OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
Postfix Archives: selecting MX record

selecting MX record


Subject: selecting MX record
From: Geoff Gibbs (ggibbshgmp.mrc.ac.uk)
Date: Tue Jan 04 2000 - 08:50:28 CST


I have a problem where Postfix appears to be behaving differently from
sendmail, in that it appears to select a different MX record to use.

I think Postfix is behaving correctly, but sendmail is actually delivering
the mail !

On our site, sendmail is still handling ordinary e-mail, while Postfix
is being used to send out to our mailing lists. In syslog I can see
two entries to the same address, one succeeds and one fails :-

Jan 4 14:04:20 mercury sendmail[25848]: OAA25843: to=FB7A070nw01.rrz.uni-hamburg.de.filtered, ctladdr=ggibbshgmp.mrc.ac.uk (5042/5000), delay=00:02:48,
xdelay=00:02:48, mailer=esmtp, relay=rzaixsrv2.rrz.uni-hamburg.de. [134.100.32.12], stat=Sent (PAA40118 Message accepted for delivery)
Jan 4 14:04:21 mercury postfix/smtp[23478]: BE78717A63: to=<FB7A070nw01.rrz.uni-hamburg.de>, relay=nw01.rrz.uni-hamburg.de[134.100.1.50], delay=124,
status=bounced (host nw01.rrz.uni-hamburg.de[134.100.1.50] refused to talk to me: 520 Connection not authorised from this address.)

The difference that I can see is the different "relay=" entry. So I looked
at the DNS entries :-

Non-authoritative answer:
nw01.rrz.uni-hamburg.de preference = 10, mail exchanger = rzdspc1.informatik.uni-hamburg.de
nw01.rrz.uni-hamburg.de preference = 5, mail exchanger = rzaixsrv2.rrz.uni-hamburg.de
nw01.rrz.uni-hamburg.de preference = 0, mail exchanger = nw01.rrz.uni-hamburg.de
nw01.rrz.uni-hamburg.de preference = 100, mail exchanger = ws-ham1.win-ip.dfn.de

It appears that Postfix is using the "0" preference, while sendmail is
using the "5" preference. My current problem is that the "0" preference
does not appear to want to talk to us, but I don't think I can blame
Postfix for that ;-) Although if anyone can solve that I would be most
grateful.

(The ".filtered" part of the address is because sendmail pipes messages
through procmail)

All the best

Geoff Gibbs

UK-Human Genome Mapping Project-Resource Centre,
Hinxton, Cambridge, CB10 1SB, UK
Tel: +44 1223 494530 Fax: +44 1223 494512 E-mail: G.Gibbshgmp.mrc.ac.uk



This archive was generated by hypermail 2b27 : Tue Jan 04 2000 - 09:01:17 CST