* Two "maybe" related problems
@ 2004-05-18 14:57 Michael Buesch
2004-05-18 15:09 ` Vladimir Saveliev
0 siblings, 1 reply; 4+ messages in thread
From: Michael Buesch @ 2004-05-18 14:57 UTC (permalink / raw)
To: reiserfs-list
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi,
I had a reiser4 partition.
While it was mounted, I deleted the partition with
cfdisk (yes, blame on me ;)
Then I rebooted the machine. I got the following oops
on booting (That's my first problem):
May 18 17:11:07 lfs kernel: Resume Machine: disabled
May 18 17:11:07 lfs kernel: found reiserfs format "3.6" with standard journal
May 18 17:11:07 lfs kernel: Reiserfs journal params: device hda6, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
May 18 17:11:07 lfs kernel: reiserfs: checking transaction log (hda6) for (hda6)
May 18 17:11:07 lfs kernel: Using r5 hash to sort names
May 18 17:11:07 lfs kernel: VFS: Mounted root (reiserfs filesystem) readonly.
May 18 17:11:07 lfs kernel: Freeing unused kernel memory: 152k freed
May 18 17:11:07 lfs kernel: Unable to find swap-space signature
May 18 17:11:07 lfs kernel: EXT2-fs warning: mounting unchecked fs, running e2fsck is recommended
May 18 17:11:07 lfs kernel: attempt to access beyond end of device
May 18 17:11:07 lfs kernel: hda13: rw=0, want=6023432, limit=5859441
(hda13 was the reiser4 partition, before I deleted it.
Here my bootscript tries to boot the non-existing partition)
May 18 17:11:07 lfs kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000
May 18 17:11:07 lfs kernel: printing eip:
May 18 17:11:07 lfs kernel: c01a86fe
May 18 17:11:07 lfs kernel: *pde = 00000000
May 18 17:11:07 lfs kernel: Oops: 0000 [#1]
May 18 17:11:07 lfs kernel: PREEMPT
May 18 17:11:07 lfs kernel: CPU: 0
May 18 17:11:07 lfs kernel: EIP: 0060:[<c01a86fe>] Not tainted
May 18 17:11:07 lfs kernel: EFLAGS: 00010206 (2.6.5)
May 18 17:11:07 lfs kernel: EIP is at znodes_tree_done+0x21/0x1eb
May 18 17:11:07 lfs kernel: eax: 00000000 ebx: dfd13014 ecx: 00000000 edx: dfd13014
May 18 17:11:07 lfs kernel: esi: 00002000 edi: 00000000 ebp: dfd13038 esp: df923d58
May 18 17:11:07 lfs kernel: ds: 007b es: 007b ss: 0068
May 18 17:11:07 lfs kernel: Process mount (pid: 37, threadinfo=df922000 task=df9472e0)
May 18 17:11:07 lfs kernel: Stack: 00000000 dfd13014 dfd16200 df923d9c 00000000 c01b058b dfd13014 dfd13014
May 18 17:11:07 lfs kernel: c01cb25f dfd13014 0000000c fffffffb c01cb537 dfd16200 df923d9c df89d000
May 18 17:11:07 lfs kernel: 00000000 4b1b5d0b 00000000 00000001 df923da8 df923da8 df923db0 df923db0
May 18 17:11:07 lfs kernel: Call Trace:
May 18 17:11:07 lfs kernel: [<c01b058b>] done_tree+0x3e/0x67
May 18 17:11:07 lfs kernel: [<c01cb25f>] _done_formatted_fake+0x25/0x3e
May 18 17:11:07 lfs kernel: [<c01cb537>] reiser4_fill_super+0x5e/0x7d
May 18 17:11:07 lfs kernel: [<c0153913>] sb_set_blocksize+0x1f/0x45
May 18 17:11:07 lfs kernel: [<c0153325>] get_sb_bdev+0xef/0x145
May 18 17:11:07 lfs kernel: [<c01674ff>] alloc_vfsmnt+0x93/0xce
May 18 17:11:07 lfs kernel: [<c01c56a4>] reiser4_get_sb+0x2f/0x33
May 18 17:11:07 lfs kernel: [<c01cb4d9>] reiser4_fill_super+0x0/0x7d
May 18 17:11:07 lfs kernel: [<c015357b>] do_kern_mount+0x5b/0xe1
May 18 17:11:07 lfs kernel: [<c01688ba>] do_mount+0x570/0x765
May 18 17:11:07 lfs kernel: [<c0192608>] remove_save_link+0xbe/0xe8
May 18 17:11:07 lfs kernel: [<c015aaa3>] __user_walk+0x5c/0x72
May 18 17:11:07 lfs kernel: [<c0136a27>] __alloc_pages+0xb9/0x364
May 18 17:11:07 lfs kernel: [<c01682d1>] copy_mount_options+0x7d/0xf6
May 18 17:11:07 lfs kernel: [<c0168e64>] sys_mount+0xc2/0x137
May 18 17:11:07 lfs kernel: [<c01063dd>] sysenter_past_esp+0x52/0x71
May 18 17:11:07 lfs kernel:
May 18 17:11:07 lfs kernel: Code: 8b 3c 81 85 ff 75 07 83 c0 01 39 c6 77 f2 31 c0 85 ff 89 fa
Now, I ignored it and continued booting.
