Contents of /xml/htdocs/proj/en/glep/glep-0059.html

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1.5 - (show annotations) (download) (as text)
Sun Jan 31 07:55:45 2010 UTC (9 years, 1 month ago) by robbat2
Branch: MAIN
Changes since 1.4: +46 -40 lines
File MIME type: text/html
Revise GLEP59 per Calchan questions: Python 2.5 is widely deployed now and provides SHA512. RIPEMD160 is broken. WHIRLPOOL added. Migration plan detail added.

1 <?xml version="1.0" encoding="utf-8" ?>
2 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
3 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
5 <head>
6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
7 <meta name="generator" content="Docutils 0.6: http://docutils.sourceforge.net/" />
8 <title>GLEP 59 -- Manifest2 hash policies and security implications</title>
9 <link rel="stylesheet" href="tools/glep.css" type="text/css" /></head>
10 <body bgcolor="white">
11 <table class="navigation" cellpadding="0" cellspacing="0"
12 width="100%" border="0">
13 <tr><td class="navicon" width="150" height="35">
14 <a href="http://www.gentoo.org/" title="Gentoo Linux Home Page">
15 <img src="http://www.gentoo.org/images/gentoo-new.gif" alt="[Gentoo]"
16 border="0" width="150" height="35" /></a></td>
17 <td class="textlinks" align="left">
18 [<b><a href="http://www.gentoo.org/">Gentoo Linux Home</a></b>]
19 [<b><a href="http://www.gentoo.org/proj/en/glep">GLEP Index</a></b>]
20 [<b><a href="http://www.gentoo.org/proj/en/glep/glep-0059.txt">GLEP Source</a></b>]
21 </td></tr></table>
22 <table class="rfc2822 docutils field-list" frame="void" rules="none">
23 <col class="field-name" />
24 <col class="field-body" />
25 <tbody valign="top">
26 <tr class="field"><th class="field-name">GLEP:</th><td class="field-body">59</td>
27 </tr>
28 <tr class="field"><th class="field-name">Title:</th><td class="field-body">Manifest2 hash policies and security implications</td>
29 </tr>
30 <tr class="field"><th class="field-name">Version:</th><td class="field-body">1.4</td>
31 </tr>
32 <tr class="field"><th class="field-name">Last-Modified:</th><td class="field-body"><a class="reference external" href="http://www.gentoo.org/cgi-bin/viewcvs.cgi/xml/htdocs/proj/en/glep/glep-0059.txt?cvsroot=gentoo">2010/01/13 03:26:53</a></td>
33 </tr>
34 <tr class="field"><th class="field-name">Author:</th><td class="field-body">Robin Hugh Johnson &lt;robbat2&#32;&#97;t&#32;gentoo.org&gt;,</td>
35 </tr>
36 <tr class="field"><th class="field-name">Status:</th><td class="field-body">Draft</td>
37 </tr>
38 <tr class="field"><th class="field-name">Type:</th><td class="field-body">Standards Track</td>
39 </tr>
40 <tr class="field"><th class="field-name">Content-Type:</th><td class="field-body"><a class="reference external" href="glep-0002.html">text/x-rst</a></td>
41 </tr>
42 <tr class="field"><th class="field-name">Requires:</th><td class="field-body"><a class="reference external" href="http://www.gentoo.org/proj/en/glepglep-0044.html">44</a></td>
43 </tr>
44 <tr class="field"><th class="field-name">Created:</th><td class="field-body">October 2006</td>
45 </tr>
46 <tr class="field"><th class="field-name">Updated:</th><td class="field-body">November 2007, June 2008, July 2008, October 2008, January 2010</td>
47 </tr>
48 <tr class="field"><th class="field-name">Updates:</th><td class="field-body">44</td>
49 </tr>
50 <tr class="field"><th class="field-name">Post-History:</th><td class="field-body">December 2009, January 2010</td>
51 </tr>
52 </tbody>
53 </table>
54 <hr />
55 <div class="contents topic" id="contents">
56 <p class="topic-title first">Contents</p>
57 <ul class="simple">
58 <li><a class="reference internal" href="#abstract" id="id1">Abstract</a></li>
59 <li><a class="reference internal" href="#motivation" id="id2">Motivation</a></li>
60 <li><a class="reference internal" href="#specification" id="id3">Specification</a><ul>
61 <li><a class="reference internal" href="#the-bad-news" id="id4">The bad news</a></li>
62 <li><a class="reference internal" href="#how-fast-can-md5-be-broken" id="id5">How fast can MD5 be broken?</a></li>
63 <li><a class="reference internal" href="#the-good-news" id="id6">The good news</a></li>
64 <li><a class="reference internal" href="#what-should-be-done" id="id7">What should be done</a></li>
65 <li><a class="reference internal" href="#checksum-depreciation-timing" id="id8">Checksum depreciation timing</a></li>
