How to recover volume
Created: 16 Nov 2010 | Updated: 09 Dec 2010 | 12 comments
This issue has been solved. See solution.
df -k in solaris, vol01 was I/O error. After replaced diskarray controller. The vol01 still can not been used. How to do next using vxvm to recover volume? Maybe use command vxrecover?
Thank you
Discussion Filed Under:
Comments 12 Comments • Jump to latest comment
Hello Kelvin,
there could be multiple things...
-- what type of volume it is ? mirror/stripe/concat ?
-- whether the Luns of the volume are still intact ?
-- do you see any disk/lun in failed state ?
can you paste what is the current state of the volume ?
# vxprint -qthg <diskgroup> <volume>
vxrecover may work if all the disks are visible in the background (no failed disk)
Depending on volume state will let you know the commands to recover..
Gaurav
PS: If you are happy with the answer provided, please mark the post as solution. You can do so by clicking link "Mark as Solution" below the answer provided.
Hi Gaurav,
Thank you for your past reply
type:concat
RAID-1
All the status of dg were enabled and all the disks were good, so I think that we should update the file system with vx command.
After vxrecover, the vol01 was still I/O error in "df -k"
After replace diskarray controller, should I update the volo01 status? Thank you
Kelvin
Hello Kelvin,
little confused, RAID-1 is a mirror .. it can not be concat...
can you paste the vxprint output which I asked above ?
to update the state in "df -k", you will need to unmount & re-mount the filesystem.. but before doing that, I would like to ensure the state of volume is healthy... a healthy volume is indicated in vxprint output with "Enabled Active" state
Gaurav
PS: If you are happy with the answer provided, please mark the post as solution. You can do so by clicking link "Mark as Solution" below the answer provided.
Hello Gaurav,
Sorry for wrong choice.
The system was RAID 1+0.
vxprint like the following:
dm bgw1dgc4t40d0 - - - - NODEVICE
dm bgw1dgc5t40d0 c3t40d0s2 auto 16063 857028480 -
pl vol01-01 vol01 DISABLED NODEVICE 2550603840 CONCAT - RW
sd bgw1dgc3t40d1-01 vol01-01 bgw1dgc3t40d1 0 836546880 0 c1t40d1 ENA
sd bgw1dgc4t40d0-01 vol01-01 bgw1dgc4t40d0 0 857028480 836546880 - NDEV
sd bgw1dgc5t40d0-01 vol01-01 bgw1dgc5t40d0 0 857028480 1693575360 c3t40d0 ENA
DEVICE TYPE DISK GROUP STATUS
c0t0d0s2 auto:sliced rootdisk rootdg online
c0t1d0s2 auto:sliced rootmirror rootdg online
c1t40d0s2 auto:sliced ora1dgc3t40d0 ora1dg online
c1t40d1s2 auto:sliced bgw1dgc3t40d1 bgw1dg online
c2t40d0s2 auto:sliced - - online
c3t40d0s2 auto:sliced bgw1dgc5t40d0 bgw1dg online
- - bgw1dgc4t40d0 bgw1dg failed was:c2t40d0s2
Hello Kelvin
ok, so that clears some smoke ....
Now you can see, your volume is still disabled active .. that means it is not ready to serve your filesystem... & reason for this is, plex is in nodevice state... that means one of the disk is missing.... & missing disk is second subdisk which is flagged as NDEV
v vol01 - DISABLED ACTIVE 2550603776 SELECT - fsgen
pl vol01-01 vol01 DISABLED NODEVICE 2550603840 CONCAT - RW
sd bgw1dgc3t40d1-01 vol01-01 bgw1dgc3t40d1 0 836546880 0 c1t40d1 ENA
sd bgw1dgc4t40d0-01 vol01-01 bgw1dgc4t40d0 0 857028480 836546880 - NDEV
sd bgw1dgc5t40d0-01 vol01-01 bgw1dgc5t40d0 0 857028480 1693575360 c3t40d0 ENA
& you can see disk is reported as failed in "vxdisk list"
bgw1dgc4t40d0 bgw1dg failed was:c2t40d0s2
however same disk is also shown as online in vxdisk list
c2t40d0s2 auto:sliced - - online
Can you run following commands & show me outcomes:
# vxdisk -e list
# vxreattach -c c2t40d0s2
# vxdisk -e list
# vxdisk -o alldgs list
Gaurav
PS: If you are happy with the answer provided, please mark the post as solution. You can do so by clicking link "Mark as Solution" below the answer provided.
Hi Gaurav,
There was a disk broken, and I had replaced it before.
After replaced it, "bgw1dgc4t40d0 bgw1dg failed was:c2t40d0s2" still in vxdisk list
We need to re-import disk c2t40d0s2 into vxvm?
Hi Kelvin,
well if you have replaced disk & its a concat volume, you will not get the data ... you will need to restore the data from some backup...
you can put the new disk back into the diskgroup with the same name so your volume will be reconstructed but for sure if c2t40d0s2 is the newly replaced disk, you will not get data into vol01 volume... are you ok with that ?
That's the reason I asked you in first comment, if the failed disk is available again (i.e same disk with same data -untouched disk) then you can get data back, but if disk is replaced, data will not be there
Gaurav
PS: If you are happy with the answer provided, please mark the post as solution. You can do so by clicking link "Mark as Solution" below the answer provided.
Hi Gaurav,
Thanks for your help. I am doing some I.N. research. And the system need vcs.
It is a great help.
sorry but I am not clear with actions/objective ....
you are saying you need VCS, do you want to configure this volume in VCS ?
what about volume, do you want to re-enable the volume (data loss ofcourse) or you will rebuild your filesystem on your own ?
Gaurav
PS: If you are happy with the answer provided, please mark the post as solution. You can do so by clicking link "Mark as Solution" below the answer provided.
Yes, I wanna configure this volume in VCS and after replace the disk-array, I re-enable it and rebuild it by myself.
Great... have a look on VCS Users guide... that will give you good instructions on how to configure VCS ...
You can find the guide here:
https://sort.symantec.com/documents/doc_details/sf...
Gaurav
PS: If you are happy with the answer provided, please mark the post as solution. You can do so by clicking link "Mark as Solution" below the answer provided.
Moved to Storage Foundation forum as the original query was regarding how to recover vxvm volume(s)
If this post has helped you, please vote or mark as solution
Would you like to reply?
Login or Register to post your comment.