Annotation of www/security.html, Revision 1.422
1.294 david 1: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
2: "http://www.w3.org/TR/html4/loose.dtd">
1.1 deraadt 3: <html>
4: <head>
1.20 deraadt 5: <title>OpenBSD Security</title>
1.345 tom 6: <link rel="alternate" type="application/rss+xml" title="OpenBSD errata (external)" href="http://www.undeadly.org/cgi?action=errata">
1.294 david 7: <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
1.1 deraadt 8: <meta name="description" content="OpenBSD advisories">
1.403 sthen 9: <meta name="copyright" content="This document copyright 1997-2014 by OpenBSD.">
1.421 sthen 10: <link rel="canonical" href="http://www.openbsd.org/security.html">
1.1 deraadt 11: </head>
12:
1.274 david 13: <body bgcolor="#ffffff" text="#000000" link="#23238E">
1.210 jsyn 14: <a href="index.html"><img alt="[OpenBSD]" height="30" width="141" src="images/smalltitle.gif" border="0"></a>
1.106 deraadt 15: <p>
1.294 david 16: <h2><font color="#e00000">Security</font></h2>
17: <hr>
1.1 deraadt 18:
1.406 deraadt 19: For security advisories for specific releases, click below:
1.294 david 20: <p>
1.406 deraadt 21:
1.418 tedu 22: <a href="errata21.html">2.1</a>,
23: <a href="errata22.html">2.2</a>,
24: <a href="errata23.html">2.3</a>,
25: <a href="errata24.html">2.4</a>,
26: <a href="errata25.html">2.5</a>,
27: <a href="errata26.html">2.6</a>,
28: <a href="errata27.html">2.7</a>,
29: <a href="errata28.html">2.8</a>,
30: <a href="errata29.html">2.9</a>,
31: <a href="errata30.html">3.0</a>,
32: <a href="errata31.html">3.1</a>,
33: <a href="errata32.html">3.2</a>,
34: <a href="errata33.html">3.3</a>,
35: <a href="errata34.html">3.4</a>,
36: <a href="errata35.html">3.5</a>,
37: <a href="errata36.html">3.6</a>,
1.420 schwarze 38: <a href="errata37.html">3.7</a>,
1.365 deraadt 39: <br>
1.418 tedu 40: <a href="errata38.html">3.8</a>,
41: <a href="errata39.html">3.9</a>,
42: <a href="errata40.html">4.0</a>,
43: <a href="errata41.html">4.1</a>,
44: <a href="errata42.html">4.2</a>,
45: <a href="errata43.html">4.3</a>,
46: <a href="errata44.html">4.4</a>,
47: <a href="errata45.html">4.5</a>,
48: <a href="errata46.html">4.6</a>,
49: <a href="errata47.html">4.7</a>,
50: <a href="errata48.html">4.8</a>,
51: <a href="errata49.html">4.9</a>,
52: <a href="errata50.html">5.0</a>,
53: <a href="errata51.html">5.1</a>,
54: <a href="errata52.html">5.2</a>,
55: <a href="errata53.html">5.3</a>,
1.420 schwarze 56: <a href="errata54.html">5.4</a>,
1.406 deraadt 57: <br>
1.419 jsg 58: <a href="errata55.html">5.5</a>,
1.420 schwarze 59: <a href="errata56.html">5.6</a>,
60: <a href="errata57.html">5.7</a>.
