problem with corrupted filesystem

problem with corrupted filesystem

[Marcus Becker]

I had some folders left after failed emerge in /var/tmp/portage, so I tried to remove the folders in there and got the error "directory not empty"
I was able to move the subfolder "locale" to my /root folder and then remove the directory with all other subdirectories
in the /root folder I still wasn't able to remove "locale"

After fsck.reiser4 --fix /dev/sda3 problem is solved, here is the relevant output:

disi-laptop ~ # uname -a
Linux disi-laptop 2.6.33-zen1 #3 ZEN SMP Sat Mar 27 17:00:34 GMT 2010 
i686 Intel(R) Pentium(R) M processor 1.60GHz GenuineIntel GNU/Linux

disi-laptop ~ # mkfs.reiser4 --version
mkfs.reiser4 1.0.7
Copyright (C) 2001-2005 by Hans Reiser, licensing governed by 
reiser4progs/COPYING.
disi-laptop ~ # ls
locale  xorg.conf.new
disi-laptop ~ # ls -l
total 5
drwx------ 2 root root    4 Mar 27 15:35 locale
-rw-r--r-- 1 root root 3944 Mar 27 15:05 xorg.conf.new
disi-laptop ~ # man ls
disi-laptop ~ # ls -li
total 5
509520 drwx------ 2 root root    4 Mar 27 15:35 locale
1584446 -rw-r--r-- 1 root root 3944 Mar 27 15:05 xorg.conf.new
disi-laptop ~ # mv locale locale_moved
disi-laptop ~ # ls -li
total 5
509520 drwx------ 2 root root    4 Mar 29 16:58 locale_moved
1584446 -rw-r--r-- 1 root root 3944 Mar 27 15:05 xorg.conf.new
disi-laptop ~ # rm -rf locale_moved/
rm: cannot remove `locale_moved': Directory not empty
disi-laptop ~ # ls -lai locale_moved/
total 1
509520 drwx------ 2 root root 4 Mar 29 16:58 .
99103 drwx------ 3 root root 9 Mar 29 16:58 ..
disi-laptop ~ # rmdir locale_moved/
rmdir: failed to remove `locale_moved/': Directory not empty
-- 3 root root 9 Mar 29 16:58 ..

fsck_sda3.log:
FSCK: obj40_repair.c: 223: obj40_stat_unix_check: Node (411349), item (3), 
[7c641:6c6f63616c6500:7c650] (stat40): wrong bytes (264), Fixed to (100). 
FSCK: obj40_repair.c: 350: obj40_stat_lw_check: Node (411349), item (3),
[7c641:6c6f63616c6500:7c650] (stat40): wrong size (4), Fixed to (2). 


disi-laptop ~ # ls
locale_moved  xorg.conf.new
disi-laptop ~ # ls -li
total 5
509520 drwx------ 2 root root    2 Mar 29 16:58 locale_moved
1584446 -rw-r--r-- 1 root root 3944 Mar 27 15:05 xorg.conf.new
disi-laptop ~ # mv locale_moved locale
disi-laptop ~ # ls -li
total 5
509520 drwx------ 2 root root    2 Mar 29 17:20 locale
1584446 -rw-r--r-- 1 root root 3944 Mar 27 15:05 xorg.conf.new
disi-laptop ~ # rm -r locale
disi-laptop ~ # ls -li
total 4
1584446 -rw-r--r-- 1 root root 3944 Mar 27 15:05 xorg.conf.new

Here is the related forum post: http://forums.gentoo.org/viewtopic-p-6223315.html#6223315

thanks,

Marcus
___________________________________________________________
WEB.DE DSL: Internet, Telefon und Entertainment für nur 19,99 EUR/mtl.!
http://produkte.web.de/go/02/
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: problem with corrupted filesystem

[Marcus Becker]

kk, thanks for the answer

I just read that other mail from Jonas and thought I give it a shot.
Nothing changed on the filesystem.
If you need more information, let me know.

