cancel
Showing results for 
Search instead for 
Did you mean: 

Disk Error

Mamady123
Not applicable
Dear All,

Below is the error messages  i am getting from my server

root@gtsun3
# vxdisk list
DEVICE       TYPE      DISK         GROUP        STATUS
c0t0d0s2     sliced    rootdisk     rootdg       online failing
c1t0d0s2     sliced    c1t0d0s2     rootdg       online
c9t0d0s2     sliced    -            -            error
c9t0d2s2     sliced    -            -            error
c9t0d3s2     sliced    -            -            error
c9t0d4s2     sliced    -            -            error

Could someone help me sort out this issue.

Your prompt response will be very helpful.

thanks in advanced.
1 ACCEPTED SOLUTION

Accepted Solutions

Gaurav_S
Moderator
Moderator
   VIP    Certified
Hello,

Regarding your first disk c0t0d0s2 , it shows a failing flag which indicates that there was an IO error while writing to public region of the disk...
For this, check /var/adm/messages for scsi errors (or transport errors), also you can check iostat

# iostat -En | grep -i hard   

If the disk shows good no. of hard errors, likely that disk is having issues & needs replacement, best would be to monitor hard/transport errors, if they are growing then disk has issues..  If the errors are not increasing, you can clear the failing flag by following command:

# vxedit -g <diskgroup> set failing=off <diskname>

Regarding your other disks which are on c9 controller, there could be multiple possibilities...

a) Check from storage if disks are presented correctly
b) Check whether disks have proper label on it  (prtvtoc /dev/rdsk/c9txdxsx), if it shows no label, then you need to label them (If there was existing data on the disk, there is a risk of data gone), once labelled, run "vxdctl enable" to rescan the disks.
c) Check multipaths of each disk, if all the paths to a disk are unavailable then disk will land in error state, you can check the same with

# vxdisk list c9txdxsx      (check at bottom for multipathing information, atleast one path should be enabled)
OR
# vxdisk path

d) Check if c9 controller is configured (cfgadm -al)  if not then configure it...
e) check if controller is enabled from DMP
# vxdmpadm listctlr all

if it is disabled then you can enable with

# vxdmpadm enable ctlr=c9

Hope this helps

Gaurav

View solution in original post

3 REPLIES 3

Gaurav_S
Moderator
Moderator
   VIP    Certified
Hello,

Regarding your first disk c0t0d0s2 , it shows a failing flag which indicates that there was an IO error while writing to public region of the disk...
For this, check /var/adm/messages for scsi errors (or transport errors), also you can check iostat

# iostat -En | grep -i hard   

If the disk shows good no. of hard errors, likely that disk is having issues & needs replacement, best would be to monitor hard/transport errors, if they are growing then disk has issues..  If the errors are not increasing, you can clear the failing flag by following command:

# vxedit -g <diskgroup> set failing=off <diskname>

Regarding your other disks which are on c9 controller, there could be multiple possibilities...

a) Check from storage if disks are presented correctly
b) Check whether disks have proper label on it  (prtvtoc /dev/rdsk/c9txdxsx), if it shows no label, then you need to label them (If there was existing data on the disk, there is a risk of data gone), once labelled, run "vxdctl enable" to rescan the disks.
c) Check multipaths of each disk, if all the paths to a disk are unavailable then disk will land in error state, you can check the same with

# vxdisk list c9txdxsx      (check at bottom for multipathing information, atleast one path should be enabled)
OR
# vxdisk path

d) Check if c9 controller is configured (cfgadm -al)  if not then configure it...
e) check if controller is enabled from DMP
# vxdmpadm listctlr all

if it is disabled then you can enable with

# vxdmpadm enable ctlr=c9

Hope this helps

Gaurav

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
Mamady123

Which version of VxVM? Up to version 3.5 it was quite normal for unitialized disks to show 'error' in status column.

As far as the failing status of the rootdisk is concerned, I fully agree with Gaurav.

cela19O5
Not applicable
Yeah sometimes I meet this problem but I do not know how to recover :(  It is not easy at all ...