RAID error

HI I have SunFire V880 and it has 12 HDD. This morning I saw this message in /var/adm/messages

Jan 4 07:35:42 etpdb1 md_raid: [ID 371651 kern.warning] WARNING: md d0: write error on /dev/dsk/c1t5d0s2 Jan 4 07:35:42 etpdb1 last message repeated 1 time Jan 4 07:35:43 etpdb1 md_raid: [ID 104909 kern.warning] WARNING: md: d0: /dev/dsk/c1t5d0s2 needs maintenance

# metastat -i

MD_IOCPROBE_DEV: Invalid argument

MD_IOCPROBE_DEV: Invalid argument

d0: RAID

State: Needs Maintenance

Invoke: metareplace d0 c1t5d0s2 <new device>

Interlace: 32 blocks

Size: 213369984 blocks

Original device:

Size: 213371808 blocks

Device Start Block Dbase StateHot Spare

c1t5d0s23219YesMaintenance

c1t4d0s23219YesOkay

c1t3d0s23219YesOkay

c1t2d0s23219YesOkay

d20: RAID

State: Okay

Interlace: 16 blocks

Size: 716685504 blocks

Original device:

Size: 716694800 blocks

Device Start Block Dbase StateHot Spare

c1t8d0s210346NoOkay

c1t9d0s210346NoOkay

c1t10d0s2 10346NoOkay

c1t11d0s2 10346NoOkay

c1t12d0s2 10346NoOkay

c1t13d0s2 10346NoOkay

Can anyone please advise the next step.

[1238 byte] By [tejaspandyaa] at [2007-11-26 14:04:05]
# 1

It looks like the disk /dev/dsk/c1t5d0s2 is going bad, or is having issues. You may need to replace the drive.

If you already have a hot spare drive, run the command

#metareplace d0 c1t5d0s2 c1t#d0s2 (# being the target of the spare).

If you don't have a hot spare, you can replace the bad drive with another of the same size, and resync the d0:RAID after the new drive is in place.

mlhardwickea at 2007-7-8 1:47:53 > top of Java-index,General,Sys Admin Best Practices...