metadata/glsa/glsa-201612-01.xml


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE glsa SYSTEM "http://www.gentoo.org/dtd/glsa.dtd">
<glsa id="201612-01">
  <title>GnuPG: RNG output is predictable</title>
  <synopsis>Due to a design flaw, the output of GnuPG's Random Number Generator
    (RNG) is predictable.
  </synopsis>
  <product type="ebuild">gnupg</product>
  <announced>December 02, 2016</announced>
  <revised>December 02, 2016: 1</revised>
  <bug>591536</bug>
  <access>local</access>
  <affected>
    <package name="app-crypt/gnupg" auto="yes" arch="*">
      <unaffected range="ge">1.4.21</unaffected>
      <vulnerable range="lt">1.4.21</vulnerable>
    </package>
  </affected>
  <background>
    <p>The GNU Privacy Guard, GnuPG, is a free replacement for the PGP suite of
      cryptographic software.
    </p>
  </background>
  <description>
    <p>A long standing bug (since 1998) in Libgcrypt (see “GLSA 201610-04”
      below) and GnuPG allows an attacker to predict the output from the
      standard RNG. Please review the “Entropy Loss and Output Predictability
      in the Libgcrypt PRNG” paper below for a deep technical analysis.
    </p>
  </description>
  <impact type="normal">
    <p>An attacker who obtains 580 bytes of the random number from the standard
      RNG can trivially predict the next 20 bytes of output.
    </p>
    
    <p>This flaw does not affect the default generation of keys, because
      running gpg for key creation creates at most 2 keys from the pool. For a
      single 4096 bit RSA key, 512 bytes of random are required and thus for
      the second key (encryption subkey), 20 bytes could be predicted from the
      the first key.
    </p>
    
    <p>However, the security of an OpenPGP key depends on the primary key
      (which was generated first) and thus the 20 predictable bytes should not
      be a problem.  For the default key length of 2048 bit nothing will be
      predictable.
    </p>
  </impact>
  <workaround>
    <p>There is no known workaround at this time.</p>
  </workaround>
  <resolution>
    <p>All GnuPG 1 users should upgrade to the latest version:</p>
    
    <code>
      # emerge --sync
      # emerge --ask --oneshot --verbose "&gt;=app-crypt/gnupg-1.4.21"
    </code>
    
  </resolution>
  <references>
    <uri link="http://nvd.nist.gov/nvd.cfm?cvename=CVE-2016-6313">CVE-2016-6313</uri>
    <uri link="http://formal.iti.kit.edu/~klebanov/pubs/libgcrypt-cve-2016-6313.pdf">
      Entropy Loss and Output Predictability in the Libgcrypt PRNG
    </uri>
    <uri link="http://security.gentoo.org/glsa/201610-04">GLSA 201610-04</uri>
  </references>
  <metadata tag="requester" timestamp="Wed, 30 Nov 2016 18:28:25 +0000">whissi</metadata>
  <metadata tag="submitter" timestamp="Fri, 02 Dec 2016 09:38:37 +0000">whissi</metadata>
</glsa>