David,
On Tue, 4 May 2004, David Woodhouse wrote:
> On Tue, 2004-05-04 at 21:33 +0100, Tony Finch wrote:
> > Incorrect. After a sub-condition fails, subsequent parts of and{} are
> > parsed but not evaluated.
>
> Hmm. I could have sworn I did this...
You could well have done, if prior to 4.32 ...
>
> # Drop the connection if the reverse DNS isn't in the HELO domain
> drop log_message = HELO MISMATCH Forged HELO for ($sender_helo_name)
> message = You are not really $sender_helo_name. Go Away.
> - condition = ${if and { \
> - {!eq{$acl_m9}{}} \
> - {!match{$sender_host_name}{${rxquote:$acl_m9}\N$\N}} \
> - } \
> - {yes}{no}}
> + message = You are not really $sender_helo_name. Go Away.
> + condition = ${if !eq{$acl_m9}{} {1}}
> + condition = ${if !match{$sender_host_name}{${rxquote:$acl_m9}\N$\N} {1}}
> delay = 5s
>
>
> ... to avoid the evaluation of $sender_host_name, upon empirical
> evidence (it was bitching about 2002::/16 IPv6 addresses not having
> reverse DNS, even though I could have told it not to bother trying).
From ChangeLog-4.32:-
5. A reference to $sender_host_name in the part of a conditional expansion
that was being skipped was still causing a DNS lookup. This no longer
occurs
> dwmw2
HTH,
Richard
