First off this is a repost of a post I made on LinuxQuestions.org.
I have a fresh build of the latest sendmail release on a Fedora Core 7 box.
I have had no problems, compiling, installing and putting everything in place. However, after running sendmail I've found that it isn't listening for any incoming smtp connections what so ever. I've kept my sendmail.mc file down to the bare minimum and it follows:
VERSIONID(`$Id: generic-linux.mc,v 8.1 1999/09/24 22:48:05 gshapiro Exp $')dnl
FEATURE(`virtusertable', `hash /etc/mail/virtusertable')dnl
The above worked to generate a cf file that got sendmail working perfectly on my previous RedHat 9 box. I have rebuilt and reinstalled the .cf files at least a dozen times with various minor changes, restarting sendmail every time.
I have tried using various recommendations of the DAEMON_OPTIONS definition none of which have made any difference.
The reason I know sendmail isn't listening is that when I run:
tcp 0 0 *:mysql *:* LISTEN
tcp 0 0 *:ftp *:* LISTEN
tcp 0 0 localhost.localdom:postgres *:* LISTEN
tcp 0 0 *:pop3 *:* LISTEN
tcp 0 0 *:http *:* LISTEN
tcp 0 0 *:ssh *:* LISTEN
tcp 0 0 *:telnet *:* LISTEN
unix 2 [ ACC ] STREAM LISTENING 6649 /tmp/.s.PGSQL.5432
unix 2 [ ACC ] STREAM LISTENING 6557 /var/lib/mysql/mysql.sock
unix 2 [ ACC ] STREAM LISTENING 6264 /var/run/dbus/system_bus_socket
unix 2 [ ACC ] STREAM LISTENING 6936 @/var/run/hald/dbus-fi0qSqDJgZ
unix 2 [ ACC ] STREAM LISTENING 6939 @/var/run/hald/dbus-Q4EGBjjoWD
The sendmail processes running when I restart sendmail are:
3510 ? Ss 0:00 sendmail: Queue runner@01:00:00 for /var/spool/mqueue
3517 ? Ss 0:00 sendmail: Queue runner@01:00:00 for /var/spool/clientmqueue
I have also done the following:
Telneting to port 25 from the machine itself gives:
[root@server cf]# telnet server.domain.com 25
telnet: connect to address [ipaddress]: Connection refused
where [ipaddress] is the address of the external ethernet connection (i.e direct connection to the internet).
Running lsof -i :25 yields no result
Inspecting the mail log immediately after starting sendmail gives:
Sep 19 11:05:25 server sendmail: alias database /etc/mail/aliases rebuilt by simon
Sep 19 11:05:25 server sendmail: /etc/mail/aliases: 1 aliases, longest 5 bytes, 10 bytes total
Sep 19 11:05:25 obiwan sendmail: starting daemon (8.14.1): queueing@01:00:00
Sep 19 11:05:25 server sm-msp-queue: starting daemon (8.14.1): queueing@01:00:00
Sep 19 11:05:25 server sm-msp-queue: l8I6tQgb026691: to=root, delay=18:09:59, xdelay=00:00:00, mailer=relay, pri=211353, relay=[127.0.0.1] [127.0.0.1], dsn=4.0.0, stat=Deferred: Connection refused by [127.0.0.1]
My latest efforts to solve the problem follow:
First, I tried building an older version of sendmail on my FC7 box. This version of sendmail is one I had successfully compiled and run on a Redhat 9 box. This yielded the same results as above.
Second, I scratch built a second FC7 box with an older machine. With this I kept the absolute most basic generic linux configs that come with the sendmail source. This also yielded the same result as above.
Third, I used the sendmail 8.14.1 source to build sendmail on an FC5 box, both with and without Berkeley db. Both builds worked first time without a problem.
None of the above has brought me any closer to solving my problem, other than to draw the conclusion that there is something to do with Fedora Core 7 that I have missed.
Thankyou for your help in advance.