Re: Paranoia about DegradedArray
On Wednesday 29 October 2008, Hendrik Boom wrote:
> I got the message (via email)
>
> This is an automatically generated mail message from mdadm
> running on april
>
> A DegradedArray event had been detected on md device /dev/md0.
>
> Faithfully yours, etc.
>
> P.S. The /proc/mdstat file currently contains the following:
>
> Personalities : [raid1]
> md0 : active raid1 hda3[0]
> 242219968 blocks [2/1] [U_]
>
> unused devices: <none>
>
You don't mention that you've checked the array with
mdadm --detail /dev/md0. Try that and it will give you some good
information.
I've never used /proc/mdstat because the --detail option gives me more
data in one shot. From what I remember, this is a raid1, right? It
looks like it has 2 devices and one is still working, but I might be
wrong. Again --detail will spell out a lot of this explicitly.
> Now I gather from what I've googled that somehow I've got to get the
> RAID to reestablish the failed drive by copying from the nonfailed
> drive. I do believe the hardware is basically OK, and that what I've
> got is probably a problem due to a power failure (We've had a lot of
> these recently) or something transient.
>
> (a) How do I do this?
If a drive has actually failed, then mdadm --remove /dev/md0 /dev/hdxx.
If the drive has not failed, then you need to fail it first with --fail
as an option/switch for mdadm.
> (b) is hda3 the failed drive, or is it the one that's still working?
That's one of the things mdadm --detail /dev/md0 will tell you. It will
list the active drives and the failed drives.
Hal
Reply to: