Re: Just received a fail event from mdadm (UncorrectableError), is my drive dead?
On Mon, Oct 30, 2006 at 11:47:15AM -0700, Michael Loftis wrote:
> --On October 30, 2006 1:31:02 PM -0500 Mike Garey <random51k@gmail.com>
> wrote:
>
> >I just received an email from mdadm monitoring saying that "A Fail
> >event had been detected on md device /dev/md0." According to dmesg, I
> >see the following:
> <...>
>
> >does this mean my drive is dead and should be replaced? Or is it a bad
> >block that's been remapped to another part of the drive, and I just
> >need to re-add my drive to the array to get it to re-sync? Thanks,
>
> With IDE drives it's about impossible to tell the difference. I'd try via
> mdadm removing the failed part of the mirror, then hotadding it back in.
> let it resync. If it occurs again, especially during resync, the drive is
> going or gone, and it's time to replace it ASAP.
Not to simply be a 'me too' reply, but reinforcement of Mike's
statement. If the drive comes back online, count yourself lucky and get
another drive ordered now while it's an easy task to rebuild.
This error message from md IS YOUR WARNING. Drives very often give us
these warnings but we procrastinate on acting on them. Act now before
the rest of the array fails.
Good luck.
j
Reply to: