On Fri, Oct 06, 2000 at 11:17:31AM +0100, Corin Hartland-Swann wrote:
> On Fri, 6 Oct 2000, Joel Becker wrote:
> > Please reply personally. I just had my kernel go nuts in
> > do_try_to_free_pages(), and when I bounced the box (sysrq sync + reboot)
>
> What kernel were you using? I assume you're aware of this bug in
> 2.2.15/16... I dropped back to 14 to stop this from happening, but am
> currently testing 2.2.17 (final) to see if that's OK.
I was using 2.2.18pre10 + latest .17 raid patches. My
understanding is that do_try_to_free_pages() thing is a VM issue, not
necessarily a RAID issue.
> Your superblock should /NOT/ have been corrupted. Can you post more
> information about this?
It is beginning to look like a bad disk, but I cannot be sure.
Specifically, I/O errors to 0:1:0 (sdb), which is the first disk in the
array. I recovered the array by removing sdb1 from the raidtab and
starting up on sdc1 (the mirror). I then synced sdb1 back in. It
failed again later in the evening, with more I/O errors. This could be
bus, card, or drive, though I'm pretty sure I seated the bus well when I
put the drives in. The second failure had no "do_try_to_free_pages()"
issues.
Basically, it doesn't start the array on reboot, it says
(paraphrased) "bad superblock on sdb1, giving up". I'd like to see it
say "bad superblock on sdb1, lets try sdc1", but it doesn't seem to do
that.
Oh I miss my days with real error reporting.
> IIRC all that should have happened is it should have fsck'd one of the
> disks (the first, I think?), and then resync'd that to the other one.
Is this still valid for the 0.9 RAID stuff? I saw it in the 0.4
documentation online, but when I tried to mount one of the disks, the
kernel positively vomited errors. So if I'm reading you correctly, each
drive of a RAID-1 is a normal partition in every mountable way?
Thanks for the reply,
Joel
--
"We'd better get back, `cause it'll be dark soon,
and they mostly come at night. Mostly."
http://www.jlbec.org/
[EMAIL PROTECTED]
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to [EMAIL PROTECTED]