1.406 deraadt 61: <br>
1.56 deraadt 62: <hr>
63:
1.294 david 64: <a name="goals"></a>
1.278 deraadt 65: <ul>
1.406 deraadt 66: <li><h3><font color="#e00000">Goals</font></h3><p>
1.22 deraadt 67:
1.14 deraadt 68: OpenBSD believes in strong security. Our aspiration is to be NUMBER
1.22 deraadt 69: ONE in the industry for security (if we are not already there). Our
70: open software development model permits us to take a more
71: uncompromising view towards increased security than Sun, SGI, IBM, HP,
72: or other vendors are able to. We can make changes the vendors would
1.27 deraadt 73: not make. Also, since OpenBSD is exported with <a href=crypto.html>
1.45 deraadt 74: cryptography</a>, we are able to take cryptographic approaches towards
75: fixing security problems.<p>
1.18 deraadt 76:
1.288 matthieu 77: <a name="disclosure"></a>
1.294 david 78: <li><h3><font color="#e00000">Full Disclosure</font></h3><p>
1.106 deraadt 79:
1.45 deraadt 80: Like many readers of the
1.196 jufi 81: <a href="http://online.securityfocus.com/archive/1">
1.18 deraadt 82: BUGTRAQ mailing list</a>,
1.106 deraadt 83: we believe in full disclosure of security problems. In the
84: operating system arena, we were probably the first to embrace
85: the concept. Many vendors, even of free software, still try
86: to hide issues from their users.<p>
87:
88: Security information moves very fast in cracker circles. On the other
89: hand, our experience is that coding and releasing of proper security
90: fixes typically requires about an hour of work -- very fast fix
91: turnaround is possible. Thus we think that full disclosure helps the
92: people who really care about security.<p>
93:
1.288 matthieu 94: <a name="process"></a>
1.294 david 95: <li><h3><font color="#e00000">Audit Process</font></h3><p>
1.15 deraadt 96:
1.12 deraadt 97: Our security auditing team typically has between six and twelve
1.45 deraadt 98: members who continue to search for and fix new security holes. We
99: have been auditing since the summer of 1996. The process we follow to
100: increase security is simply a comprehensive file-by-file analysis of
1.106 deraadt 101: every critical software component. We are not so much looking for
102: security holes, as we are looking for basic software bugs, and if
1.138 deraadt 103: years later someone discovers the problem used to be a security
1.106 deraadt 104: issue, and we fixed it because it was just a bug, well, all the
105: better. Flaws have been found in just about every area of the system.
106: Entire new classes of security problems have been found during our
107: audit, and often source code which had been audited earlier needs
108: re-auditing with these new flaws in mind. Code often gets audited
109: multiple times, and by multiple people with different auditing
110: skills.<p>
1.12 deraadt 111:
1.94 deraadt 112: Some members of our security auditing team worked for Secure Networks,
113: the company that made the industry's premier network security scanning
114: software package Ballista (Secure Networks got purchased by Network
115: Associates, Ballista got renamed to Cybercop Scanner, and well...)
116: That company did a lot of security research, and thus fit in well
1.106 deraadt 117: with the OpenBSD stance. OpenBSD passed Ballista's tests with flying
118: colours since day 1.<p>
1.31 deraadt 119:
1.34 deraadt 120: Another facet of our security auditing process is its proactiveness.
1.45 deraadt 121: In most cases we have found that the determination of exploitability
122: is not an issue. During our ongoing auditing process we find many
123: bugs, and endeavor to fix them even though exploitability is not
124: proven. We fix the bug, and we move on to find other bugs to fix. We
125: have fixed many simple and obvious careless programming errors in code
126: and only months later discovered that the problems were in fact
127: exploitable. (Or, more likely someone on
1.197 jufi 128: <a href="http://online.securityfocus.com/archive/1">BUGTRAQ</a>
1.45 deraadt 129: would report that other operating systems were vulnerable to a `newly
130: discovered problem', and then it would be discovered that OpenBSD had
131: been fixed in a previous release). In other cases we have been saved
132: from full exploitability of complex step-by-step attacks because we
133: had fixed one of the intermediate steps. An example of where we
1.94 deraadt 134: managed such a success is the lpd advisory that Secure Networks put out.
135: <p>
1.29 deraadt 136:
1.288 matthieu 137: <a name="newtech"></a>
1.294 david 138: <li><h3><font color="#e00000">New Technologies</font></h3><p>
1.278 deraadt 139:
140: As we audit source code, we often invent new ways of solving problems.
