DegradedArray event on /dev/md1

Melonheadjr44 · 11-03-2009, 09:36 AM

Hi,

For the past weeks, when restarting the server, I get this email about a degradedArray. So we bought a new harddrive. But I am unsure which one to replace and how to do so.

Also, when booting the server right now, it automatically goes on hard disk 0 and then the server doesnt boot, so I have to select the boot device and boot on hard disk 1 for our server to boot up.

Here's the email error message I get:

Quote:

A DegradedArray event had been detected on md device /dev/md1.

Faithfully yours, etc.

P.S. The /proc/mdstat file currently contains the following:

Personalities : [raid1]
md1 : active raid1 sdb2[0]
290953088 blocks [2/1] [U_]

md0 : active raid1 sdb1[0] sda1[1]
2008000 blocks [2/2] [UU]

unused devices: <none>

Why is it md1 that is degraded if it has more blocks? What do I do from here?

Thanks

bertl · 11-04-2009, 12:02 PM

It's more likely to come across a bad sector because it's larger, basically. The fact that md0 is still in one piece seems to indicate your disk as such is still working (spinning, delivering data) but you might want to try reading/writing from/to md0 a bit too, if it is mounted.

It even happens that the raid is considered broken because there was a harmless disk timeout, that happens too. An example from my raid set:

Code:

# cat /proc/mdstat
Personalities : [raid1] 
..
md0 : active raid1 sda1[0] sdb1[1]
      96256 blocks [2/2] [UU]
      
unused devices: <none>

Now I set sdb1 to forced faulty

Code:

# mdadm --manage --fail /dev/md0 /dev/sdb1
mdadm: set /dev/sdb1 faulty in /dev/md0
# cat /proc/mdstat 
Personalities : [raid1] 
..
md0 : active raid1 sda1[0] sdb1[2](F)
      96256 blocks [2/1] [U_]
      
unused devices: <none>

This would be a typical situation after hardware failure, or sometimes a more harmless problem such as a timeout. In your case it isn't marked faulty, more like after

Code:

# mdadm --manage --remove /dev/md0 /dev/sdb1
mdadm: hot removed /dev/sdb1
# cat /proc/mdstat 
Personalities : [raid1] 
..    
md0 : active raid1 sda1[0]
      96256 blocks [2/1] [U_]
      
unused devices: <none>

So what you could try is

Code:

# mdadm --manage --add /dev/md0 /dev/sdb1
mdadm: re-added /dev/sdb1
# cat /proc/mdstat 
Personalities : [raid1] 
..      
md0 : active raid1 sdb1[2] sda1[0]
      96256 blocks [2/1] [U_]
      [===========>.........]  recovery = 58.9% (57920/96256) finish=0.0min speed=28960K/sec
      
unused devices: <none>

It should do a resync, which can take quite long if you have a large partition, even longer when the machine is also very busy. If this doesn't work, or if in a couple of days your RAID set fails again, you really do need to replace the disk. A kind soul wrote up a howto for that at http://www.howtoforge.com/replacing_..._a_raid1_array

-Bert