fsck out of memory

fsck out of memory

[Marco C. Mason]

Stephan--

I don't know if anyone mentioned it or not, but the block addresses in your
error messages appear suspiciously close to 2^29.  I'm suspecting an 
internal
overflow in a calculation somewhere...

--marco

Re:2TB+ fs ext3 (was fsck out of memory)

[Stephan van Hienen]

On Mon, 10 Feb 2003, Marco C. Mason wrote:
> I don't know if anyone mentioned it or not, but the block addresses in your
> error messages appear suspiciously close to 2^29.  I'm suspecting an
> internal
> overflow in a calculation somewhere...

615381536 blocks can this be over the limit ?

(-b 8192 gives me warning : 'Warning: blocksize 8192 not usable on most
systems.')

Re:2TB+ fs ext3 (was fsck out of memory)

[Stephan van Hienen]

On Tue, 11 Feb 2003, Bryan O'Sullivan wrote:

> Ugh, that stuff is ancient.
>
> Peter Chubb has a backport of his (very thorough) 2.5 patch to support
> large block devices on 32-bit platforms, against much newer kernels
> (e.g. 2.4.20) than Ben's stuff.  His site seems to be down, but you
> should be able to get it from somewhere under
> http://www.gelato.unsw.edu.au/~peterc/
correct link : http://www.gelato.unsw.edu.au/patches-index.html
(but down)

i used this patch, but i got this comment :

> Well, that's the most likely candidate, because it's the least tested
> component.  Are you using Ben LaHaise's LBD fixes for the md devices?
> Without those, md and lvm are not LBD-safe.

makes me wonder if i need another patch besides the 'Peter Chubb patch'
when using md raid ?

> I haven't used Peter's patch, but a similar patch, developed
> independently, definitely allows ext3 filesystems of up to 8TB in size
> to work fine on x86, under 2.4.

look at my other posts, i can't create/work with a 2348 Gigabyte /dev/md0

Re:2TB+ fs ext3 (was fsck out of memory)

[Stephan van Hienen]

On Mon, 10 Feb 2003, Helge Hafting wrote:

> For 1MB average filesize use -i 1048576

tried to mke2fs with '-i 1048576' :

----
]# mke2fs -i 1048576 /dev/md0 -R stride=16 -m 0 -j
mke2fs 1.32 (09-Nov-2002)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
2403840 inodes, 615381536 blocks
0 blocks (0.00%) reserved for the super user
First data block=0
18780 block groups
32768 blocks per group, 32768 fragments per group
128 inodes per group
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632,
2654208,
        4096000, 7962624, 11239424, 20480000, 23887872, 71663616,
78675968,
        102400000, 214990848, 512000000, 550731776

Writing inode tables: done
Creating journal (8192 blocks): done
Writing superblocks and filesystem accounting information: done

This filesystem will be automatically checked every 23 mounts or
180 days, whichever comes first.  Use tune2fs -c or -i to override.
----
dmesg output :

md: mke2fs(pid 10600) used obsolete MD ioctl, upgrade your software to use
new ioctls.
----
then i try to mount :

----
]# mount /dev/md0  /raid/
mount: wrong fs type, bad option, bad superblock on /dev/md0,
       or too many mounted file systems
---
dmesg output :

raid5: switching cache buffer size, 4096 --> 1024
raid5: switching cache buffer size, 1024 --> 4096
EXT3-fs error (device md(9,0)): ext3_check_descriptors: Block bitmap for
group 4480 not in group (block 15)!
EXT3-fs: group descriptors corrupted !
----

and e2fsck :

----
e2fsck /dev/md0
e2fsck 1.32 (09-Nov-2002)
Group descriptors look bad... trying backup blocks...
Block bitmap for group 6528 is not in group.  (block 15)
Relocate<y>? yes

Inode bitmap for group 6528 is not in group.  (block 3145728)
Relocate<y>? yes

Inode table for group 6528 is not in group.  (block 0)
WARNING: SEVERE DATA LOSS POSSIBLE.
Relocate<y>? yes

Block bitmap for group 6529 is not in group.  (block 0)
Relocate<y>? yes

Inode bitmap for group 6529 is not in group.  (block 0)
Relocate<y>? yes

Inode table for group 6529 is not in group.  (block 0)
WARNING: SEVERE DATA LOSS POSSIBLE.
Relocate<y>? yes

Block bitmap for group 6530 is not in group.  (block 0)
Relocate<y>? yes

Inode bitmap for group 6530 is not in group.  (block 0)
Relocate<y>? yes

Inode table for group 6530 is not in group.  (block 0)
WARNING: SEVERE DATA LOSS POSSIBLE.
Relocate<y>? yes

.....