Marcus

-----Ursprüngliche Nachricht-----
Von: Edward Shishkin <edward.shishkin@gmail.com>
Gesendet: Mar 30, 2010 11:07:32 PM
An: Marcus Becker 
Betreff: Re: problem with corrupted filesystem

>Marcus Becker wrote:
>> Hi,
>>
>> I tried using hash=tea as mount option with the same result.
>>   
>
>I don't think that reiser4 supports such mount option
>
>> Is this something I can change on an already formatted partition?
>>   
>
>What do you want to change? IMHO the type of hash can be assigned only
>in mkfs time. There was an interface to change it on-line per-directory (via
>so-called pseudo-files), but this is still deprecated and is not used.
>
>> Where can I check, which hash is used?
>
>In the plugin table of (root) directory of debugfs.reiser4 -t output.
>Attributes of other objects are inherited through the semantic tree
>by default and, hence, are not shown.
>
>In your case (that you have sent me) we have the following:
>
>NODE (715370) LEVEL=1 ITEMS=22 SPACE=0 MKFS ID=0x7c4de18e FLUSH=0x0
>#0  SD (stat40): [29:1(SD):0:2a:0] OFF=28, LEN=94, flags=0x0
>[...]
>permission : id = 0
>formatting : id = 2 (smart)
>hash : id = 1 (r5_hash) <<<<<<<<<<<<<<<
>fibration : id = 2 (ext_1_fibre)
>statdata : id = 0 (stat40)
>diritem : id = 2 (cde40)
>crypto : id = 0
>digest : id = 0
>compress : id = 0 (lzo1)
>compressMode : id = 4 (conv)
>cluster : id = 0 (64K)
>create : id = 0 (reg40)
>
>Edward.
>
>>  /proc/mounts doesn't list the option.
>>
>> Thanks,
>>
>> Marcus
>>
>
___________________________________________________________
GRATIS für alle WEB.DE-Nutzer: Die maxdome Movie-FLAT!
Jetzt freischalten unter http://movieflat.web.de
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: problem with corrupted filesystem

[Edward Shishkin]

Marcus Becker wrote:
> Hi,
>
> I tried using hash=tea as mount option with the same result.
>   

I don't think that reiser4 supports such mount option

> Is this something I can change on an already formatted partition?
>   

What do you want to change? IMHO the type of hash can be assigned only
in mkfs time. There was an interface to change it on-line per-directory (via
so-called pseudo-files), but this is still deprecated and is not used.

> Where can I check, which hash is used?

In the plugin table of (root) directory of debugfs.reiser4 -t output.
Attributes of other objects are inherited through the semantic tree
by default and, hence, are not shown.

In your case (that you have sent me) we have the following:

NODE (715370) LEVEL=1 ITEMS=22 SPACE=0 MKFS ID=0x7c4de18e FLUSH=0x0
#0  SD (stat40): [29:1(SD):0:2a:0] OFF=28, LEN=94, flags=0x0
[...]
permission : id = 0
formatting : id = 2 (smart)
hash : id = 1 (r5_hash) <<<<<<<<<<<<<<<
fibration : id = 2 (ext_1_fibre)
statdata : id = 0 (stat40)
diritem : id = 2 (cde40)
crypto : id = 0
digest : id = 0
compress : id = 0 (lzo1)
compressMode : id = 4 (conv)
cluster : id = 0 (64K)
create : id = 0 (reg40)

Edward.

>  /proc/mounts doesn't list the option.
>
> Thanks,
>
> Marcus
>
>
> -----Urspr

Re: problem with corrupted filesystem

[Marcus Becker]

Hi,

I tried using hash=tea as mount option with the same result.
Is this something I can change on an already formatted partition?
Where can I check, which hash is used? /proc/mounts doesn't list the option.

Thanks,

Marcus


-----Urspr

Re: problem with corrupted filesystem

[Marcus Becker]

Nothing wrong with it, just couldn't clean up afterwards. No crash etc.
emerge uses /var/tmp/portage as a build directory and when the binaries are finished it moves them to the live system and removes all files in there.
It cleans all other files/folders but cannot remove that folder (same as me).

Actual, I built gcc just again to see if the same happens again.
this time I ran it again after moving the folder to /root and without fsck. 
It ended up in a segmentation fault
Also I have another "locale" folder that cannot be removed.
Third time, same happens.

I'll now go ahead and do fsck again, remove those 2 folders and see afterwards if gcc builds fine again.

