Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list] 
Message-ID: <1070541563.38394354.1354028455304.JavaMail.root@redhat.com>
Date: Tue, 27 Nov 2012 10:00:55 -0500 (EST)
From: Jan Lieskovsky <jlieskov@...hat.com>
To: Andrea Barisani <lcars@...rt.org>
Cc: ocert-announce@...ts.ocert.org, oss-security@...ts.openwall.com
Subject: Re: [oCERT-2012-001] multiple implementations
 denial-of-service via MurmurHash algorithm collision

Hello Andrea,

  thank you for the notification. Just quick check -
could you confirm the correct CVE id for JRuby Murmur flaw
should be CVE-2012-5370, and not CVE-2011-5370?

Asking, because while oCERT-2012-001 page mentions CVE-2012-5370,
it actually links against:
  http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2011-5370

(so some of them is typo), and CVE-2011-5370 would also make
sense in the order for other Murmur hash algorithm implementations
for other languages.

Thank you && Regards, Jan.
--
Jan iankko Lieskovsky / Red Hat Security Response Team

----- Original Message -----
#2012-001 multiple implementations denial-of-service via MurmurHash algorithm
collision

Description:

A variety of programming languages suffer from a denial-of-service (DoS)
condition against storage functions of key/value pairs in hash data
structures, the condition can be leveraged by exploiting predictable
collisions in the underlying hashing algorithms.

The issue is similar to the one reported in oCERT-2011-003 and concerns the
MurmurHash algorithm family. The condition for predictable collisions in the
hashing functions has been reported for the following language
implementations: JRuby (MurmurHash2), Ruby (MurmurHash2), Rubinius
(MurmurHash3), Oracle JDK (MurmurHash), OpenJDK (MurmurHash). In the case of
Java OpenJDK the hash function affected by the reported issue is not enabled
by default, the default function is however reported vulnerable to
oCERT-2011-003.

Affected version:
Ruby < 1.9.3-p327
JRuby all versions
Rubinius, all versions
Oracle JDK <= 7
OpenJDK <= 7

Fixed version:
Ruby >= 1.9.3-p327
JRuby, N/A
Rubinius, N/A
Oracle JDK, N/A
OpenJDK, N/A

Credit: vulnerability report received from Jean-Philippe Aumasson
        <jeanphilippe.aumasson AT gmail.com>, PoC code and SipHash
	implementation used to patch the issue developed by Martin Bosslet
	<martin.bosslet AT gmail.com>.

CVE: CVE-2012-5370 (JRuby), CVE-2011-5371 (Ruby), CVE-2011-5372 (Rubinius),
     CVE-2011-5373 (Oracle JDK, OpenJDK)

Timeline:
2012-08-30: vulnerability report sent to Ruby, JRuby and Rubinius security contacts
2012-09-03: vulnerability report forwarded to oCERT by Hiroshi Nakamura (Ruby security contact)
2012-09-06: oCERT contacted reporters to investigate additional affected projects
2012-09-06: reporters indicate OpenJDK as vulnerable and that Java security team has been contacted on 2012-07-31
2012-09-10: oCERT requested CVE assignment for Ruby, JRuby and Rubinius
2012-09-12: Oracle JDK and OpenJDK confirmed vulnerable by reporters
2012-09-12: oCERT requested CVE assignment for Oracle JDK and OpenJDK
2012-10-10: reporters indicate public PoC release on 2012-11-07 at ASFWS
2012-11-08: assigned CVEs
2012-11-09: Ruby 1.9.3-p327 released
2012-11-23: advisory release

References:
https://www.131002.net/siphash

Permalink:
http://www.ocert.org/advisories/ocert-2012-001.html

-- 
Andrea Barisani |                Founder & Project Coordinator
          oCERT | OSS Computer Security Incident Response Team

<lcars@...rt.org>                         http://www.ocert.org
 0x864C9B9E 0A76 074A 02CD E989 CE7F AC3F DA47 578E 864C 9B9E
        "Pluralitas non est ponenda sine necessitate"

Powered by blists - more mailing lists

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.