Quite a story. Makes me wonder what drives you added. WD Caviar Green is notorious for RAID issues, WD has special (expensive!) RE (Raid Edition) drives. Even Caviar Black is not recommended for RAID purposes.
I've seen issues with fluffy sata cables as well. Only one remedy: get proper ones. You don want to lose data just because of a cheap cable being intermittent.
RAID5 failure of one partition, not others
-
gerald_clark
- Posts: 10642
- Joined: 2005/08/05 15:19:54
- Location: Northern Illinois, USA
RAID5 failure of one partition, not others
Use RAID 6 or RAID 10. RAID 5 is dangerous.
-
nosebreaker
- Posts: 72
- Joined: 2010/08/09 16:10:26
Re: Failed RAID5 volume (see latest post) [solved]
All drives are Seagate Barracuda drives, 5900 RPM 2TB. The older ones aren't labeled "green", but the new model is named that.
I ordered replacement SATA-3 cables (with locking tab) from monoprice right away, and I am moving the data off it now and I plan to rebuild as RAID6. I only had 3 drives to start with, which is why I did RAID5 originally. Now that I have 6 (soon to be 8) I plan to use RAID6.
I ordered replacement SATA-3 cables (with locking tab) from monoprice right away, and I am moving the data off it now and I plan to rebuild as RAID6. I only had 3 drives to start with, which is why I did RAID5 originally. Now that I have 6 (soon to be 8) I plan to use RAID6.
-
nosebreaker
- Posts: 72
- Joined: 2010/08/09 16:10:26
Re: Failed RAID5 volume (see latest post) [solved]
Aw crap, new problem. There must be some sort of hardware failure here, now the RAID6 array has failed!
I might post this in a new thread, but when I try the mdadm repair method from above, it no longer works. It says things like:
mdadm: /dev/sdb has wrong super-minor.
mdadm: /dev/sdb3 has wrong super-minor.
mdadm: /dev/sdb2 has wrong super-minor.
mdadm: /dev/sdb1 has wrong super-minor.
And so on for all the drives, and at the end says:
mdadm: no devices found for /dev/md2
So I did the following:
mdadm --assemble /dev/md0 /dev/sda1 /dev/sdb1 /dev/sdc1 (etc with drives)
I did that for /dev/md1 and /dev/md2 as well, it was nice enough to tell me when I added the wrong drives with an error, and at the end of it all I have access to the data again. But in the process I damaged /dev/md0 and /dev/md1 which is my /boot and swap, but at least I have the data again. I damaged it by marking drives as failed to remove them form the array instead of just stopping the array. So don't do that!
I might post this in a new thread, but when I try the mdadm repair method from above, it no longer works. It says things like:
mdadm: /dev/sdb has wrong super-minor.
mdadm: /dev/sdb3 has wrong super-minor.
mdadm: /dev/sdb2 has wrong super-minor.
mdadm: /dev/sdb1 has wrong super-minor.
And so on for all the drives, and at the end says:
mdadm: no devices found for /dev/md2
So I did the following:
mdadm --assemble /dev/md0 /dev/sda1 /dev/sdb1 /dev/sdc1 (etc with drives)
I did that for /dev/md1 and /dev/md2 as well, it was nice enough to tell me when I added the wrong drives with an error, and at the end of it all I have access to the data again. But in the process I damaged /dev/md0 and /dev/md1 which is my /boot and swap, but at least I have the data again. I damaged it by marking drives as failed to remove them form the array instead of just stopping the array. So don't do that!