will be displayed.

Why is this error message seenonly in OSN(Operating System-based Naming) mode?
It is a Sun bug. BecauseSun is not using our VxVM device detecting APIs correctly, they compare thedmpnode name with OS device name, in EBN(Enclosure-based Naming) mode they arenot equal, so only in OSN mode it works and error message isdisplayed.


Workaround 1
Change the namingscheme toebn.

# vxddladm set namingscheme=ebn
# vxddladm getnamingscheme
NAMING_SCHEME      PERSISTENCE    LOWERCASE      USE_AVID
============================================================
EnclosureBased    Yes            Yes            Yes

#dumpadm
     Dump content: kernelpages
      Dump device: /dev/dsk/c0t0d0s1(dedicated)
Savecore directory:/var/crash/lab2
 Savecore enabled: yes
# dumpadm -d/dev/dsk/c0t1d0s1
     Dump content: kernelpages
      Dump device: /dev/dsk/c0t1d0s1(dedicated)
Savecore directory:/var/crash/lab2
 Savecore enabled:yes


Workaround 2
Use force update mode to changedump device with -u option:

# dumpadm -u -d/dev/dsk/c0t1d0s1

In force update mode Sun will not check if thedevice is in use as VxVM disk.