On Tuesday 11 January 2005 23:33, Simon Raffeiner wrote:
> Hi,
>
> yesterday my workstation refused to boot Linux using a Reiser4 formatted /
> partition. It mocks about an error with an already cached block and about
> skipping the "entire subtree", I also can remember the filename "znode.c".
> ATM I had to boot another PC and connect the harddisk there to make some
> diagnostics so I cannot post the original error message, but this can be
> done on request.
>
> I used to access the partition with Kernel 2.6.9-mm1 for about a month and
> switched to 2.6.10-mm1 three days ago with no problems. According to
> dd_rescue and dmesg there are no bad blocks on the disk (at least the drive
> does not give any error messages about).
>
> I tried to run fsck.reiser4 from the X-2004 Version of "Kanotix" and got a
> segmentation fault right after "CHECKING STORAGE TREE". I tried the same
> with reiser4progs 1.0.1 on Gentoo with the same result, also the version
> from 1.0.3 breaks. This is the output from fsck.reiser4 1.0.3 on Gentoo
> Linux:
>
>
> playstation etc # fsck.reiser4 /dev/hdc3
> *******************************************************************
> This is an EXPERIMENTAL version of fsck.reiser4. Read README first.
> *******************************************************************
>
> Fscking the /dev/hdc3 block device.
> Will check the consistency of the Reiser4 SuperBlock.
> Will check the consistency of the Reiser4 FileSystem.
> Continue?
> (Yes/No): Yes
> ***** fsck.reiser4 started at Tue Jan 11 21:27:03 2005
> Fatal: Failed to open the reiser4 backup.
> Fatal: Cannot open the FileSystem on (/dev/hdc3).
>
> 1 fatal corruptions were detected in SuperBlock. Run with --build-sb option
> to fix them.
>
>
> playstation etc # fsck.reiser4 --build-sb /dev/hdc3 -L fsck.log
> *******************************************************************
> This is an EXPERIMENTAL version of fsck.reiser4. Read README first.
> *******************************************************************
>
> Fscking the /dev/hdc3 block device.
> Will build the Reiser4 SuperBlock.
> Will check the consistency of the Reiser4 FileSystem.
> Continue?
> (Yes/No): Yes
> ***** fsck.reiser4 started at Tue Jan 11 21:24:10 2005
> Enter the key plugin name [key_large]:
> Reiser4 fs was detected on /dev/hdc3.
> Master super block (16):
> magic: ReIsEr4
> blksize: 4096
> format: 0x0 (format40)
> uuid: 97920906-596c-4a63-9d49-2a268254f6c6
> label: <none>
>
> Format super block (17):
> plugin: format40
> description: Disk-format for reiser4.
> magic: ReIsEr40FoRmAt
> flushes: 0
> mkfs id: 0x751d8925
> blocks: 14462532
> free blocks: 10761350
> root block: 119159
> tail policy: 0x2 (smart)
> next oid: 0x585b59
> file count: 555932
> tree height: 5
> key policy: LARGE
>
>
> CHECKING STORAGE TREE
> Segmentation fault
>
>
> The harddisk and the file system label seem to be perfectly okay but there
> is some kind of corruption that breaks the Reiser4 code in Linux 2.6.9-mm1,
> Linux 2.6.10-mm1, reiser4progs 1.0.1 and reiser4progs 1.0.3. Google gave me
> no answer on this and I am not able to track the problem down using strace.
> The fsck.log remains empty after the command above.
can you pack the metadata with
debugfs.reiser4 -P /dev/hdc3 | bzip2 -c > hdc3.bz2
using 1.0.3 progs and provide it for downloading?
--
Thanks,
Vitaly Fertman
