[linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[Rasmus Andersen]

Hi.

I have built a 2.2.20-2 kernel with kiobuf-2.2.18pre24-B.tar.gz
applied and patched with the patch lvm_0.9.1_beta7 created.

I have begun to get the following in my /var/log/messages in
conjunction with various programs (tar, cp) failing:

Jun  4 21:55:05 firewall kernel: attempt to access beyond end of device 
Jun  4 21:55:05 firewall kernel: 08:01: rw=0, want=530424, limit=410608 
Jun  4 21:55:05 firewall kernel: dev 3a:03 blksize=1024 blocknr=622593 sector=1060840 size=4096 count=1 
Jun  4 21:55:05 firewall kernel: EXT2-fs error (device lvm(58,3)): read_inode_bitmap: Cannot read inode bitmap - block_group = 19, inode_bitmap = 622593 
Jun  4 21:55:05 firewall kernel: EXT2-fs error (device lvm(58,3)): ext2_check_inodes_bitmap: Wrong free inodes count in super block, stored = 307461, counted = 291461 


I can do a 'dd if=/dev/<diskX> of=/dev/null' for the disks that are
part of the volume group without problems but if I do it for /dev/vg1/
lvol4, I get the error above (without the EXT2-part).

I am now trying to get whatever data I can off the FS and would like
1) helpful suggestions, and 2) to provide whatever feedback I can
before I have to reclaim the disk space. Probably for something
non-LVM for the moment.

Please mail me for additional information. I am rather new to LVM
and have little clue as to what information you might find useful.
-- 
        Rasmus(rasmus@jaquet.dk)

Gates' Law: Every 18 months, the speed of software halves
  -- Anonymous

Re: [linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[Heinz J. Mauelshagen]

Hi Rasmus.

On Tue, Jun 05, 2001 at 09:24:36AM +0200, Rasmus Andersen wrote:
> Hi.
> 
> I have built a 2.2.20-2 kernel with kiobuf-2.2.18pre24-B.tar.gz
> applied and patched with the patch lvm_0.9.1_beta7 created.
> 
> I have begun to get the following in my /var/log/messages in
> conjunction with various programs (tar, cp) failing:
> 
> Jun  4 21:55:05 firewall kernel: attempt to access beyond end of device 
> Jun  4 21:55:05 firewall kernel: 08:01: rw=0, want=530424, limit=410608 
> Jun  4 21:55:05 firewall kernel: dev 3a:03 blksize=1024 blocknr=622593 sector=1060840 size=4096 count=1 
> Jun  4 21:55:05 firewall kernel: EXT2-fs error (device lvm(58,3)): read_inode_bitmap: Cannot read inode bitmap - block_group = 19, inode_bitmap = 622593 
> Jun  4 21:55:05 firewall kernel: EXT2-fs error (device lvm(58,3)): ext2_check_inodes_bitmap: Wrong free inodes count in super block, stored = 307461, counted = 291461 
> 

This sounds like a well known ext2 problem.
Did you run fsck successfully? And if so, didn't it work again?


> 
> I can do a 'dd if=/dev/<diskX> of=/dev/null' for the disks that are
> part of the volume group without problems but if I do it for /dev/vg1/
> lvol4, I get the error above (without the EXT2-part).

Did you *really* run "dd if=/dev/vg1/lvol4 of=/dev/null" and got that error
in the log? Never have seen that before.

> 
> I am now trying to get whatever data I can off the FS and would like
> 1) helpful suggestions, and 2) to provide whatever feedback I can
> before I have to reclaim the disk space. Probably for something
> non-LVM for the moment.
> 
> Please mail me for additional information. I am rather new to LVM
> and have little clue as to what information you might find useful.
> -- 
>         Rasmus(rasmus@jaquet.dk)
> 
> Gates' Law: Every 18 months, the speed of software halves
>   -- Anonymous
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com
> http://lists.sistina.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html

-- 

Regards,
Heinz    -- The LVM Guy --

*** Software bugs are stupid.
    Nevertheless it needs not so stupid people to solve them ***

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Sistina Software Inc.
Senior Consultant/Developer                       Am Sonnenhang 11
                                                  56242 Marienrachdorf
                                                  Germany
Mauelshagen@Sistina.com                           +49 2626 141200
                                                       FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Re: [linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[Rasmus Andersen]

On Tue, Jun 05, 2001 at 12:34:17PM +0200, Heinz J. Mauelshagen wrote:
> 
> This sounds like a well known ext2 problem.
> Did you run fsck successfully? And if so, didn't it work again?

Uhm. Known problem? Could you elaborate a bit? I have not seen it
before and would like to know if it is something I will encounter
regularily using ext2 and LVM.

I have run e2fsck a number of times now, and while I do not get
the kernel log message after the reboot (se below) fsck complain
and fix stuff every time in lost+found which contain a lot of
junk I cannot delete manually either.

> 
> 
> > 
> > I can do a 'dd if=/dev/<diskX> of=/dev/null' for the disks that are
> > part of the volume group without problems but if I do it for /dev/vg1/
> > lvol4, I get the error above (without the EXT2-part).
> 
> Did you *really* run "dd if=/dev/vg1/lvol4 of=/dev/null" and got that error
> in the log? Never have seen that before.

I will try to reproduce. But trying to get as much as the (used) 
filesystem with off the disk with tar resulted in a kernel panic
(this was just after my last mail):

panic: brw_page: page not locked for IO

This was put after a couple of lines of 

Jun  5 10:04:35 firewall kernel: attempt to access beyond end of device 
Jun  5 10:04:35 firewall kernel: 08:01: rw=0, want=530420, limit=410608 
Jun  5 10:04:35 firewall kernel: dev 3a:03 blksize=1024 blocknr=622592 sector=1060832 size=4096 count=1 

on the console.