Always different inode, same problem with that "locale" folder
I can emerge other ebuilds without problems.

Cheers,

Marcus

-----Urspr

Re: problem with corrupted filesystem

[Edward Shishkin]

Marcus Becker wrote:
> Hi Edward,
>
> I uploaded the file here:
> http://omploader.org/vM3pxYw
>   

Ok, got it. Thanks.

So what's up with your emerge? Why did it fail? With what error?
Was is a system crash, power loss, etc?

Edward.

> hope I can help,
>
> Marcus
>
>
> -----Ursprüngliche Nachricht-----
> Von: Edward Shishkin <edward.shishkin@gmail.com>
> Gesendet: Mar 30, 2010 12:24:04 PM
> An: Marcus Becker 
> Betreff: Re: problem with corrupted filesystem
>
>   
>> Marcus Becker wrote:
>>     
>>> the problem is actual reproduceable. I emerged gcc again and the excact same happened:
>>>
>>> disi-laptop portage # rm -r *
>>> rm: cannot remove `sys-devel/gcc-4.4.3/work/gcc-4.4.3/libjava/classpath/resource/gnu/java/locale': Directory not empty
>>>   
>>>       
>> Hello.
>>
>> Would you please pack the metadata by
>> debugfs.reiser4 -P /dev/xxx | gzip > meta.gz
>> and let me download the file meta.gz
>>
>> Thanks,
>> Edward.
>>
>>
>>     
>>>  
>>>
>>> I am not sure, why it doesn't like that particular folder..
>>>       
> ___________________________________________________________
> WEB.DE DSL: Internet, Telefon und Entertainment für nur 19,99 EUR/mtl.!
> http://produkte.web.de/go/02/
>
>   

--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: problem with corrupted filesystem

[Marcus Becker]

Hi Edward,

I uploaded the file here:
http://omploader.org/vM3pxYw

hope I can help,

Marcus


-----Ursprüngliche Nachricht-----
Von: Edward Shishkin <edward.shishkin@gmail.com>
Gesendet: Mar 30, 2010 12:24:04 PM
An: Marcus Becker 
Betreff: Re: problem with corrupted filesystem

>Marcus Becker wrote:
>> the problem is actual reproduceable. I emerged gcc again and the excact same happened:
>>
>> disi-laptop portage # rm -r *
>> rm: cannot remove `sys-devel/gcc-4.4.3/work/gcc-4.4.3/libjava/classpath/resource/gnu/java/locale': Directory not empty
>>   
>
>Hello.
>
>Would you please pack the metadata by
>debugfs.reiser4 -P /dev/xxx | gzip > meta.gz
>and let me download the file meta.gz
>
>Thanks,
>Edward.
>
>
>>  
>>
>> I am not sure, why it doesn't like that particular folder..
>
___________________________________________________________
WEB.DE DSL: Internet, Telefon und Entertainment für nur 19,99 EUR/mtl.!
http://produkte.web.de/go/02/
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: problem with corrupted filesystem

[Edward Shishkin]

Marcus Becker wrote:
> the problem is actual reproduceable. I emerged gcc again and the excact same happened:
>
> disi-laptop portage # rm -r *
> rm: cannot remove `sys-devel/gcc-4.4.3/work/gcc-4.4.3/libjava/classpath/resource/gnu/java/locale': Directory not empty
>   

Hello.

Would you please pack the metadata by
debugfs.reiser4 -P /dev/xxx | gzip > meta.gz
and let me download the file meta.gz

Thanks,
Edward.


>  
>
> I am not sure, why it doesn't like that particular folder..

problem with corrupted filesystem

[Marcus Becker]

the problem is actual reproduceable. I emerged gcc again and the excact same happened:

disi-laptop portage # rm -r *
rm: cannot remove `sys-devel/gcc-4.4.3/work/gcc-4.4.3/libjava/classpath/resource/gnu/java/locale': Directory not empty

 

I am not sure, why it doesn't like that particular folder..
___________________________________________________________
GRATIS für alle WEB.DE-Nutzer: Die maxdome Movie-FLAT!
Jetzt freischalten unter http://movieflat.web.de
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html