RE: [Exim] lookuphost, MX and A records discrepancy ?

Startseite
Nachricht löschen
Nachricht beantworten
Autor: Bernard Stern
Datum:  
To: Philip Hazel
CC: exim-users
Betreff: RE: [Exim] lookuphost, MX and A records discrepancy ?
On Thu, 20 Jul 2000 12:55:14 +0100 (BST) Philip Hazel wrote:

> On Thu, 20 Jul 2000, Bernard Stern wrote:


>> >> news_bounce:
>> >> driver = lookuphost
>> >> domains = *
>> >> local_parts = news
>> >> condition = ${if and {{> {$message_age}{43200}}{eq {$sender_address}{""}}} {yes}{no}}
>> >> # 43200 s = 12 h
>> >> transport = trash


>> Yes, I think I'm getting the point. I have looked at some more logs
>> and the error is either "connection refused" or "connection timed
>> out". So Exim doesn't get any SMTP return code and is bound to
>> try again.


> But that should then eventually hit that router and be routed
> differently. I was previously guessing that the actual routing was
> unable to complete. Can you try a -d9 delivery on a message that is in
> this state, and older than 43200?


> Wait! You have a bug in your condition that I didn't spot before. You
> shouldn't test $sender_address for {""} but instead you should test it
> for {} because the {} are the delimiters, not "". The condition you have
> will never work.


Shame! Yes, you are absolutely right, this is exactly the point.
Many stuck NDRs magically disapparated to trash!

This brings another little problem: when this router (news_bounce)
is triggered by a normal queue run, exim is perfectly happy. Now
if I force delivery (using the eximon), exim is no longer happy
and barfs via the panic log (3 lines here for legibility):

2000-07-20 14:01:26 13E0z8-00006p-00 failed to expand condition
"${if and {{> {$message_age}{43200}}{eq {$sender_address}{}}} {yes}{no}}"
for news_bounce router: unknown variable name "message_age"

Could it be that this variable is not available for such forced
deliveries? The spec document about this variable doesn't make
me feel that it is only available at some specific moments. Or
maybe the text "at the start of a delivery attempt" refers
to a "normal (unforced) delivery attempt"? If this is the case,
the spec should reflect it, IMHO.

Many thanks Philip for your always valuable help.

Regards,

Bernard Stern, SWITCH

____________S_W_I_T_CH___Swiss Academic_______________________________________
        mail: SWITCH Head Office       a          Tel: +41 1 268 1520
              Limmatquai 138           n          Fax: +41 1 268 1568
              CH-8001 Zurich           d        e-mail: stern@???
________________________________________Reseach Network_______________________