Brian Spraker wrote:
> When I run the "exim -bh" option and run through the procedure, I see where 
> Exim 
> checks the ACL that I made to check for whitelisting of the reverse DNS.
> 
> Exim doesn't believe that it is in there.

*snip* (prior discussion)


A) w/r that test - first swap the position in the list, and/or add a couple of 
real or dummy entries and see what happens.

B) IF the acl clauses differ ONLY in an SQL call vs a flat-file lsearch (or 
whatever), first thing I would do is convert 'deny' or 'accept' to 'warn' 
verbs, 
  put BOTH claues into active use, and add copious 'logwrite' and 'log_message' 
so as to test BOTH mechanisms at the same time on the same traffic.

If that has to be done on a production server, just put all the 'warn' stuff 
ahead of whatever sort-of-works until you get sight of where the problem lies, 
then slectively comment-OFF those 'instrumentation' clauses and apply what has 
been learned to production clauses.

You may be chasing a MySQL vagary or a stored-record anomaly that has nothing 
to 
do with Exim itself.

I've now and then found a trailing space or even a non-printing character 
sucked 
into a record, for example... (as I run PostgreSQL in UTF...) cured it only by 
deleting and re-entering the record.


'page two..'

JFWIW - I may have one of the most insanely PostgreSQL-driven Exim configs in 
production use.

BUT .... ALL of my *many* 'local' whitelist/blacklist tests are done with 
lsearch, wildlsearch, dsearch, and the like - none with SQL calls.

I can no longer remember WHY that is, but it JFW, and I am sure I had a good 
reason at the time... (Exim 4.4X onward...)

;-)

HTH,

Bill





-- 
## List details at http://lists.exim.org/mailman/listinfo/exim-users 
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/

Reply via email to