1 |
ok lets move past the what if. or could be. I have read most of the |
2 |
threads since the inception. Who is the moderator or top level person for |
3 |
security for gentoo. Have they posted or addressed this on this thread or |
4 |
in general? that part I was unable to find? |
5 |
|
6 |
I dont mean to simplify or over complicate the matter. But is there a |
7 |
legitamate solution or proposal on the table. I have read some great |
8 |
ideas, and some very dis-heartening ones as well. Like most of you I |
9 |
love gentoo, I love the concept behind it. So now lets move toward a |
10 |
solution. There will be flaws with any of these ideas, but there has to |
11 |
be a solution that will be a bit more secure and resolve some of the issue |
12 |
at hand. wether that is PGP or .jar or DSA keys. |
13 |
|
14 |
|
15 |
----- Original Message ----- |
16 |
From: "Andrew Jaquith" <ajaquith@×××××××.com> |
17 |
To: <gentoo-security@l.g.o> |
18 |
Cc: "Peter Simons" <simons@××××.to> |
19 |
Sent: Monday, November 08, 2004 8:19 AM |
20 |
Subject: *****SPAM***** Re: [gentoo-security] Re: Let's blow the whistle |
21 |
|
22 |
|
23 |
> I've lost track on where we are on this thread. |
24 |
> |
25 |
> That said -- there is a perfectly good standard for signing and |
26 |
> verifying files in large-scale collections of code. It's called a |
27 |
> signed JAR file. The 'jarsigner -verify my.jar' command works very |
28 |
> nicely. |
29 |
> |
30 |
> http://java.sun.com/j2se/1.3/docs/guide/jar/jar.html |
31 |
> |
32 |
> The file format itself is ZIP rather than tar or gzip. |
33 |
> |
34 |
> Peter -- I appreciate your efforts to raise the visibility of this |
35 |
> issue. |
36 |
> |
37 |
> Andrew Jaquith |
38 |
> Senior Project Manager |
39 |
> Symantec Professional Services |
40 |
> 196 Broadway |
41 |
> Cambridge, MA 02139 |
42 |
> USA |
43 |
> |
44 |
> Direct: 617.768.2711 |
45 |
> Mobile: 617.501.3278 |
46 |
> Fax: 617.621.1478 |
47 |
> Email: ajaquith@×××××××.com |
48 |
> PGP key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x898CF546 |
49 |
> On Nov 8, 2004, at 10:02 AM, Peter Simons wrote: |
50 |
> |
51 |
> > Brian G Peterson writes: |
52 |
> > |
53 |
> >> I assume that you intend to 'blow the whistle' because |
54 |
> >> you are incapable or unwilling to submit a patch for the |
55 |
> >> issue yourself? |
56 |
> > |
57 |
> > If you read the recent messages carefully, you'll find that |
58 |
> > I have tried _numerous_ times to provide details how to |
59 |
> > remedy the situation. |
60 |
> > |
61 |
> > |
62 |
> >> I agree that there is a lot of room for improvement in |
63 |
> >> the portage security system. |
64 |
> > |
65 |
> > Then why don't we stop discussing what I know or don't know, |
66 |
> > do or won't do, and talk about a solution? The vast majority |
67 |
> > of text posted in this thread is concerned with all kinds of |
68 |
> > things BUT finding a good, technical solution to a |
69 |
> > vulnerability that _does_ exist. |
70 |
> > |
71 |
> > Generating a signed hash list of all files is really not |
72 |
> > that difficult. It would solve the problem in a matter of |
73 |
> > hours for those who are concerned about it, and it would |
74 |
> > probably set things in motion for a better solution to be |
75 |
> > developed that solves the problem for all users as well as |
76 |
> > possible. |
77 |
> > |
78 |
> > So why is the Gentoo team so incredibly reluctant to do |
79 |
> > anything about it? |
80 |
> > |
81 |
> > Again: |
82 |
> > |
83 |
> > (1) Configure your main site to update the portage tree |
84 |
> > from CVS in a time interval that's sufficient large to |
85 |
> > allow for the hash list to be generated. Someone else |
86 |
> > already suggested once an hour. I can't say what is |
87 |
> > appropriate since I don't know your setup. |
88 |
> > |
89 |
> > (2) Calculate hashes for all files in the /usr/portage |
90 |
> > hierarchy. One could probably use a trivial Makefile to |
91 |
> > generate hashes incrementally, even, to ease the load |
92 |
> > on the machine. |
93 |
> > |
94 |
> > (3) Sign the hash file with a GPG key. That means that |
95 |
> > either someone has to enter the pass phrase manually, |
96 |
> > or you'll have to set up a pass phrase agent, or you'll |
97 |
> > have to use a key without a password at all. |
98 |
> > |
99 |
> > Everything but the first solution is sub-optimal but |
100 |
> > still a _lot_ better than what we have now. If someone |
101 |
> > manages to compromise the main site, we all have far |
102 |
> > greater problems than a lost secret key, so even _if_ |
103 |
> > the pass phrase is empty we still gain security. |
104 |
> > |
105 |
> > (4) Distribute the signed hash file with the portage tree. |
106 |
> > |
107 |
> > (5) Provide scripts that verify the integrity of the tree |
108 |
> > after an emerge sync _before_ any other code is run |
109 |
> > that has been obtained from the network. |
110 |
> > |
111 |
> > (6) Make the matching public key available on the key |
112 |
> > servers, on the web site, and every other place that |
113 |
> > you can think about. Give an expiry date of, say 3 |
114 |
> > months to make clear that this is an intermediate |
115 |
> > solution that will change. |
116 |
> > |
117 |
> > (7) Get as many people to sign the key as possible to |
118 |
> > properly authenticate it. |
119 |
> > |
120 |
> > (8) Write a security advisory that educates the users about |
121 |
> > the problem. |
122 |
> > |
123 |
> > Peter |
124 |
> > |
125 |
> > |
126 |
> > -- |
127 |
> > gentoo-security@g.o mailing list |
128 |
> > |
129 |
> |
130 |
> |
131 |
> -- |
132 |
> gentoo-security@g.o mailing list |
133 |
> |
134 |
> |
135 |
|
136 |
|
137 |
|
138 |
-- |
139 |
gentoo-security@g.o mailing list |