[CentOS-virt] Xen box down

Tue Jan 5 18:29:04 UTC 2010
Ben M. <centos at rivint.com>

Christopher G. Stach II wrote:
> ----- "Ben M." <centos at rivint.com> wrote:
>> Error is: "duplicate or bad block in use"
> It's probably just that fsck can't automatically fix some dirtiness and not a big deal. If you aren't prompted for a password or to log in to fix manually, get to the grub menu, edit the grub command line, stick ``single'' and/or ``init=/bin/sh'' on the end, boot, and run fsck manually. If you just want the machine up in a possibly slightly fucked state, just answer "yes" to everything. If not and you care a little bit about maybe getting some data back, see the next paragraph. (It's usually not that bad unless you have a skilled enemy or very bad luck.) You can probably have someone do all of it over the phone.
> If it doesn't even get that far or fsck can't fix it automatically, you're probably screwed. Whenever that has happened to me, I just do a block level dump of the partition/disk and recover from that image. It's a lot easier. Anyway, it is probably fine. If it isn't, you can always try pulling each of the disks or setting it back to use a single disk to try and isolate the problem.
> Also, switch to MD RAID. :)

Have at touch of flu or something, sorry for delays answering. Not a 
good way to approach the issue. Not thinking as clearly as would prefer.

I am thinking of running SpinRite first. It tends to do a good job with 
SMART issues. Anyone see an issue with this?