Almost there, I wasn't quite clear in my last post, the cname should be the fqdn of your host. As I mentioned in my last post I believe your server FQDN is lcrm1.lcrm.co.uk - you should put that in.
So, is this the solution::
Code:@ IN SOA lcrm.co.uk. support.lcrm.co.uk. ( 2006091301; serial 28800; refresh, seconds 7200; retry, seconds 604800; expire, seconds 86400 ); minimum, seconds ; NS ns.lcrm.co.uk.; NS ns2.lcrm.co.uk.; ; MX 10 lcrm.co.uk.; ; * A 188.8.131.52 lcrm1 A 184.108.40.206 www A 220.127.116.11 ns A 18.104.22.168 ns2 A 22.214.171.124 webmail A 126.96.36.199 mail CNAME lcrm1.lcrm.co.uk
Right, let's do it this way.
That, with a bit of luck and a following wind, should be it.Code:@ IN SOA lcrm.co.uk. support.lcrm1.lcrm.co.uk. ( 2005110200 10800 3600 604800 38400 ) @ IN NS ns.lcrm.co.uk. @ IN NS ns2.lcrm.co.uk. @ IN MX 10 lcrm1.lcrm.co.uk. @ IN A 188.8.131.52 lcrm1.lcrm.co.uk. IN A 184.108.40.206 mail1 IN CNAME lcrm1.lcrm.co.uk. mail2 IN CNAME lcrm1.lcrm.co.uk. * A 220.127.116.11 www A 18.104.22.168 ns A 22.214.171.124 ns2 A 126.96.36.199 webmail A 188.8.131.52
Thanks very much Bill,
I have changed the zone file as to your recommendation and will let yo know in a couple of days if all has cleared up.
Any chance I can become part of the team Bill?
Sure there is something I can help out with. I did send my C.V. a few times but heard no reply.
Mail is currently not being delivered outside the network.
I have two MTA's with one on the edge and one on the main store inside the firewall.
I had hoped to use a single MTA on the edge. That gave me the "localhost loops back to myself" error.
I know that this is related to postfix, I know that the issue is also based upon DNS configuration. Can someone assist me with the next best step in troubleshooting?
I would like to help, can you provide some more info please?
like are the two servers in the same domain?
what are the dns servers for the both of them?
are they acting independently or as a cluster?
is the edge a proxy server or an actual email server \
is the internal server used as a backup or do you have one internal for local mail and the edge for external mail?
in which case does the internal receive mail and then send it to the edge to send out to the internet?
rather confused myself there -- can you enlighten the situation?
Both servers resolve to one internal DNS server. An external dns handles our mx to the edge.
They are both in the same domain.
The two servers are not clustered.
The edge server is a zimbra MTA with an external IP
I don't see that the internal actually delivers mail to the edge for external delivery. I think that it should make the communication itself. I tried (grasped at the possibility of a difference) setting the MTA relay on the internal to the external MTA, but that seemed ridiculous and the result was no different. My traffic is stuck at the internal server with the familiar "localhost mail loops back to myself".
