Scalix Forums

Hi,

OS: SuSE OSS 10.1
Scalix: 11.0.2

A client of mine has changed ADSL supplier recently. I have changed the linux.mc file and the client-info auth file and run omsendin. I also restarted sendmail again for good measure!

According to the mail logs sendmail is still using the previous smarthost.

I have also rebooted the server just in case and tried using the 'DSesmtp:[<IP>]' syntax in the .cf file to no avail...

Help, as I am now looking at a re-install!

Thanks, Graham

I would sugest using the DNS name. Put

Dsmtp.domain.com

in the correct place in the .cf, restart sendmail and see if it works. If it does, your definition in the .mc must have been unrecognised.

I thought it was 'DSsmtp.domain.com' but I have tried both and restarted sendmail (/etc/sendmail.cf) no joy.

If I stop the sendmail service and then try and send an email using SWA it appears as if sendmail is still running - sending fails with old smtp in logs as before...

ps aux | grep sendmail - shows nothing.
killall -9 sendmail - shows nothing

Yet it still appears to be running and trying to start it from the command line '/usr/bin/sendmail -bD' I get the socket in use error.

As a note the client recently moved the server (read unplugged it - yes I told him not to) and I don't know if that could have anything to do with it.

Sorry, that was a typo, should have been DS. Does lsof -i:25 output anything regarding sendmail? Also, what is the DAEMON_PORT_OPTIONS in your mc set to?

Mail can be sent even when sendmail is not running, Scalix launches sendmail -bs for outgoing mail and it will at least end up in the queues, even if it's not properly sent.

lsof -i:25

sendmail 9178 root 3u IPv4 622355 TCP localhost:smtp (LISTEN)
omsmtpd 32294 root 5u IPv4 368356 TCP <FQDN of server>:smtp (LISTEN)
omsmtpd 32294 root 7u IPv4 368357 TCP 10.8.0.1:smtp (LISTEN)

DAEMON_PORT_OPTIONS

Pattern not found in MC or CF files.

Scalix launches sendmail -bs for outgoing mail

Ahh, that had crossed my mind!

I suppose this is a sign not to answer questions while on the phone

Sendmail is running and listening to the correct port, according to lsof. And if DS is set correctly, it should go to the correct server after a restart.

Could it possibly be DNS related? Does the hostname resolve correctly from the Scalix server?

They have moved from Pipex to Entanet:

I have changed the nameservers in YaST.

PING smtp.enta.net (62.249.192.219) 56(84) bytes of data.
64 bytes from smtp1.enta.net (62.249.192.219): icmp_seq=1 ttl=57 time=32.4 ms

----------------------------------------
nslookup smtp.enta.net

Server: 127.0.0.1
Address: 127.0.0.1#53

Non-authoritative answer:
Name: smtp.enta.net
Address: 62.249.192.218
Name: smtp.enta.net
Address: 62.249.192.219
Name: smtp.enta.net
Address: 87.127.246.178

--------------------------------------
dig smtp.enta.net MX

; <<>> DiG 9.3.2 <<>> smtp.enta.net MX
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 737
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 0

;; QUESTION SECTION:
;smtp.enta.net. IN MX

;; AUTHORITY SECTION:
enta.net. 10800 IN SOA enta.net. hostmaster.enta.net. 2008011102 28800 7200 604800 86400

;; Query time: 30 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Fri Jan 11 14:24:19 2008
;; MSG SIZE rcvd: 78

Looking at this some more it would appear that Scalix is using an old configuration file when it is invoking Sendmail to send email out.

This is the only way I can explain the references to the old SMTP smarthost in the mail log...

Does anyone know which config file Scalix is using and where it is located?

By default it should use either /etc/mail/sendmail.cf or /etc/sendmail.cf. Where it's modified depends on your distro, on RH (and possibly SuSE) you need to check the startup script (/etc/init.d/sendmail) and the config file (/etc/sysconfig/sendmail) and see if there's anything defined there.