66 </ul>
67 </li>
68 <li><a class="reference internal" href="#backwards-compatibility" id="id9">Backwards Compatibility</a></li>
69 <li><a class="reference internal" href="#references" id="id10">References</a></li>
70 <li><a class="reference internal" href="#thanks-to" id="id11">Thanks to</a></li>
71 <li><a class="reference internal" href="#copyright" id="id12">Copyright</a></li>
72 </ul>
73 </div>
74 <div class="section" id="abstract">
75 <h1><a class="toc-backref" href="#id1">Abstract</a></h1>
76 <p>While Manifest2 format allows multiple hashes, the question of which
77 checksums should be present, why, and the security implications of such
78 have never been resolved. This GLEP covers all of these issues, and
79 makes recommendations as to how to handle checksums both now, and in
80 future.</p>
81 </div>
82 <div class="section" id="motivation">
83 <h1><a class="toc-backref" href="#id2">Motivation</a></h1>
84 <p>This GLEP is being written as part of the work on signing the Portage
85 tree, but is only tangentially related to the actual signing of
86 Manifests. Checksums present one possible weak point in the overall
87 security of the tree - and a comprehensive security plan is needed.</p>
88 </div>
89 <div class="section" id="specification">
90 <h1><a class="toc-backref" href="#id3">Specification</a></h1>
91 <div class="section" id="the-bad-news">
92 <h2><a class="toc-backref" href="#id4">The bad news</a></h2>
93 <p>First of all, I'd like to cover the bad news in checksum security.
94 A much discussed point, as been the simple question: What is the
95 security of multiple independent checksums on the same data?
96 The most common position (and indeed the one previously held by myself),
97 is that multiple checksums would be an increase in security, but we
98 could not provably quantify the amount of security this added.
99 The really bad news, is that this position is completely and utterly
100 wrong. Many of you will be aghast at this. There is extremely little
101 added security in multiple checksums as noted by Joux [J04]. For any set
102 of checksums, the actual strength lies in that of the strongest
103 checksum.</p>
104 <p>Wang et al [W04] extended Joux's [J04] work on SHA-0 to cover MD4, MD5,
105 HAVAL-128 and RIPEMD families of hashes.</p>
106 </div>
107 <div class="section" id="how-fast-can-md5-be-broken">
108 <h2><a class="toc-backref" href="#id5">How fast can MD5 be broken?</a></h2>
109 <p>For a general collision, not a pre-image attack, since the announcement
110 by Wang et al [W04], the time required to break MD5 has been massively
111 reduced. Originally at 1 hour on a near-supercomputer (IBM P690) and
112 estimated at 64 hours with a Pentium-3 1.7Ghz. This has gone down to
113 less than in two years, to 17 seconds [K06a].</p>
114 <p>08/2004 - 1 hour, IBM pSeries 690 (32x 1.7Ghz POWER4+) = 54.4 GHz-Hours
115 03/2005 - 8 hours, Pentium-M 1.6Ghz = 12.8 Ghz-Hours
116 11/2005 - 5 hours, Pentium-4 1.7Ghz = 8.5 Ghz-Hours
117 03/2006 - 1 minute, Pentium-4 3.2Ghz = .05 Ghz-Hours
118 04/2006 - 17 seconds, Pentium-4 3.2Ghz = .01 Ghz-Hours</p>
119 <p>If we accept a factor of 800x as a sample of how much faster a checksum
120 may be broken over the course of 2 years (MD5 using the above data is
121 &gt;2000x), then existing checksums do not stand a significant chance of
122 survival in the future. We should thus accept that whatever checksums we
123 are using today, will be broken in the near future, and plan as best as
124 possible. (A brief review [H04] of the SHA1 attacks indicates an
125 improvement of ~600x in the same timespan).</p>
126 <p>And for those that claim implementation of these procedures is not yet
127 feasible, see [K06b] for an application that can produce two
128 self-extracting EXE files, with identical MD5s, and whatever payload you
129 want.</p>
130 </div>
131 <div class="section" id="the-good-news">
132 <h2><a class="toc-backref" href="#id6">The good news</a></h2>
133 <p>Of the checksums presently used by Manifest2 (SHA1, SHA256, RIPEMD160),
134 one stands close to being completely broken: SHA1; and another is
135 significantly weakened: RIPEMD160. The SHA2 series has suffered some
136 attacks, but still remains reasonably solid [G07],[K08].</p>
137 <p>To reduce the potential for future problems and any single checksum
138 break leading to a rapid decrease in security, we should incorporate the
