Re: CVE Request: MySQL: MyISAM temporary file issue

[Marcus Meissner <meissner () suse de>]

On Wed, Sep 10, 2014 at 06:00:08PM +0200, Salvatore Bonaccorso wrote:
> Hi
>
> The changes for MySQL 5.5.39[1] and 5.6.20[2] contain a reference to
> the following issue, which could be exploited by a local user to run
> arbitrary code in context of the mysqld server.
>
> MyISAM temporary files could be used to mount a code-execution attack.
> (Bug #18045646).
>
> This is also tracked in[3] and [4] mentioning as relevant fix [5].
>
> Was a CVE already requested for this issue? If not, could one be
> assigned?
>
> Regards,
> Salvatore
>
>  [1] https://dev.mysql.com/doc/relnotes/mysql/5.5/en/news-5-5-39.html
>  [2] https://dev.mysql.com/doc/relnotes/mysql/5.6/en/news-5-6-20.html
>  [3] https://bugzilla.redhat.com/show_bug.cgi?id=1126271
>  [4] https://bugs.gentoo.org/show_bug.cgi?id=518718
>  [5] https://bazaar.launchpad.net/~mysql/mysql-server/5.5/revision/4638

Hi folks.

By reviewing our mysql and mariadb issues we correlated this unknown issue
to a bugreport submitted to us by an external researcher.

CVE-2014-4274

https://bugzilla.suse.com/show_bug.cgi?id=857678

The bugreport also has a reproducer.

Stefan "nordi" Nordhausen reported this issue to both Oracle and to us on
7th of January 2014.

Oracle fixed it in their 2014oct CPU.
MariaDB fixed it in 10.0.13 I think.


We had sadly lacked disclosure timeline control on this issue and reactions on
Oracle side were also, errm, delayed.

Ciao, Marcus