-- 
Regards,
        Rasmus(rasmus@jaquet.dk)

"I'm not going to have some reporters pawing through our papers. We are
the president."  -Hillary Clinton commenting on the release of subpoenaed
documents.

Re: [linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[Rasmus Andersen]

On Tue, Jun 05, 2001 at 03:46:57PM +0200, Rasmus Andersen wrote:
> On Tue, Jun 05, 2001 at 12:34:17PM +0200, Heinz J. Mauelshagen wrote:
> > 
> > This sounds like a well known ext2 problem.
> > Did you run fsck successfully? And if so, didn't it work again?
> 

Forgot to state that before the panic->reboot fsck could not complete.
It ran up against a short read with the kernel log entry as reported.
This was (of course) on /dev/vg1/lvol4, not mounted.
-- 
        Rasmus(rasmus@jaquet.dk)

It's a recession when your neighbour loses his job; it's a depression 
when you lose yours. -- Harry S. Truman 

Re: [linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[Rasmus Andersen]

On Tue, Jun 05, 2001 at 03:46:57PM +0200, Rasmus Andersen wrote:
[...]
> > Did you *really* run "dd if=/dev/vg1/lvol4 of=/dev/null" and got that error
> > in the log? Never have seen that before.
> 
> I will try to reproduce. But trying to get as much as the (used) 
[...]

To finish this off: I cannot reproduce it after my reboot.
-- 
        Rasmus(rasmus@jaquet.dk)

We're going to turn this team around 360 degrees.
-Jason Kidd, upon his drafting to the Dallas Mavericks

Re: [linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[Heinz J. Mauelshagen]

On Tue, Jun 05, 2001 at 09:02:48PM +0200, Rasmus Andersen wrote:
> On Tue, Jun 05, 2001 at 03:46:57PM +0200, Rasmus Andersen wrote:
> [...]
> > > Did you *really* run "dd if=/dev/vg1/lvol4 of=/dev/null" and got that error
> > > in the log? Never have seen that before.
> > 
> > I will try to reproduce. But trying to get as much as the (used) 
> [...]
> 
> To finish this off: I cannot reproduce it after my reboot.

Oops. Bad luck then.
In case you are able again, please provide the configuration information and
log here.

> -- 
>         Rasmus(rasmus@jaquet.dk)
> 
> We're going to turn this team around 360 degrees.
> -Jason Kidd, upon his drafting to the Dallas Mavericks
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com
> http://lists.sistina.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html

-- 

Regards,
Heinz    -- The LVM Guy --

*** Software bugs are stupid.
    Nevertheless it needs not so stupid people to solve them ***

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Sistina Software Inc.
Senior Consultant/Developer                       Am Sonnenhang 11
                                                  56242 Marienrachdorf
                                                  Germany
Mauelshagen@Sistina.com                           +49 2626 141200
                                                       FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Re: [linux-lvm] Problems with 0.9.1b7 and 2.2.20-2

[Heinz J. Mauelshagen]

On Tue, Jun 05, 2001 at 03:46:57PM +0200, Rasmus Andersen wrote:
> On Tue, Jun 05, 2001 at 12:34:17PM +0200, Heinz J. Mauelshagen wrote:
> > 
> > This sounds like a well known ext2 problem.
> > Did you run fsck successfully? And if so, didn't it work again?
> 
> Uhm. Known problem? Could you elaborate a bit?

I was refering to ext2 consistency problems in older kernels which have been
fixed in the meantime. You need to have recent ext2 tool as well.

> I have not seen it
> before and would like to know if it is something I will encounter
> regularily using ext2 and LVM.

No, you shouldn't in case you have recent ext2 software.

> 
> I have run e2fsck a number of times now, and while I do not get
> the kernel log message after the reboot (se below) fsck complain
> and fix stuff every time in lost+found which contain a lot of
> junk I cannot delete manually either.
> 
> > 
> > 
> > > 
> > > I can do a 'dd if=/dev/<diskX> of=/dev/null' for the disks that are
> > > part of the volume group without problems but if I do it for /dev/vg1/
> > > lvol4, I get the error above (without the EXT2-part).
> > 
> > Did you *really* run "dd if=/dev/vg1/lvol4 of=/dev/null" and got that error
> > in the log? Never have seen that before.
> 
> I will try to reproduce. But trying to get as much as the (used) 
> filesystem with off the disk with tar resulted in a kernel panic
> (this was just after my last mail):
> 
> panic: brw_page: page not locked for IO
> 
> This was put after a couple of lines of 
> 
> Jun  5 10:04:35 firewall kernel: attempt to access beyond end of device 
> Jun  5 10:04:35 firewall kernel: 08:01: rw=0, want=530420, limit=410608 
> Jun  5 10:04:35 firewall kernel: dev 3a:03 blksize=1024 blocknr=622592 sector=1060832 size=4096 count=1 
> 
> on the console.
> 
> -- 
> Regards,
>         Rasmus(rasmus@jaquet.dk)
> 
> "I'm not going to have some reporters pawing through our papers. We are
> the president."  -Hillary Clinton commenting on the release of subpoenaed
> documents.
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com
> http://lists.sistina.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html

-- 

Regards,
Heinz    -- The LVM Guy --

*** Software bugs are stupid.
    Nevertheless it needs not so stupid people to solve them ***

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Sistina Software Inc.
Senior Consultant/Developer                       Am Sonnenhang 11
                                                  56242 Marienrachdorf
                                                  Germany
Mauelshagen@Sistina.com                           +49 2626 141200
                                                       FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-