Re: Unable to fsck 4.5TB FFSv2 fs

[Stephen Borrill <netbsd%precedence.co.uk@localhost>]

On Tue, 23 Sep 2008, Christos Zoulas wrote:
> In article <Pine.NEB.4.64.0809231024100.16305%ugly.precedence.co.uk@localhost>,
> Stephen Borrill  <netbsd%precedence.co.uk@localhost> wrote:
> > I have:
> >
> > ld1 at aac0 unit 1: RAID 5
> > ld1: 4656 GB, 607897 cyl, 255 head, 63 sec, 512 bytes/sect x 9765867520
> > sectors
> >
> > Which I formatted with:
> > newfs -O 2 -f 4096 -b 32768 -I -F -s 9765867520 /dev/rld1d
> >
> > This worked fine, but apparently (the machine is at a customer's) it
> > locked and so they power-cycled it and thus it wants to fsck (NetBSD 4.0
> > so no WAPBL).
> >
> > This is giving:
> > ** /dev/rld1d
> > ** Last Mounted on /usr/backup
> > ** Phase 1 - Check Blocks and Sizes
> > cannot alloc 1243616332 bytes for inoinfo
[snip]
> Can you try this patch? It should reduce the total space needed, so if you die
> in the second alloc, this should probably fix it. I have not tested it, so you
> are on your own if it eats your disk, but visually it looks safe.

My first attempt tried to clear every inode which I think was down to an 
error in hand applying the patch (missed out the memset). So I newfsed and 
started again. The copy eventually died and the machine dropped to ddb:
http://mail-index.netbsd.org/netbsd-users/2008/09/25/msg002023.html

So I then tried the correctly patched fsck_ffs and its demands have gone 
up!

** /dev/rld1d
** Last Mounted on /usr/backup
** Phase 1 - Check Blocks and Sizes
cannot alloc 4294967292 bytes for inoinfo

--
Stephen