On 8/9/05, Tony Finch <dot@???> wrote:
> On Mon, 8 Aug 2005, Cliff Pratt wrote:
> >
> > Obviously the way to solve this is to stop the application from
> > generating these invalid addresses, but I'd like to understand why the
> > difference.
>
> You didn't say what email software the machines are running. You didn't
> quote relevant lines from the logs on the relevant machines. You didn't
> examine the debugging output.
>
> Tony.
>
Yes, I did say what software (Debian package exim4 version 4.50-8) but
it was right at the front of the message. No, I didn't include the
relevant lines from the log, they were bog-standard ones which we've
all seen 100s of times. You are correct though, I should have included
them anyway. I'm aware of Eric Raymond's excellent guide on asking
questions (
http://www.catb.org/~esr/faqs/smart-questions.html) and I
regret that I haven't fully followed them.
I can't currently get at the machines in question, but if I don't find
anything by digging into the logs and manuals, I post the logs here.
I'd post debugging output but I didn't think that this was that sort
of a problem. I've just trying to get my head around why an attempt to
relay a message with an invalid address in the bcc that it appears to
fail *all* recipients. I suspect that the answer lies in my (lack of)
understanding of exim4 and smtp relaying via smarthost versus 'direct'
delivery via the Internet rather than anything deeper. If I knew a bit
more, I think that it would be obvious <grin>.
Cheers,
Cliff
