Re: [EXIM] rejections and envelope sender logging

Page principale
Supprimer ce message
Répondre à ce message
Auteur: Chris Thompson
Date:  
À: exim-users
Sujet: Re: [EXIM] rejections and envelope sender logging
Peter Radcliffe <pir@???> writes:
>
> One of my users is on a mailing list one which someone is sending mail
> with the from line 'From: me <me@???>' which doesn't seem to be
> valid (here.com has no A or MX records) but verification fails:
>
> DNS lookup of here.com (MX) gave TRY_AGAIN
> here.com in dns_again_means_nonexist? no (end of list)
> lookuphost router deferred here.com
> message: host lookup did not complete
> search_tidyup called
> me@??? cannot be resolved at this time:
> host lookup did not complete


The registered nameservers for here.com are NS.PSI.NET and NS2.PSI.NET
(now that's already given you a nice warm feeling, hasn't it? :-]), but
these give SERVFAIL return codes when queried about anything in the zone,
or referals to themselves if the no-recursion flag is set. This is
typical of a "lame delegation", i.e. the registered nameservers don't
actually have any knowledge of the zone. This can be variously due to
user error, malicious delegation, or misguided "we'll create the delegation
first and the zone later" policies.

> Now, presumably I can put here.com in dns_again_means_nonexist and
> it'll bounce these right away rather than accepting/rejecting but
> is this the only way to deal with domains that are continually
> broken like this ?


It's quite difficult to automatically distinguish all possible reasons for
SERVFAIL errors: if one could, then the conditions wouldn't (well, perhaps)
have been clumped together under this return code in the first place.

> P ... off to mail the tech contact for here.com


Best of luck. I must say that "me@???" sounds like a spammer's address
if ever there was one...

Chris Thompson               Cambridge University Computing Service,
Email: cet1@???    New Museums Site, Cambridge CB2 3QG,
Phone: +44 1223 334715       United Kingdom.


--
*** Exim information can be found at http://www.exim.org/ ***