[pkg-fetchmail-maint] Bug#316446: Bouncing versus forwarding to postmaster

[Roger Lynn]

Roger Lynn wrote:
> In addition to generating a bounce, the original mail is forwarded to
> postmaster, as can be seen on the last couple of lines above. My
> understanding of the man page is that either a bounce should be
> generated *or* it should be forwarded to the postmaster address, but not
> both.

Having reread the man page yet again, I can confirm that this is a bug 
and a regression from the version of Fetchmail that shipped with Woody.

>From the man page:
"If fetchmail cannot match any mailserver usernames or localdomain
addresses, the mail will be bounced.  Normally  it will be bounced to 
the sender, but if `nobounce' is on it will go to the postmaster"

I can not see any circumstances under which both should happen.

Should I clone this as a separate bug?

The combination of forwarding undeliverable mail to the postmaster 
combined with sending bounces with non-null reverse paths means that 
when spam is sent from a fake from address to a non-existent user on my 
machine I now get two copies of the spam instead of none. The first 
comes from the original spam being forwarded to me, the second from the 
bounce Fetchmail generates itself bouncing which is then forwarded to me 
as well. If the far end was also running Fetchmail this would probably 
result in a mail loop. It's bad enough that I receive spam addressed to 
me, but to receive two copies of spam addressed to someone else is just 
ridiculous.

Thanks,

Roger