On the internal server with the mail store (no external ip):Code:; ; <<>> DiG 9.2.4 <<>> helpwithzimbra.com mx ;; global options: printcmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 48042 ;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 1, ADDITIONAL: 1 ;; QUESTION SECTION: ;helpwithzimbra.com. IN MX ;; ANSWER SECTION: helpwithzimbra.com. 86400 IN MX 10 office.helpwithzimbra.com. ;; AUTHORITY SECTION: helpwithzimbra.com. 86400 IN NS ns1.helpwithzimbra.com. ;; ADDITIONAL SECTION: office.helpwithzimbra.com. 86400 IN A xxx.xxx.xxx.234 ;; Query time: 1 msec ;; SERVER: 184.108.40.206#53(220.127.116.11) ;; WHEN: Tue Feb 6 12:39:44 2007 ;; MSG SIZE rcvd: 97 [zimbra@office ]$
And from another server, external to the network:Code:; <<>> DiG 9.2.4 <<>> helpwithzimbra.com mx ;; global options: printcmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 35569 ;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 1, ADDITIONAL: 1 ;; QUESTION SECTION: ;helpwithzimbra.com. IN MX ;; ANSWER SECTION: helpwithzimbra.com. 86400 IN MX 10 office.helpwithzimbra.com. ;; AUTHORITY SECTION: helpwithzimbra.com. 86400 IN NS ns1.helpwithzimbra.com. ;; ADDITIONAL SECTION: office.helpwithzimbra.com. 86400 IN A xxx.xxx.xxx.234 ;; Query time: 0 msec ;; SERVER: 18.104.22.168#53(22.214.171.124) ;; WHEN: Tue Feb 6 12:45:40 2007 ;; MSG SIZE rcvd: 97 [zimbra@zimbra ~]$
I see the A record for all but the MTA on the internal server. That one does not have an mx. But it is only for outgoing mail.Code:; <<>> DiG 9.2.4 <<>> helpwithzimbra.com mx ;; global options: printcmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 37115 ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 4, ADDITIONAL: 5 ;; QUESTION SECTION: ;helpwithzimbra.com. IN MX ;; ANSWER SECTION: helpwithzimbra.com. 1187 IN MX 10 office.helpwithzimbra.com. ;; AUTHORITY SECTION: helpwithzimbra.com. 1187 IN NS remote1.easydns.com. helpwithzimbra.com. 1187 IN NS remote2.easydns.com. helpwithzimbra.com. 1187 IN NS ns1.easydns.com. helpwithzimbra.com. 1187 IN NS ns2.easydns.com. ;; ADDITIONAL SECTION: office.helpwithzimbra.com. 1187 IN A xxx.xxx.xxx.234 ns1.easydns.com. 379 IN A 126.96.36.199 ns2.easydns.com. 79698 IN A 188.8.131.52 remote1.easydns.com. 95421 IN A 184.108.40.206 remote2.easydns.com. 52301 IN A 220.127.116.11 ;; Query time: 17 msec ;; SERVER: 18.104.22.168#53(22.214.171.124) ;; WHEN: Tue Feb 6 13:02:44 2007 ;; MSG SIZE rcvd: 231 [root@www postgrey]#
Last edited by davdunc; 02-06-2007 at 02:21 PM. Reason: fix a little grammer
Yay! I figured it out.
Here is the issue - The global settings for the relay server and MTA were set to "localhost"
I configured two mail servers one for the mail store and one for the edge.
The server designated as the mail store had the inherited the MTA global settings as did the server designated for edge MTA. So both servers inherited localhost for web MTA and relay host.
On top of the local delivery this caused the real culprit! The reason I was getting the "localhost mail loops back to myself" was related to the setting for the relay host. The global setting for MTA relay was set to localhost. As a result, my original settings on both hosts was set to localhost. postconf relayhost showed localhost:25 as the relay for each one. That looped the outgoing mail to the localhost.
I removed the settting for the relayhost from my edge MTA configuration in the Admin tool. The administration console froze completely. I had to restart zimbra on the mail store to get the admin console to unfreeze. Long story short, I couldn't get the setting to complete in the server configuration. Since it wasn't working on the server, I thought that I should just try to remove the setting in the Global Configuration. That worked!.
I then moved back to the server configuration and pressed the "Reset to Global Settings" button and saved the settings.
Mail began to flow without issue.
During the troubleshooting, I reinstalled the edge MTA services and broke the logger. That will be my next problem to tackle. No statistics are currently visible in the admin console, try as I might to fix it, but fix it I will. Thanks for the help!
Last edited by davdunc; 02-09-2007 at 01:23 PM. Reason: Clarify the point.
i have the same problem too.
the zimbra postfix (runs on local port 2525) delivers his mail to a relay host (an exim4 on Port 25).
When i send via Webgui, everything is ok...if i send via Zimbra SMTP (on Port 2525), they said " status=deferred (mail for localhost loops back to myself"
What is wrong?
MTA Relay Host enabled = true
DNS lookups enabled = false
There are currently 1 users browsing this thread. (0 members and 1 guests)