141: Sometimes these ideas have been used before in some random application
142: written somewhere, but perhaps not taken to the degree that we do.
143: <p>
144:
145: <ul>
146: <li>strlcpy() and strlcat()
147: <li>Memory protection purify
148: <ul>
149: <li>W^X
150: <li>.rodata segment
151: <li>Guard pages
152: <li>Randomized malloc()
153: <li>Randomized mmap()
154: <li>atexit() and stdio protection
155: </ul>
1.295 otto 156: <li>Privilege separation
1.278 deraadt 157: <li>Privilege revocation
158: <li>Chroot jailing
159: <li>New uids
160: <li>ProPolice
161: <li>... and others
162: </ul>
163: <p>
164:
1.294 david 165: <li><h3><font color="#e00000">The Reward</font></h3><p>
1.106 deraadt 166:
1.45 deraadt 167: Our proactive auditing process has really paid off. Statements like
1.35 deraadt 168: ``This problem was fixed in OpenBSD about 6 months ago'' have become
1.45 deraadt 169: commonplace in security forums like
1.197 jufi 170: <a href="http://online.securityfocus.com/archive/1">BUGTRAQ</a>.<p>
1.35 deraadt 171:
1.45 deraadt 172: The most intense part of our security auditing happened immediately
1.80 espie 173: before the OpenBSD 2.0 release and during the 2.0->2.1 transition,
1.45 deraadt 174: over the last third of 1996 and first half of 1997. Thousands (yes,
175: thousands) of security issues were fixed rapidly over this year-long
176: period; bugs like the standard buffer overflows, protocol
177: implementation weaknesses, information gathering, and filesystem
178: races. Hence most of the security problems that we encountered were
179: fixed before our 2.1 release, and then a far smaller number needed
180: fixing for our 2.2 release. We do not find as many problems anymore,
181: it is simply a case of diminishing returns. Recently the security
182: problems we find and fix tend to be significantly more obscure or
183: complicated. Still we will persist for a number of reasons:<p>
1.36 deraadt 184:
1.35 deraadt 185: <ul>
1.45 deraadt 186: <li>Occasionally we find a simple problem we missed earlier. Doh!
1.35 deraadt 187: <li>Security is like an arms race; the best attackers will continue
1.45 deraadt 188: to search for more complicated exploits, so we will too.
189: <li>Finding and fixing subtle flaws in complicated software is
190: a lot of fun.
1.35 deraadt 191: </ul>
1.106 deraadt 192: <p>
1.15 deraadt 193:
1.14 deraadt 194: The auditing process is not over yet, and as you can see we continue
1.28 deraadt 195: to find and fix new security flaws.<p>
1.12 deraadt 196:
1.288 matthieu 197: <a name="default"></a>
1.294 david 198: <li><h3><font color="#e00000">"Secure by Default"</font></h3><p>
1.106 deraadt 199:
200: To ensure that novice users of OpenBSD do not need to become security
201: experts overnight (a viewpoint which other vendors seem to have), we
202: ship the operating system in a Secure by Default mode. All non-essential
203: services are disabled. As the user/administrator becomes more familiar
204: with the system, he will discover that he has to enable daemons and other
205: parts of the system. During the process of learning how to enable a new
206: service, the novice is more likely to learn of security considerations.<p>
207:
208: This is in stark contrast to the increasing number of systems that
209: ship with NFS, mountd, web servers, and various other services enabled
210: by default, creating instantaneous security problems for their users
211: within minutes after their first install.<p>
212:
1.288 matthieu 213: <a name="crypto"></a>
1.294 david 214: <li><h3><font color="#e00000">Cryptography</font></h3><p>
1.106 deraadt 215:
216: And of course, since the OpenBSD project is based in Canada, it is possible
217: for us to integrate cryptography. For more information, read the page
1.116 deraadt 218: outlining <a href=crypto.html>what we have done with cryptography</a>.</p>
1.106 deraadt 219:
1.294 david 220: <li><h3><font color="#e00000">Advisories</font></h3><p>
1.106 deraadt 221:
1.418 tedu 222: Please refer to the links at the top of this page.
