mailing list archives
Re: MySQL 5.1/5.5 WiNDOWS REMOTE R00T (mysqljackpot)
From: Jeffrey Walton <noloader () gmail com>
Date: Wed, 5 Dec 2012 20:38:53 -0500
On Mon, Dec 3, 2012 at 11:03 AM, king cope
<isowarez.isowarez.isowarez () googlemail com> wrote:
Yes I agree, we should discard this default remote vulnerability
because it is documented.
Devil's advocate: Does a questionable design choice/feature that is
documented make it any less vulnerable?
How does a Mom and Pop shop who were told to get mySQL to support
<some business software> mitigate this issue when its insecure out of
the box and there are no IT resources?
2012/12/2 Sergei Golubchik <serg () askmonty org>:
2012/12/2 Kurt Seifried <kseifried () redhat com>:
*** FARLiGHT ELiTE HACKERS LEGACY R3L3ASE ***
Attached is the MySQL Windows Remote Exploit (post-auth, udf
technique) including the previously released mass scanner. The
exploit is mirrored at the farlight website
So in the case of this issue it appears to be documented (UDF, do
not run MySQL as administrator, etc.). As I understand CVE
assignment rules this issue does not require a CVE, however just to
be on the safe side I'm CC'ing MySQL, Oracle, MariaDB, OSS-SEC,
Steven Christey, cve-assign and OSVDB to the CC so that everyone is
aware of what is going on.
Just to confirm - yes, it's documented.
UDF is a feature that allows to run any code in the MySQL server
process. FILE privilege allows to create files. So yes, sure, with the
appropriate privileges and the appropriately configured server
one can create a file and load it as UDF. As expected.
Full-Disclosure - We believe in it.
Hosted and sponsored by Secunia - http://secunia.com/