linux-kernel - Re: Fwd: EXT3 File System Corruption 2.6.34

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <4C0DA578.9010709@sandeen.net>
Date:	Mon, 07 Jun 2010 21:05:44 -0500
From:	Eric Sandeen <sandeen@...deen.net>
To:	Jeffrey Merkey <jeffmerkey@...il.com>
CC:	linux-kernel@...r.kernel.org
Subject: Re: Fwd: EXT3 File System Corruption 2.6.34

Jeffrey Merkey wrote:

> OK.  I will set this up.  You may want to make this option the default
> in the build scripts.  here is a corrupted file.  

It was default, but Linus changed it a while back.

> This was a .gif
> image file I saved THEN AFTER SAVING THE FILE I pulled the power to
> the machine and during recovery the file was FUCKED.  

I assume your application did not sync the data, and buffered data
loss is expected on a power loss.

> At any rate,
> this does not happen with 2.6.28.

that I can't explain for sure.... different timing perhaps.

> I dumped the file with xdump a util I use internally for my own use so
> you could see the file contents as text and I could post it here.
> This was an image file but look what ended up in it -- directory
> blocks and such.  Take a look:

As I said, stale blocks exposed due to data=writeback.  Known behavior,
unfortunately the default for ext3.  If you find similar problems
when mounted data=ordered, it's a more interesting report.

-Eric

>           0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F
> 00000000  6C 73 0A 63 64 20 2E 2E 0A 63 6C 73 0A 6C 73 0A  ls.cd ...cls.ls.
> 00000010  63 64 20 6C 69 6E 75 78 2D 32 2E 36 2E 33 34 2D  cd linux-2.6.34-
> 00000020  6D 64 62 2F 0A 63 6C 73 0A 6C 73 0A 63 64 20 2E  mdb/.cls.ls.cd .
> 00000030  2E 0A 63 6C 73 0A 6C 73 0A 63 64 20 6C 69 6E 75  ..cls.ls.cd linu

<giant snip>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/