1 |
W dniu pią, 06.07.2018 o godzinie 13∶34 +0200, użytkownik Ulrich Mueller |
2 |
napisał: |
3 |
> > > > > > On Fri, 6 Jul 2018, Marc Schiffbauer wrote: |
4 |
> > * Michał Górny schrieb am 06.07.18 um 11:33 Uhr: |
5 |
> > > If you don't see it for 5 years, how can you be sure that it is |
6 |
> > > even still there? |
7 |
> > Are you serious? Who tells you that I do not check from time to |
8 |
> > time? |
9 |
> > I am sure there will always be some scenario which makes a key |
10 |
> > unacessible in some way. I do not disagree with that. Its a matter |
11 |
> > of propability. |
12 |
> > And for the worst case there is a revoke-Certificate which can be |
13 |
> > used. |
14 |
> |
15 |
> Note that the revocation certificate is still listed under |
16 |
> recommendations only, so devs need not create one. Making this a |
17 |
> requirement would be a real improvement, IMHO. |
18 |
|
19 |
How are you going to enforce it? I didn't make it a requirement because |
20 |
we simply can't verify it being met. |
21 |
|
22 |
> Instead, the GLEP draft is focusing on short expiration times. |
23 |
> It won't help much if your compromised key will expire within one |
24 |
> year, but you cannot revoke it. |
25 |
|
26 |
You're conflating two unrelated concepts. Expiration is not meant to |
27 |
replace revocation, or in any way amend it. Expiration is meant to |
28 |
cover the case of both the key and the revocation certificate being |
29 |
destroyed or otherwise becoming inaccessible. |
30 |
|
31 |
> |
32 |
> Suggestions: |
33 |
> - Change the minimum requirement for key expiry to at most 3 years |
34 |
> (which is what in version 1 is recommended). |
35 |
> - Recommend at most 15 months of key expiry, to be renewed at least |
36 |
> 2 weeks before the expiry date. |
37 |
> - Make creation of a revocation certificate (and storing it in a place |
38 |
> separate from the key) mandatory. |
39 |
> |
40 |
> Ulrich |
41 |
|
42 |
-- |
43 |
Best regards, |
44 |
Michał Górny |