Got an alert from one of my log parsing scripts that there was something wrong with my RAID (on a remote box) after a reboot, so logged in to check. From dmesg:
Code:
ad4: 152627MB <SAMSUNG SP1614C SW100-34> at ata2-master SATA150
ad6: 152627MB <SAMSUNG SP1614C SW100-34> at ata3-master SATA150
ad8: 152627MB <SAMSUNG SP1614C SW100-34> at ata4-master SATA150
ad10: 152627MB <SAMSUNG SP1614C SW100-34> at ata5-master SATA150
ar0: 305255MB <FreeBSD PseudoRAID RAID0+1 (stripe 64 KB)> status: DEGRADED
ar0: disk0 READY (master) using ad4 at ata2-master
ar0: disk1 DOWN no device found for this subdisk
ar0: disk2 DOWN no device found for this subdisk
ar0: disk3 READY (mirror) using ad10 at ata5-master
OK, so my RAID0+1 is degrade, icky. Best check what's going on and rebuild:
Code:
# atacontrol status ar0
ar0: ATA RAID0+1 stripesize=128 status: DEGRADED
subdisks:
0 ad4 ONLINE
1 ---- MISSING
2 ---- MISSING
3 ad10 ONLINE
# atacontrol rebuild ar0
atacontrol: ioctl(IOCATARAIDREBUILD): Input/output error
Hmmm, check some forums and attempt to detach one of the faulty disks (which atacontol info has no problem seeing):
Code:
# atacontrol detach ata3
And the system crashes, requiring a remote power cycle.
At this point I discover someone's swiped my remote serial console, and much swearing occurs. Get all this sorted, and the box restarted, only to find the RAID is REALLY unhappy now, and has two arrays:
Code:
ar0: ATA RAID0+1 stripesize=128 status: BROKEN
subdisks:
0 ad4 ONLINE
1 ---- MISSING
2 ---- MISSING
3 ---- MISSING
ar1: ATA RAID0+1 stripesize=128 status: BROKEN
subdisks:
1 ---- MISSING
2 ---- MISSING
3 ---- MISSING
4 ad10 ONLINE
Any suggestions on steps I can take to get the array recognised again?