Re: [Evolution] attachments issue

[Patrick O'Callaghan <poc usb ve>]

On Fri, 2013-05-17 at 15:23 -0400, Adam Tauno Williams wrote:
> On Fri, 2013-05-17 at 12:06 -0430, Patrick O'Callaghan wrote:
> > On Fri, 2013-05-17 at 17:48 +0200, Thomas Prost wrote:
> > > Am Freitag, den 17.05.2013, 10:52 -0430 schrieb Patrick O'Callaghan: 
> > > > On Fri, 2013-05-17 at 12:15 +0200, Thomas Prost wrote:
> > > > > > pD9548C44.dip0.t-ipconnect.de (probably your own machine)
> > > > > > mrelayeu.kundenserver.de (your local mail server?)
> > > > > > moutng.kundenserver.de (some intermediate relay)
> > > > > That's all my carrier, whose policy I know ...
> > > > That's entirely fortuitous. There's no way in general that you can
> > > > predict what sequence of relays a message is going to go through.
>
> ?? This is just incorrect.  The relay sequence of a message is
> determined primarily by end-point relay hosts and MX records.  The
> policy within end-points will be consistent [otherwise that policy is
> *insane*] so end-point hosts [sites] can be thought of as a single
> multi-component apparatus.  So generally there is only my-end-point and
> the-other-end-point.   Communication between those is determined almost
> exclusively my MX records - which anyone can query, and change very
> rarely.

I guess we have different notions of what "predict" means. It's easy to
imagine a scenario where an intermediate relay fails after the message
has left the source host and an alternate is used. The alternate need
not have the same characteristics as the primary.

Anyway, we're getting OT. The point I was trying to make is that you
can't just assume that large messages are OK for your destination,
especially if it's one you haven't corresponded with before.

The other point I made in the previous thread on this is that people use
email for everything even when it's not the best tool. Then they wonder
why their message with an attached movie file doesn't work.

poc