Re: [jdev] Re: TLS and self-signed certs

2004-11-18 Thread Matthias Wimmer
Hi JD! JD Conley schrieb am 2004-11-12 09:18:46: Not sure ... there are valid reasons to change your s2s certificate: - Key expired - Key has been compromised - Key has been lost Well, if the cert changed you could then verify the key again with a dialback and reset the cache if

RE: [jdev] Re: TLS and self-signed certs

2004-11-18 Thread JD Conley
-Original Message- From: Matthias Wimmer [mailto:[EMAIL PROTECTED] Hi JD! JD Conley schrieb am 2004-11-12 09:18:46: Not sure ... there are valid reasons to change your s2s certificate: - Key expired - Key has been compromised - Key has been lost Well, if the

Re: [jdev] Re: TLS and self-signed certs

2004-11-18 Thread Jacek Konieczny
On Thu, Nov 18, 2004 at 09:33:05AM -0800, JD Conley wrote: If an attacker attempts to connect and provides a certificate that is not on record for the host they are claiming to be, a dialback is performed against the authority of the host. The attacker, unless they have control of DNS or the

Re: [jdev] Re: TLS and self-signed certs

2004-11-18 Thread David Waite
Nothing can be done without trust. We are using Verisign today as a trusted body for providing correct DNS records and references. -David Waite On Thu, 18 Nov 2004 05:14:02 -0800, Neil Stevens [EMAIL PROTECTED] wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thursday 18 November

Re: [jdev] Re: TLS and self-signed certs

2004-11-13 Thread maqi
On Fri, 12 Nov 2004, Peter Saint-Andre wrote: Well, I was thinking of this paragraph: While this is a good choice to establish coordinates links between servers, it does not work for servers, where the administrators did not setup the encryption by agreement. Just before that paragraph: I

Re: [jdev] Re: TLS and self-signed certs

2004-11-12 Thread Matthias Wimmer
Hi Peter! Peter Saint-Andre schrieb am 2004-11-11 17:53:30: Allowing self signed (or otherwise untrusted) certs with STARTTLS + EXTERNAL is opening yourself up for a serious security breach. Well, that's another story. But that claim on the URL I provided was that it is technically

Re: [jdev] Re: TLS and self-signed certs

2004-11-12 Thread Matthias Wimmer
Hi Justin! Justin Karneges schrieb am 2004-11-11 22:07:54: And now that I think about it, the whole use dialback for the first connection, SASL EXTERNAL for all after concept would be a good way to optimize s2s. Not sure ... there are valid reasons to change your s2s certificate: - Key

RE: [jdev] Re: TLS and self-signed certs

2004-11-12 Thread JD Conley
From: Matthias Wimmer [mailto:[EMAIL PROTECTED] Sent: Friday, November 12, 2004 5:07 AM Hi Justin! Justin Karneges schrieb am 2004-11-11 22:07:54: And now that I think about it, the whole use dialback for the first connection, SASL EXTERNAL for all after concept would be a good way to

Re: [jdev] Re: TLS and self-signed certs

2004-11-12 Thread David Waite
I doubt someone arguing that ssh is more widely used (either in terms of traffic or in terms of users) than SSL/TLS, or even than just the https usage of SSL/TLS, would be taken very seriously. If Jabber servers had started with S2S defaulting to SSL, then trusted issuers would be the only way to

Re: [jdev] Re: TLS and self-signed certs

2004-11-11 Thread Justin Karneges
On Thursday 11 November 2004 04:53 pm, Peter Saint-Andre wrote: In article [EMAIL PROTECTED], JD Conley [EMAIL PROTECTED] wrote: Allowing self signed (or otherwise untrusted) certs with STARTTLS + EXTERNAL is opening yourself up for a serious security breach. Well, that's another story.

Re: [jdev] Re: TLS and self-signed certs

2004-11-11 Thread David Waite
On Thu, 11 Nov 2004 19:54:49 -0800, Neil Stevens [EMAIL PROTECTED] wrote: Also, remember that different people have different threat models to address. Someone in the old hypothetical revolutionary conspiracy can't afford to depend on large institutional corporations to sign their