On Wed, 13 Oct 2004, Christiaan den Besten wrote:
> Ohw, I feel lame ..... 'user error' :). Well, somewhat ...
>
> I use Mailscanner for mail filtering, so the 'input' exim daemon has
> 'queue only' set in the configuration. But exim requires to have a
> 'valid' router and transport anyway ....
...
> It should simply accepts all mail (yes, should have used an 'accept' router,
> juist build that one :) ) and should never realy be used because the daemon
> should just (and does) queue the message.
...
> Is there any particular reason for exim to check for 'valid'
> transports/routers in a queue_only mode?
I do much the same. But you do want some valid routers: because Exim
wants to run through these when doing verification steps if you are
requiring them (and you should have at least some). For example, you
probably don't want to accept mail from a source that, by verification,
you find you can't actually return mail to. You'll presumably also want
to run verification for your internal users, if they aren't local to that
box you'll probably need some routers to go through that verification
step; likewise if you are relaying for other hosts.
So, set up enough routers to meet your verification needs, and set:
verify_only
on them.
You may need some (dummy?) transports too. If you are doing callout
verification, you'll probably need a remote_smtp one at least.
Jethro.
>
> ----- Original Message ----- From: "Philip Hazel" <ph10@???>
> To: "Christiaan den Besten" <chris@???>
> Cc: <exim-users@???>
> Sent: Wednesday, October 13, 2004 4:14 PM
> Subject: Re: [exim] "retry time not reached for any host after a
> longfailureperiod" (2)
>
>
> > On Wed, 13 Oct 2004, Christiaan den Besten wrote:
> >
> > > What usually happend:
> > > - I send a message to a recipient sending me a message.
> > > - It bounces right back to me.
> > > - I send another messages to the same recipient within the same minute.
> > > No
> > > problems, delivered promptly without errors.
> >
> > On the face of it, that is completely weird. I can think of ways this
> > could happen, but they are unusual. For example: If the recipient is
> > remote, and Exim has to use the DNS, and the name servers for the domain
> > have got out of step. One time one server might yield one IP address and
> > the next time a different server yields a different address. So, for a
> > first check: Are the IP addresses that Exim uses in both cases the same?
> > They should be listed in the log lines.
> >
> > > Can I somehow send some debug info into syslog explaining me what info
> > > exim is
> > > retrieving from is retry database or something?
> >
> > You can't get info into syslog, but you can run Exim with debugging
> > turned on to do a single delivery. For example
> >
> > exim -d someone@???
> > Your message here
> > .
> >
> > You will get a lot of debugging output send to the standard error.
> >
> > --
> > Philip Hazel University of Cambridge Computing Service,
> > ph10@??? Cambridge, England. Phone: +44 1223 334714.
> > Get the Exim 4 book: http://www.uit.co.uk/exim-book
> >
> >
>
>
> --
> ## List details at http://www.exim.org/mailman/listinfo/exim-users Exim
> details at http://www.exim.org/ ##
>
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Jethro R Binks
Computing Officer, IT Services
University Of Strathclyde, Glasgow, UK