I wanted to create a new reiserfs-3.6 filesystem on the already
exiting hda8. I fired up mkreiserfs:
root@lfs:~> mkreiserfs /dev/hda8
mkreiserfs 3.6.11 (2003 www.namesys.com)
A pair of credits:
Joshua Macdonald wrote the first draft of the transaction manager. Yuri Rupasov
did testing and benchmarking, plus he invented the r5 hash (also used by the
dcache code). Yura Rupasov, Anatoly Pinchuk, Igor Krasheninnikov, Grigory
Zaigralin, Mikhail Gilula, Igor Zagorovsky, Roman Pozlevich, Konstantin
Shvachko, and Joshua MacDonald are former contributors to the project.
Oleg Drokin was the debugger for V3 during most of the time that V4 was under
development, and was quite skilled and fast at it. He wrote the large write
optimization of V3.
Guessing about desired format.. Kernel 2.6.5 is running.
Format 3.6 with standard journal
Count of blocks on the device: 732430
Number of blocks consumed by mkreiserfs formatting process: 8234
Blocksize: 4096
Hash function used to sort names: "r5"
Journal Size 8193 blocks (first block 18)
Journal Max transaction length 1024
inode generation number: 0
UUID: dd3f1273-77a2-4a25-9289-384ba5a723e8
ATTENTION: YOU SHOULD REBOOT AFTER FDISK!
ALL DATA WILL BE LOST ON '/dev/hda8'!
Continue (y/n):y
Initializing journal - 0%....20%....40%....60%....80%....100%
Syncing..
mkreiserfs is stuck in unkillable D state at this point.
(That's my second problem).
Kernel is 2.6.5 with reiser4 snapshot 2004.03.26
Please CC: me, as I'm not subscribed to this list.
- --
Regards Michael Buesch [ http://www.tuxsoft.de.vu ]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
iD8DBQFAqiRqFGK1OIvVOP4RAnDRAKDTBq/AhGQe23gnXEfp4h2eDYgi7wCgk4An
ZT5P3f2CyF/mLiSPbrxt7Gs=
=tUh1
-----END PGP SIGNATURE-----
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Two "maybe" related problems
2004-05-18 14:57 Two "maybe" related problems Michael Buesch
@ 2004-05-18 15:09 ` Vladimir Saveliev
2004-05-19 11:57 ` Michael Buesch
0 siblings, 1 reply; 4+ messages in thread
From: Vladimir Saveliev @ 2004-05-18 15:09 UTC (permalink / raw)
To: Michael Buesch; +Cc: reiserfs-list
Hello
On Tue, 2004-05-18 at 18:57, Michael Buesch wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi,
>
> I had a reiser4 partition.
> While it was mounted, I deleted the partition with
> cfdisk (yes, blame on me ;)
> Then I rebooted the machine. I got the following oops
> on booting (That's my first problem):
>
> May 18 17:11:07 lfs kernel: Resume Machine: disabled
> May 18 17:11:07 lfs kernel: found reiserfs format "3.6" with standard journal
> May 18 17:11:07 lfs kernel: Reiserfs journal params: device hda6, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
> May 18 17:11:07 lfs kernel: reiserfs: checking transaction log (hda6) for (hda6)
> May 18 17:11:07 lfs kernel: Using r5 hash to sort names
> May 18 17:11:07 lfs kernel: VFS: Mounted root (reiserfs filesystem) readonly.
> May 18 17:11:07 lfs kernel: Freeing unused kernel memory: 152k freed
> May 18 17:11:07 lfs kernel: Unable to find swap-space signature
> May 18 17:11:07 lfs kernel: EXT2-fs warning: mounting unchecked fs, running e2fsck is recommended
> May 18 17:11:07 lfs kernel: attempt to access beyond end of device
> May 18 17:11:07 lfs kernel: hda13: rw=0, want=6023432, limit=5859441
> (hda13 was the reiser4 partition, before I deleted it.
