On Fri, 2011-08-19 at 22:14 -0700, Paul Allen Newell wrote: > On 8/18/2011 9:49 PM, Paul Allen Newell wrote: > > > > > > Your earlier email indicates that to solve the mail/mailx between > > machines, I need comment out that line in sendmail.mc and I will get > > the results you indicated. Sure worth a try, let me give it a go. > > > Craig: > > I comment out the 127.0.0.1 line in sendmail.mc as suggested, did a > service sendmail restart > > I added a line in access for "Connect:192.168.2 RELAY" to match the > format of the existing entries. All the other entries were Relay and so > were all the online docs I could find. > > I did this on two machines (sender and receiver) > > I did a mail from one machine to another and got a "No route to host". > > I then did a service iptables stop and tried again ... got a "Connection > refused" > > Netstat -an | grep 25 gives me the same output as your email shows. > Iptables -L also gives me the same output. ---- If it looks something like this on your SMTP server # netstat -an Active Internet connections (servers and established) Proto R-Q S-Q Local Address Foreign Address State tcp 0 0 0.0.0.0:25 0.0.0.0:* LISTEN then indeed it is listening on all interfaces (0.0.0.0 port 25) for connections from all addresses any originating port (0.0.0.0:*) which is what we're after. If you get connection refused, there is probably only a few possibilities... 1 - /etc/hosts.deny - is there something in that file that would cause your system to reject connections? 2 - your network hardware... I have seen some 2-wire DSL modems that ridiculously block network traffic between 2 computers. This wouldn't seem likely because the error message changed when you shut down iptables service but after having been bitten by that nonsense, I always try to keep in mind. It's possible you have some access control on your network switch but that generally only possible with 'managed switches' or just a defective switch. 3 - your network configurations - All computers need to have unique IP addresses (192.168.2.10, 192.168.2.11 and so forth with the only difference being the last number in the octet). They must have the same netmask - the typical netmask for a 192.168.x.x network would be a Class C - 255.255.255.0 which allows communication from 192.168.2.0 through 192.168.2.255, the former being known as the 'network address' and the latter known as the 'broadcast address'. No computers should be assigned to either the network address or the broadcast address. Baby steps... start on the computer that is the SMTP server. telnet 127.0.0.1 25 # does it connect? telnet 192.168.2.10 25 # or whatever address is listed for 'eth0' - does it connect? After you can connect to port 25 on both the lo and eth0 addresses, you can move to another machine and see if you can telnet to the same 192.168.2.x IP address used above. It should work or it's almost certain that the problem is caused by 1 of the 3 things I listed above. Craig -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- users mailing list users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
