On 27/Oct/11 13:45, Matus UHLAR - fantomas wrote:
> On 15.08.11 15:02, Sam Varshavchik wrote:
>> There are several possibilities. If the STARTTLS command itself 
>> fails, it's an SMTP error no difference then any other SMTP error, 
>> and will be either permanent or transient depending on its numerical 
>> code. If the STARTTLS command succeeds, but TLS negotiation fails, 
>> its a permanent error. But in either case there is really no fallback 
>> path.
> 
> There would be a fallback path, if courier returned and reported tempfail 
> in such case: we could set up esmtproutes for such host that would 
> disable using starttls there, and the mail would get delivered.

Better yet, Courier could add the server to some "starttls-ignorant"
file containing a list of servers that improperly advertize TLS
support.  Keeping unsent messages in a queue until a 5xx reply or
timeout is just normal.

I think ESMTP_USE_STARTTLS could safely default to 1 if such a
mechanism were implemented.

>> For practical purposes TLS for SMTP is fundamentally broken. Many TLS 
>> servers simply use self-signed certs, making TLS fundamentally 
>> useless as means for effective encryption.
> 
> Many does not mean all - those who use certificates signed by trusted 
> authorities are safe. And we can still configure other certificates as 
> trusted.

We don't need to actually trust the certificate if all what is wanted
is to encrypt the channel for privacy.

-- 
































------------------------------------------------------------------------------
Get your Android app more play: Bring it to the BlackBerry PlayBook 
in minutes. BlackBerry App World™ now supports Android™ Apps 
for the BlackBerry® PlayBook™. Discover just how easy and simple 
it is! http://p.sf.net/sfu/android-dev2dev
_______________________________________________
courier-users mailing list
[email protected]
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users

Reply via email to