139 strongest hash available from each family of checksums, and be prepared
140 to retire old checksums actively, unless there is a overriding reason to
141 keep a specific checksum, such as part of a migration plan.</p>
142 </div>
143 <div class="section" id="what-should-be-done">
144 <h2><a class="toc-backref" href="#id7">What should be done</a></h2>
145 <p>Portage should always try to verify all supported hashes that are
146 available in a Manifest2, starting with the strongest ones as maintained
147 by a preference list. Over time, the weaker checksums should be removed
148 from Manifest2 files, once all old Portage installations have had
149 sufficient time to upgrade. We should be prepared to add stronger
150 checksums wherever possible, and to remove those that have been
151 defeated.</p>
152 <p>As soon as feasible, we should add the SHA512 and WHIRLPOOL algorithms.
153 In future, as stream-based checksums are developed (in response to the
154 development by NIST [AHS]), they should be considered and used.</p>
155 <p>The SHA512 algorithm is available in Python 2.5, which has been a
156 dependency of Portage since approximately Python</p>
157 <p>The WHIRLPOOL checksum is not available within the PyCrypto library or
158 hashlib that is part of Python 2.5, but there are multiple alternative
159 Python implementations available, ranging from pure Python to C-based
160 (python-mhash).</p>
161 <p>The existence unsupported hash is not considered to be a failure unless
162 no supported hashes are available for a given Manifest entry.</p>
163 </div>
164 <div class="section" id="checksum-depreciation-timing">
165 <h2><a class="toc-backref" href="#id8">Checksum depreciation timing</a></h2>
166 <p>For the current Portage, both SHA1 and RIPEMD160 should be immediately
167 removed, as they present no advantages over the already present SHA256.
168 SHA256 cannot be replaced immediately with SHA512, as existing Portage
169 versions need at least one supported algorithm present (SHA256 support
170 was added in June 2006), so it must be retained for some while.</p>
171 <p>Immediately:
172 - Add WHIRLPOOL and SHA512.
173 - Remove SHA1 and RIPEMD160.</p>
174 <p>After the majority of Portage installations include SHA512 support:
175 - Remove SHA256.</p>
176 </div>
177 </div>
178 <div class="section" id="backwards-compatibility">
179 <h1><a class="toc-backref" href="#id9">Backwards Compatibility</a></h1>
180 <p>Old versions of Portage may support and expect only specific checksums.
181 This is accounted for in the checksum depreciation discussion.</p>
182 </div>
183 <div class="section" id="references">
184 <h1><a class="toc-backref" href="#id10">References</a></h1>
185 <dl class="docutils">
186 <dt>[AHS] NIST (2007). &quot;NIST's Plan for New Cryptographic Hash Functions&quot;,</dt>
187 <dd>(Advanced Hash Standard). <a class="reference external" href="http://csrc.nist.gov/pki/HashWorkshop/">http://csrc.nist.gov/pki/HashWorkshop/</a></dd>
188 <dt>[BOBO06] Boneh, D. and Boyen, X. (2006). &quot;On the Impossibility of</dt>
189 <dd>Efficiently Combining Collision Resistant Hash Functions&quot;; Proceedings
190 of CRYPTO 2006, Dwork, C. (Ed.); Lecture Notes in Computer Science
191 4117, pp. 570-583. Available online from:
192 <a class="reference external" href="http://crypto.stanford.edu/~dabo/abstracts/hashing.html">http://crypto.stanford.edu/~dabo/abstracts/hashing.html</a></dd>
193 <dt>[H04] Hawkes, P. and Paddon, M. and Rose, G. (2004). &quot;On Corrective</dt>
194 <dd>Patterns for the SHA-2 Family&quot;. CRYPTO 2004 Cryptology ePrint Archive,
195 Report 2004/204. Available online from:
196 <a class="reference external" href="http://eprint.iacr.org/2004/207.pdf">http://eprint.iacr.org/2004/207.pdf</a></dd>
197 <dt>[J04] Joux, Antoie. (2004). &quot;Multicollisions in Iterated Hash</dt>
198 <dd>Functions - Application to Cascaded Constructions;&quot; Proceedings of
199 CRYPTO 2004, Franklin, M. (Ed); Lecture Notes in Computer Science
200 3152, pp. 306-316. Available online from:
201 <a class="reference external" href="http://web.cecs.pdx.edu/~teshrim/spring06/papers/general-attacks/multi-joux.pdf">http://web.cecs.pdx.edu/~teshrim/spring06/papers/general-attacks/multi-joux.pdf</a></dd>
202 <dt>[K06a] Klima, V. (2006). &quot;Tunnels in Hash Functions: MD5 Collisions</dt>
203 <dd>Within a Minute&quot;. Cryptology ePrint Archive, Report 2006/105.
