0

I'm new to this so please go easy if I don't explain it lugubriously.

I'm admin'ing a debian based KVM/LVM server. It has several VM's on it but two that are the problem. Recently I had to clone VM A so I could test some things but when I restarted it wouldn't work so I had to manually reset the network on the host. I believe this is what cause my postfix smtp to stop sending on A.

The set up for postfix is it sends the message from A to VM B, our dev box, which then sends it out, I don't know much about that, still learning!!!

The problem is it has stopped sending emails back over the last day or two and checking the log it says the connection is timing out.

The log says:

connect to smtp.ourdomain.com[78.XX.YYY.ZZZ]:25: connection timed out

Before the network restart it worked fine and returned:

relay=smtp.ourdomain.com[192.XXX.YYY.Z]:25

The IP beginning with 192 is the one to our dev box which was the original setup and the IP to 78 is actually to our host machine.

The problem is I don't know how to go about changing this. I tried looking around inside main.cf for postfix but couldn't find anything, and I didn't set the system up so I don't know much about it. Thanks in advance.

boundless08
  • 13
  • 7
  • Can you connect to the 78.x.x.x server via telnet from the machine with postfix installed? It may just be a connectivity issue. – NickW Feb 06 '13 at 16:56
  • No it's not a connectivity issue, I can ping both machines from VM A, it's just changed the IP address somehow. – boundless08 Feb 06 '13 at 16:58
  • The relay entry comes from your transport map, which can be a file, or a table in a database. Have a look at your main.cf, there will be an entry telling you what type of transport map you are using, and have a look how it is entered in there. If you didn't change it, you may be obtaining the address via DNS, and the DNS entry for smtp.ourdomain.com has changed.. – NickW Feb 06 '13 at 17:00
  • I read something about transport maps but I can't find anything relating to it in main.cf. I found `relayhost = smtp.ourdomain.com` if that means anything – boundless08 Feb 06 '13 at 17:24
  • Yeah, it's hard coded, which is fine. The problem is that your DNS server (/etc/resolv.conf to find out who it is) is returning the different address. – NickW Feb 06 '13 at 17:39
  • okay `resolv.conf` contains `nameserver 192.XXX.YYY.Z`. I'm sorry about this, I don't have a clue how to fix this. – boundless08 Feb 06 '13 at 17:43
  • Do you know anything about DNS servers? If you don't you need to talk to your local admin and get him to make sure that the server is responding correctly to your requests. That or just place the IP address of smtp.ourdomain.com in the relayhost= for now. – NickW Feb 06 '13 at 17:45
  • let us [continue this discussion in chat](http://chat.stackexchange.com/rooms/7424/discussion-between-nickw-and-boundless08) – NickW Feb 07 '13 at 09:48

1 Answers1

0

Your problem is DNS related, the DNS server is responding with an incorrect address, external instead of internal, to your DNS request.

The problem isn't postfix but whatever software you have resolving domains.

My recommendation is to change the line

relayhost =

and replace the server name with the IP address in brackets [192.168.1.10] for example...

NickW
  • 10,263
  • 1
  • 20
  • 27