Please enter search query.
Search <product_name> all support & community content...
Article: 100022142
Last Published: 2019-04-03
Ratings: 0 0
Product(s): InfoScale & Storage Foundation
Problem
Downgrade diskgroup version using CBR data
Error Message
VxVM vxdg ERROR V-5-1-587 Disk group dwgdg: import failed: Disk group has no valid configuration copiesSolution
There are situations that a Veritas Volume Manager (VxVM) disk group version needs to be downgraded, in order to import the disk group on a server running an older version of VxVM.
With current and past VxVM releases, there is not a single command available to downgrade the diskgroup version.
The most commonly used method to downgrade the diskgroup version is to "rebuild" the diskgroup.
This article introduces another way to downgrade the disk group version using VxVM cbr (cbr stands for configuration backup restore) data and utility vxconfigrestore.
A disk group called dwddg is used as an example below in this document.
In the example, the VxVM version was upgraded from 4.1 to 5.0. The disk group version was upgraded to 140 from 120 after completing VxVM upgrade..
Then user realized that the disk group version needs to be downgraded back to120.
When tried to import the disk group on the server running VxVM 4.1, received the error below.
# vxdg import dwddg
VxVM vxdg ERROR V-5-1-587 Disk group dwddg: import failed: Disk group has no valid configuration copies
Attempt to Import the diskgroup with -Cf options also failed with the same error
# vxdg -Cf import dwddg
VxVM vxdg ERROR V-5-1-587 Disk group dwddg: import failed: Disk group has no valid configuration copies
Startingfrom VxVM 3.5MP4, a new feature (called cbr) is introduced in VxVM. This new feature not only backups up VxVM configuration, it also enables user to use the cbr data backed up (in /etc/vx/cbr/bk directory) to restore the VxVM configuration.
The procedures below show how the diskgroup version of dwddg is downgraded from 140 to 120..
1. check and make sure that the server is running VxVM with default diskgroup version 120(VxVM 4.1).
#modinfo |grep vx
2. check and make sure that VxVM can access all the disks for the disk group by running the two commands below
# vxdisk -o alldgs list |grep dwddg
c5t14d0s2 auto:sliced - (dwddg) online
c5t14d22s2 auto:sliced - (dwddg) online
and
#cat/etc/vx/cbr/bk/dwddg.1189258592.64.hd7d01h0/1189258592.64.hd7d01h0.cfgrec |vxprint -D - -ht | grep dm
dmhdsdwd01 c5t14d0s2 auto 7423 71162880 -
dmhdsdwd02 c5t14d22s2 auto 7423 71162880 -
It is important that the first command output shows the same number of the disks as the second command output, in this example 2.
3. run thecommand below to rebuild the disk group configuration with version 120
#/usr/lib/vxvm/bin/vxconfigrestore -p dwddg
Diskgroup dwddg configuration restoration started ......
dwddg'sdiskgroup configuration is restored (in precommit state).
Diskgroup can be accessed in read only and can be examined using
vxprint in this state.
Run:
vxconfigrestore-c dwddg ==> to commit the restoration.
vxconfigrestore-d dwddg ==> to abort the restoration.
4. check the disk group configuration restored:
# vxprint-htg dwddg
DGNAME NCONFIG NLOG MINORS GROUP-ID
STNAME STATE DM_CNT SPARE_CNT APPVOL_CNT
... snip...
DCNAME PARENTVOL LOGVOL
SPNAME SNAPVOL DCO
dgdwddg default default 119000 1189258592.64.hd7d01h0
dmhdsdwd01 c5t14d0s2 auto 7423 71162880 -
dmhdsdwd02 c5t14d22s2 auto 7423 71162880 -
v vloracle_102- ENABLED ACTIVE 8388608 SELECT - fsgen
plvloracle_102-01 vloracle_102 ENABLED ACTIVE8388608 CONCAT - RW
sdhdsdwd02-01 vloracle_102-01 hdsdwd02 0 8388608 0 c5t14d22 ENA
v vlstage - ENABLED ACTIVE 14680064 SELECT - fsgen
plvlstage-01 vlstage ENABLED ACTIVE 14680064 CONCAT - RW
sdhdsdwd01-01 vlstage-01 hdsdwd010 14680064 0 c5t14d0 ENA
v vol_oracle - ENABLED ACTIVE 10207232SELECT - fsgen
plvol_oracle-01 vol_oracle ENABLED ACTIVE 10207232CONCAT - RW
sdhdsdwd01-06 vol_oracle-01 hdsdwd01 60817408 102072320 c5t14d0 ENA
v voldwdasdev - ENABLED ACTIVE 12582912 SELECT - fsgen
plvoldwdasdev-01 voldwdasdev ENABLED ACTIVE 12582912CONCAT - RW
sdhdsdwd01-05 voldwdasdev-01 hdsdwd01 48234496 125829120 c5t14d0 ENA
v voldwdasprd - ENABLED ACTIVE 12582912 SELECT - fsgen
plvoldwdasprd-01 voldwdasprd ENABLED ACTIVE 12582912CONCAT - RW
sdhdsdwd01-02 voldwdasprd-01 hdsdwd01 14680064 125829120 c5t14d0 ENA
v voldwdastst - ENABLED ACTIVE 16777216 SELECT - fsgen
plvoldwdastst-01 voldwdastst ENABLED ACTIVE 16777216CONCAT - RW
sdhdsdwd01-03 voldwdastst-01 hdsdwd01 27262976 167772160 c5t14d0 ENA
v voldwdrepdt - ENABLED ACTIVE 4194304 SELECT - fsgen
plvoldwdrepdt-01 voldwdrepdt ENABLED ACTIVE 4194304 CONCAT - RW
sdhdsdwd01-04 voldwdrepdt-01 hdsdwd01 44040192 41943040 c5t14d0 ENA
5. check the disk group version before committing.
# vxdg list dwddg |grep -i version
version: 120
6. commit the (downgraded) configuration restored
# vxconfigrestore -c dwddg
Committing configuration restoration for diskgroup dwddg ....
dwddg's diskgroup configuration restoration is committed.
The diskgroup can now be deported and imported normal with disk group version downgraded to 120.
A successful diskgroup downgrade is based on the following two:
1. diskgroup version in cbr data is correct
2. VxVM version - the default diskgroup version must be the diskgroup version to be downgraded to.
The default diskgroup versions for each VxVM versions are:
VxVMVersion Disk Group Layout Version
3.5 90
4.0 110
4.1 120
5.0 140