> Here my bootscript tries to boot the non-existing partition)
> May 18 17:11:07 lfs kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000
> May 18 17:11:07 lfs kernel: printing eip:
> May 18 17:11:07 lfs kernel: c01a86fe
> May 18 17:11:07 lfs kernel: *pde = 00000000
> May 18 17:11:07 lfs kernel: Oops: 0000 [#1]
> May 18 17:11:07 lfs kernel: PREEMPT
> May 18 17:11:07 lfs kernel: CPU: 0
> May 18 17:11:07 lfs kernel: EIP: 0060:[<c01a86fe>] Not tainted
> May 18 17:11:07 lfs kernel: EFLAGS: 00010206 (2.6.5)
> May 18 17:11:07 lfs kernel: EIP is at znodes_tree_done+0x21/0x1eb
> May 18 17:11:07 lfs kernel: eax: 00000000 ebx: dfd13014 ecx: 00000000 edx: dfd13014
> May 18 17:11:07 lfs kernel: esi: 00002000 edi: 00000000 ebp: dfd13038 esp: df923d58
> May 18 17:11:07 lfs kernel: ds: 007b es: 007b ss: 0068
> May 18 17:11:07 lfs kernel: Process mount (pid: 37, threadinfo=df922000 task=df9472e0)
> May 18 17:11:07 lfs kernel: Stack: 00000000 dfd13014 dfd16200 df923d9c 00000000 c01b058b dfd13014 dfd13014
> May 18 17:11:07 lfs kernel: c01cb25f dfd13014 0000000c fffffffb c01cb537 dfd16200 df923d9c df89d000
> May 18 17:11:07 lfs kernel: 00000000 4b1b5d0b 00000000 00000001 df923da8 df923da8 df923db0 df923db0
> May 18 17:11:07 lfs kernel: Call Trace:
> May 18 17:11:07 lfs kernel: [<c01b058b>] done_tree+0x3e/0x67
> May 18 17:11:07 lfs kernel: [<c01cb25f>] _done_formatted_fake+0x25/0x3e
> May 18 17:11:07 lfs kernel: [<c01cb537>] reiser4_fill_super+0x5e/0x7d
> May 18 17:11:07 lfs kernel: [<c0153913>] sb_set_blocksize+0x1f/0x45
> May 18 17:11:07 lfs kernel: [<c0153325>] get_sb_bdev+0xef/0x145
> May 18 17:11:07 lfs kernel: [<c01674ff>] alloc_vfsmnt+0x93/0xce
> May 18 17:11:07 lfs kernel: [<c01c56a4>] reiser4_get_sb+0x2f/0x33
> May 18 17:11:07 lfs kernel: [<c01cb4d9>] reiser4_fill_super+0x0/0x7d
> May 18 17:11:07 lfs kernel: [<c015357b>] do_kern_mount+0x5b/0xe1
> May 18 17:11:07 lfs kernel: [<c01688ba>] do_mount+0x570/0x765
> May 18 17:11:07 lfs kernel: [<c0192608>] remove_save_link+0xbe/0xe8
> May 18 17:11:07 lfs kernel: [<c015aaa3>] __user_walk+0x5c/0x72
> May 18 17:11:07 lfs kernel: [<c0136a27>] __alloc_pages+0xb9/0x364
> May 18 17:11:07 lfs kernel: [<c01682d1>] copy_mount_options+0x7d/0xf6
> May 18 17:11:07 lfs kernel: [<c0168e64>] sys_mount+0xc2/0x137
> May 18 17:11:07 lfs kernel: [<c01063dd>] sysenter_past_esp+0x52/0x71
> May 18 17:11:07 lfs kernel:
> May 18 17:11:07 lfs kernel: Code: 8b 3c 81 85 ff 75 07 83 c0 01 39 c6 77 f2 31 c0 85 ff 89 fa
>
>
> Now, I ignored it and continued booting.
> I wanted to create a new reiserfs-3.6 filesystem on the already
> exiting hda8. I fired up mkreiserfs:
>
> root@lfs:~> mkreiserfs /dev/hda8
> mkreiserfs 3.6.11 (2003 www.namesys.com)
>
> A pair of credits:
> Joshua Macdonald wrote the first draft of the transaction manager. Yuri Rupasov
> did testing and benchmarking, plus he invented the r5 hash (also used by the
> dcache code). Yura Rupasov, Anatoly Pinchuk, Igor Krasheninnikov, Grigory
> Zaigralin, Mikhail Gilula, Igor Zagorovsky, Roman Pozlevich, Konstantin
> Shvachko, and Joshua MacDonald are former contributors to the project.
>
> Oleg Drokin was the debugger for V3 during most of the time that V4 was under
> development, and was quite skilled and fast at it. He wrote the large write
> optimization of V3.
>
> Guessing about desired format.. Kernel 2.6.5 is running.
> Format 3.6 with standard journal
> Count of blocks on the device: 732430
> Number of blocks consumed by mkreiserfs formatting process: 8234
> Blocksize: 4096
> Hash function used to sort names: "r5"
> Journal Size 8193 blocks (first block 18)
> Journal Max transaction length 1024
> inode generation number: 0
> UUID: dd3f1273-77a2-4a25-9289-384ba5a723e8
> ATTENTION: YOU SHOULD REBOOT AFTER FDISK!
