master/slave setup via VPN?

[Thorsten Schnebeck]

Hi,

(ok, just to remember in this part of the thread we are looking at the 
slave :-)

On Monday 21 July 2008 11:54:57 Alain Spineux wrote:
> On Sun, Jul 20, 2008 at 8:57 PM, Thorsten Schnebeck
> > and for some unknown reasons I get on the _slave_ for one of the mail
> > domain names a
> >
> >> : to=<info at maildomain3.com>, relay=none, delay=0.01, delays=0.01/0/0/0,
> >> dsn=5.4.6, status=bounced (mail for 10.10.10.2 loops back to myself)
>
> Did you set a transport for domain3 ?
> If slave share the same ldap data as the master,
> maybe it will not understand why you setup a transport to itself.
Yes, this is standard kolab 2.2 master/slave setup and the ldap of the slave 
is a replica of the master. 

> You should define a transport per mailbox, not for all domain3
> or customize your master.cf templates accordingly
if I use
/kolab/sbin/postmap -q "info at domain3.com" 
ldap:/kolab/etc/postfix/ldaptransport.cf
I get an empty string as "info at domain3.com" has its homeserver on the slave.
This should be ok (?) 

I do not understand what happens in this case when I use master as relaishost 
of the slave and if this setup is "bad". I know that it works as every 
incoming mail gets delivered into the right mailbox but what happens when 
there is an error in the delivering chain. Can this result in a bounce/loop?

I mean " (mail for 10.10.10.2 loops back to myself)" sounds again as this has 
something to do with my special VPN setup and a standard kolab master/slave 
setup working in "real" networks does not have this problem.

Bye

  Thorsten