204 Available online from: <a class="reference external" href="http://eprint.iacr.org/2006/105.pdf">http://eprint.iacr.org/2006/105.pdf</a></dd>
205 <dt>[K06b] Klima, V. (2006). &quot;Note and links to high-speed MD5 collision</dt>
206 <dd>proof of concept tools&quot;. Available online from:
207 <a class="reference external" href="http://cryptography.hyperlink.cz/2006/trick.txt">http://cryptography.hyperlink.cz/2006/trick.txt</a></dd>
208 <dt>[K08] Klima, V. (2008). &quot;On Collisions of Hash Functions Turbo SHA-2&quot;.</dt>
209 <dd>Cryptology ePrint Archive, Report 2008/003. Available online from:
210 <a class="reference external" href="http://eprint.iacr.org/2008/003.pdf">http://eprint.iacr.org/2008/003.pdf</a></dd>
211 <dt>[G07] Gligoroski, D. and Knapskog, S.J. (2007). &quot;Turbo SHA-2&quot;.</dt>
212 <dd>Cryptology ePrint Archive, Report 2007/403. Available online from:
213 <a class="reference external" href="http://eprint.iacr.org/2007/403.pdf">http://eprint.iacr.org/2007/403.pdf</a></dd>
214 <dt>[W04] Wang, X. et al: &quot;Collisions for Hash Functions MD4, MD5,</dt>
215 <dd>HAVAL-128 and RIPEMD&quot;, rump session, CRYPTO 2004, Cryptology ePrint
216 Archive, Report 2004/199, first version (August 16, 2004), second
217 version (August 17, 2004). Available online from:
218 <a class="reference external" href="http://eprint.iacr.org/2004/199.pdf">http://eprint.iacr.org/2004/199.pdf</a></dd>
219 </dl>
220 </div>
221 <div class="section" id="thanks-to">
222 <h1><a class="toc-backref" href="#id11">Thanks to</a></h1>
223 <dl class="docutils">
224 <dt>I'd like to thank the following folks, in no specific order:</dt>
225 <dd><ul class="first last simple">
226 <li>Ciaran McCreesh (ciaranm) - for pointing out the Joux (2004) paper,
227 and also being stubborn enough in not accepting a partial solution.</li>
228 <li>Marius Mauch (genone), Zac Medico (zmedico) and Brian Harring
229 (ferringb): for being knowledgeable about the Portage Manifest2
230 codebase.</li>
231 </ul>
232 </dd>
233 </dl>
234 </div>
235 <div class="section" id="copyright">
236 <h1><a class="toc-backref" href="#id12">Copyright</a></h1>
237 <p>Copyright (c) 2006-2010 by Robin Hugh Johnson. This material may be
238 distributed only subject to the terms and conditions set forth in the
239 Open Publication License, v1.0.</p>
240 <p>vim: tw=72 ts=2 expandtab:</p>
241 </div>
243 </div>
244 <div class="footer">
245 <hr class="footer" />
246 <a class="reference external" href="glep-0059.txt">View document source</a>.
247 Generated on: 2010-01-31 07:53 UTC.
248 Generated by <a class="reference external" href="http://docutils.sourceforge.net/">Docutils</a> from <a class="reference external" href="http://docutils.sourceforge.net/rst.html">reStructuredText</a> source.
250 </div>
251 </body>
252 </html>

  ViewVC Help
Powered by ViewVC 1.1.20