We have a pair of Sunfire v20z's (WITH LSI MPT raid) that have each had one drive in their mirrors die.
[code]
bash-2.05# cat /etc/release
Solaris 9 4/04 s9x_u6wos_08a x86
Copyright 2004 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 22 March 2004
[/code]
So that's the machine type established. Let's take a look at our array health:
[code]
# raidctl
RAID Volume RAID RAID Disk
Volume Type Status Disk Status
------------------------------------------------------
c1t0d0 IM DEGRADED c1t0d0 OK
c1t1d0 FAILED
[/code]
Ok simple enough. Since these machines aren't hot swap, I powered one off and replaced the drive. When booting the machine it spits out some warning that the array is degraded and the kernel messages show this after the system boots:
[code]
May 13 15:21:54 ns-2.vancouver.ipapp.com scsi: [ID 365881 kern.info] /pci@0,0/pci1022,7450@a/pci17c2,10@4 (mpt0):
May 13 15:21:54 ns-2.vancouver.ipapp.com Rev. 8 LSI, Inc. 1030 found.
May 13 15:21:54 ns-2.vancouver.ipapp.com scsi: [ID 365881 kern.info] /pci@0,0/pci1022,7450@a/pci17c2,10@4 (mpt0):
May 13 15:21:54 ns-2.vancouver.ipapp.com mpt0 supports power management.
May 13 15:21:54 ns-2.vancouver.ipapp.com pcplusmp: [ID 637496 kern.info] pcplusmp: pci1000,30 (mpt) instance 0 vector 0x1b ioapic 0x3 intin 0x3 is bound to cpu 0
May 13 15:22:02 ns-2.vancouver.ipapp.com scsi: [ID 365881 kern.info] /pci@0,0/pci1022,7450@a/pci17c2,10@4 (mpt0):
May 13 15:22:02 ns-2.vancouver.ipapp.com mpt0 Firmware version v1.3.27.0 (IM/IME)
May 13 15:22:02 ns-2.vancouver.ipapp.com scsi: [ID 365881 kern.info] /pci@0,0/pci1022,7450@a/pci17c2,10@4 (mpt0):
May 13 15:22:02 ns-2.vancouver.ipapp.com mpt0: IOC Operational.
May 13 15:22:13 ns-2.vancouver.ipapp.com scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci1022,7450@a/pci17c2,10@4 (mpt0):
May 13 15:22:13 ns-2.vancouver.ipapp.com Volume 0 is degraded
[/code]
So it looks like everything is p-normal so far. BUT now when I run raidctl the array is GONE!
[code]
# raidctl
No RAID volumes found
[/code]
Oh both disks are visible to the system
[code]
# format
Searching for disks...done
AVAILABLE DISK SELECTIONS:
0. c1t0d0 <DEFAULT cyl 34999 alt 2 hd 16 sec 128>
/pci@0,0/pci1022,7450@a/pci17c2,10@4/sd@0,0
1. c1t1d0 <DEFAULT cyl 34999 alt 2 hd 16 sec 128>
/pci@0,0/pci1022,7450@a/pci17c2,10@4/sd@1,0
[/code]
The system knows the volume exists AND it's status but I cannot add the second drive to the mirror and fix the degraded array. Unix junkies, please help!