1.106 deraadt 223:
1.288 matthieu 224: <a name="watching"></a>
1.294 david 225: <li><h3><font color="#e00000">Watching our Changes</font></h3><p>
1.106 deraadt 226:
1.21 deraadt 227: Since we take a proactive stance with security, we are continually
228: finding and fixing new security problems. Not all of these problems
1.80 espie 229: get widely reported because (as stated earlier) many of them are not
1.45 deraadt 230: confirmed to be exploitable; many simple bugs we fix do turn out to
231: have security consequences we could not predict. We do not have the
232: time resources to make these changes available in the above format.<p>
1.21 deraadt 233:
234: Thus there are usually minor security fixes in the current source code
235: beyond the previous major OpenBSD release. We make a limited
1.45 deraadt 236: guarantee that these problems are of minimal impact and unproven
1.44 ian 237: exploitability. If we discover that a problem definitely matters for
1.45 deraadt 238: security, patches will show up here <strong>VERY</strong> quickly.<p>
1.21 deraadt 239:
1.45 deraadt 240: People who are really concerned with security can do a number of
241: things:<p>
1.21 deraadt 242:
243: <ul>
244: <li>If you understand security issues, watch our
1.294 david 245: <a href="mail.html">source-changes mailing list</a> and keep an
1.23 deraadt 246: eye out for things which appear security related. Since
1.21 deraadt 247: exploitability is not proven for many of the fixes we make,
248: do not expect the relevant commit message to say "SECURITY FIX!".
249: If a problem is proven and serious, a patch will be available
250: here very shortly after.
1.161 horacio 251: <li>In addition to source changes, you can watch our <a href="mail.html">
1.160 ericj 252: security-announce mailing list</a> which will notify you for every
1.186 ian 253: security related item that the OpenBSD team deems as a possible threat,
1.160 ericj 254: and instruct you on how to patch the problem.
1.21 deraadt 255: <li>Track our current source code tree, and teach yourself how to do a
1.29 deraadt 256: complete system build from time to time (read /usr/src/Makefile
257: carefully). Users can make the assumption that the current
258: source tree always has stronger security than the previous release.
1.45 deraadt 259: However, building your own system from source code is not trivial;
1.397 deraadt 260: it is nearly 850MB of source code, and problems do occur as we
1.45 deraadt 261: transition between major releases.
1.115 ericj 262: <li>Install a binary snapshot for your
1.80 espie 263: architecture, which are made available fairly often. For
1.413 matthieu 264: instance, an i386 snapshot is typically made available weekly.
1.21 deraadt 265: </ul>
266:
1.9 deraadt 267: <p>
1.288 matthieu 268: <a name="reporting"></a>
1.294 david 269: <li><h3><font color="#e00000">Reporting problems</font></h3><p>
1.3 deraadt 270:
1.5 deraadt 271: <p> If you find a new security problem, you can mail it to
1.294 david 272: <a href="mailto:deraadt@openbsd.org">deraadt@openbsd.org</a>.
1.7 deraadt 273: <br>
1.5 deraadt 274: If you wish to PGP encode it (but please only do so if privacy is very
1.112 philen 275: urgent, since it is inconvenient) use this <a href="advisories/pgpkey.txt">pgp key</a>.
1.5 deraadt 276:
1.107 deraadt 277: <p>
1.288 matthieu 278: <a name="papers"></a>
1.294 david 279: <li><h3><font color="#e00000">Further Reading</font></h3><p>
1.107 deraadt 280:
1.389 lum 281: Numerous
1.404 deraadt 282: <a href="papers/index.html">papers</a> have been written by OpenBSD team members,
1.389 lum 283: many dedicated to security.
1.294 david 284: </ul>
1.1 deraadt 285:
1.24 deraadt 286: </body>
287: </html>