> ALL DATA WILL BE LOST ON '/dev/hda8'!
> Continue (y/n):y
> Initializing journal - 0%....20%....40%....60%....80%....100%
> Syncing..
>
>
> mkreiserfs is stuck in unkillable D state at this point.
> (That's my second problem).
>
This is probably caused by the fact that system oopsed already.
I would suggest you to boot of rescue system and change /etc/fstab so
that /dev/hda13 does not get mounted on booting (noauto in fourth field
of a /etc/fstab's record)
Do so for other filesystems underlying deveices of which were
repartitioned.
reboot in usual way
if it does not oops on boot (it should not) - mkreiserfs will not get
stuck
>
> Kernel is 2.6.5 with reiser4 snapshot 2004.03.26
> Please CC: me, as I'm not subscribed to this list.
>
> - --
> Regards Michael Buesch [ http://www.tuxsoft.de.vu ]
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.4 (GNU/Linux)
>
> iD8DBQFAqiRqFGK1OIvVOP4RAnDRAKDTBq/AhGQe23gnXEfp4h2eDYgi7wCgk4An
> ZT5P3f2CyF/mLiSPbrxt7Gs=
> =tUh1
> -----END PGP SIGNATURE-----
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Two "maybe" related problems
2004-05-18 15:09 ` Vladimir Saveliev
@ 2004-05-19 11:57 ` Michael Buesch
2004-05-19 12:09 ` Vladimir Saveliev
0 siblings, 1 reply; 4+ messages in thread
From: Michael Buesch @ 2004-05-19 11:57 UTC (permalink / raw)
To: Vladimir Saveliev; +Cc: reiserfs-list
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Tuesday 18 May 2004 17:09, you wrote:
> This is probably caused by the fact that system oopsed already.
> I would suggest you to boot of rescue system and change /etc/fstab so
> that /dev/hda13 does not get mounted on booting (noauto in fourth field
> of a /etc/fstab's record)
> Do so for other filesystems underlying deveices of which were
> repartitioned.
> reboot in usual way
> if it does not oops on boot (it should not) - mkreiserfs will not get
> stuck
I rebooted and updated reiserfs tools. Now it works as expected.
But should the kernel (reiser4 driver) be "allowed" to oops,
if someone tries to mount a non-existing partition, just because
there's some old and corrupt data left, from a previous filesystem?
Shouldn't it simply refuse to mount?
- --
Regards Michael Buesch [ http://www.tuxsoft.de.vu ]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
iD8DBQFAq0uvFGK1OIvVOP4RAn2YAJ9+AYo29gLPT8jZjFdjV7pLka88cACg0vHQ
tvvzO2HGeT/x73FrTbaBJ5I=
=Yg1o
-----END PGP SIGNATURE-----
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Two "maybe" related problems
2004-05-19 11:57 ` Michael Buesch
@ 2004-05-19 12:09 ` Vladimir Saveliev
0 siblings, 0 replies; 4+ messages in thread
From: Vladimir Saveliev @ 2004-05-19 12:09 UTC (permalink / raw)
To: Michael Buesch; +Cc: reiserfs-list
Hello
On Wed, 2004-05-19 at 15:57, Michael Buesch wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Tuesday 18 May 2004 17:09, you wrote:
> > This is probably caused by the fact that system oopsed already.
> > I would suggest you to boot of rescue system and change /etc/fstab so
> > that /dev/hda13 does not get mounted on booting (noauto in fourth field
> > of a /etc/fstab's record)
> > Do so for other filesystems underlying deveices of which were
> > repartitioned.
> > reboot in usual way
> > if it does not oops on boot (it should not) - mkreiserfs will not get
> > stuck
>
> I rebooted and updated reiserfs tools. Now it works as expected.
>
> But should the kernel (reiser4 driver) be "allowed" to oops,
No.
> if someone tries to mount a non-existing partition, just because
> there's some old and corrupt data left, from a previous filesystem?
> Shouldn't it simply refuse to mount?
Yes
we shall fix it
>
> - --
> Regards Michael Buesch [ http://www.tuxsoft.de.vu ]
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.4 (GNU/Linux)
>
> iD8DBQFAq0uvFGK1OIvVOP4RAn2YAJ9+AYo29gLPT8jZjFdjV7pLka88cACg0vHQ
> tvvzO2HGeT/x73FrTbaBJ5I=
> =Yg1o
> -----END PGP SIGNATURE-----
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2004-05-19 12:09 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-05-18 14:57 Two "maybe" related problems Michael Buesch
2004-05-18 15:09 ` Vladimir Saveliev
2004-05-19 11:57 ` Michael Buesch
2004-05-19 12:09 ` Vladimir Saveliev
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.