Re: [Nmh-workers] TLS support for POP merged to master

[Ken Hornstein]

>> inc(1) and msgchk(1)).
>
>You missed msgchk, as far as I can tell.

Whoops, yeah, you're right.  Sorry about that!

>inc: TLS peer aborted connection

Damn.  I am ... a bit surprised.  Our use of TLS is relatively simple;
we don't, for example, use the non-blocking interface (which requires you
to check return values and then call the the appropriate function if
TLS needs to do a read or write underneath).  So I don't _THINK_ that's
the problem.  But I could be wrong.

If I had to guess ... I would think that maybe what's going wrong is
that we might be getting the buffering wrong somewhere, and our next
command is garbled and is causing the abort.  I guess I'll have to dig
into it some more.

I don't suppose you could quantify this a bit more, could you?  Like does
it always happen on message 5, or did it happen after a large message,
or something else?  Really, any clues here would be helpful.

I don't suppose you have access to the POP server logs, do you?  If we
had a remote error message that might be helpful.  I understand it might
not be possible if it's, say gmail :-/

--Ken

--Ken