scritps send email to monitor but does not reach mail server

[aaron@mmsm.com]

I have web server that sends via email to our helpdesk if there is a problem with itself. Plus I have web forms that send email to our helpdesk. I am not getting any email. This did work on our old email server which we just replaced with Scalix Community edition. I have search these forums and the knowledge base for help but cannot find anything. Please help since our company rely's heavily on email for support.

I hate the thoughts of going back to our old email server.

Thank you in advance.

[ScalixSupport]

If you can give us some details on the logging you've checked and the route that the mail takes, that would help us as, so far, there is nothing to work with.

Cheers

Dave

[aaron@mmsm.com]

Thank you for the quick response and as much as I would like too offer an error I am not sure that I can post any errors. We don't get any errors (unless you know where I canfind them) We did check the Internet Gateway logs and nothing shows up that was sent from the webserver.

What I can do is show you what the script does....

mail -s 'unresolved summary report error' support@mmsm.com < $LOG

it does not send nor give an error anywhere that we can tell.

If we run the script then do a tail /var/log/maillog there doesnt' seem to be any activity from the webserver.

Thanks again

[DanLikesBeer]

What Aaron is trying to do is send email from a local redhat server to the local scalix email server using the 'mail' command. Used to work on the 'old' system, but now the messages seem to go nowhere.

/var/log/maillog doesn't even show an attempt at delivery

the same 'mail' command from the webserver seems to deliver fine to external addresses.

the smtpd.cfg is set:
_________________________________________
RELAY accept 127.0.0.1
RELAY accept .ourdomain.com
RELAY Log_Reject ALL

# extra rules added to prevent open relay usage
RECIPIENT Log_Reject *@*@*
RECIPIENT Log_Reject *%*
RECIPIENT Log_Reject *!*
RECIPIENT Log_Reject *#*@*
__________________________________________

Question is, what can be done to tell Scalix to accept email messages from this six-inches-away host? It is driving both of us nuts.

P.S. Tried using omsend, but get :
[OM 8013] The name/mailnode contains invalid characters

So not sure what THAT is all about either.

[DanLikesBeer]

Went to the smtpd.cfg file and disabled all the relay-blocking. Restarted the smtpd daemon and tried again. got the following in /var/log/maillog:

Code: Select all

Jun 29 11:36:27 mage sendmail[2262]: k5S2ZnWY015287: to=<root@www.ourdumbdomain.com>, delay=1+13:00:38, xdelay=00:00:00, mailer=esmtp, pri=3392321, relay=www.ourdumbdomain.com., dsn=4.0.0, stat=Deferred: Connection refused by www.ourdumbdomain.com.
Jun 29 11:36:27 mage sendmail[2262]: k5S233W4013434: to=<edi@www.ourdumbdomain.com>, delay=1+13:33:24, xdelay=00:00:00, mailer=esmtp, pri=3488648, relay=www.ourdumbdomain.com., dsn=4.0.0, stat=Deferred: Connection refused by www.ourdumbdomain.com.

So it DID try and connect that time, but still no delivery.

[ScalixSupport]

The mail command talks to sendmail, so any mail you "mail" will be processed by sendmail and not by Scalix. The log snippet you posted basically says the receiving host refused the connections, which means nobody was listening on port 25.

Sascha.

[DanLikesBeer]

Here's the rub. I CAN telnet to port 25 on the scalix server, send a message, and have it be delivered. Or does the www server I'm trying to send the mail from need to be listening?

Code: Select all

[edi@www]$ telnet mage 25
Trying 192.168.200.7...
Connected to mage.ourdumbdomain.com (192.168.200.7).
Escape character is '^]'.
220 mage.ourdumbdomain.com ESMTP Scalix SMTP Relay 10.0.1.3; Thu, 29 Jun 2006 13:14:54 -0400 (EDT)
EHLO www.ourdumbdomain.com
250-mage.ourdumbdomain.com Hello www.ourdumbdomain.com [192.168.200.4], pleased to meet you
250-AUTH GSSAPI LOGIN PLAIN CRAM-MD5 DIGEST-MD5
250-AUTH=LOGIN
250-DSN
250 8BITMIME
MAIL FROM:edi@ourdumbdomain.com
250 edi@ourdumbdomain.com... Sender ok
RCPT TO:dan@ourdumbdomain.com
250 Ok
DATA
354 Enter mail, end with "." on a line by itself (relay)
this is a test
.
250 Ok
QUIT
221 mage.ourdumbdomain.com closing connection
Connection closed by foreign host.


So does the mail/sendmail I'm using on the www server need additonal authentication 'switches' of some sort?

[ScalixSupport]

When you use the mail command, it talks to sendmail on localhost, not Scalix. All messages sent with the mail command will only show up in the sendmail logs. This is completely outside of Scalix, your problem is NOT with Scalix.

So you need to test telnet www.yourdomain.com 25 on that server to see what happens.

un 29 11:36:27 mage sendmail[2262]: k5S2ZnWY015287: to=<root@www.ourdumbdomain.com>, delay=1+13:00:38, xdelay=00:00:00, mailer=esmtp, pri=3392321, relay=www.ourdumbdomain.com., dsn=4.0.0, stat=Deferred: Connection refused by www.ourdumbdomain.com.

states that www.ourdumbdomain.com does not respond on port 25.

Sascha.

[DanLikesBeer]

When I try to telnet to port 25 on the www server (not the scalix server) I get 'connection refused'. But on a shell on the www server, I can telnet to port 25 on localhost. So sendmail is running, but only accepting connections from itself.

But where I get confused is that it still works. I can send mail from that host to any other internet based host, but not the scalix server next to it.

It's not really a big deal, as I've got a work-around in place now. Just one of those puzzling problems that irritates me. If anything, I've learned enough about sendmail to know that I don't know squat about sendmail. :oops:

Thanks for trying to help!

Dan

[ScalixSupport]

If you get connection refused this means either the Scalix SMTPD is not running or a firewall is preventing you from connecting. You have posted an SMTP conversation before which clearly showed

220 mage.ourdumbdomain.com ESMTP Scalix SMTP Relay

so Scalix SMTPD is running. I am guessing you have a firewall on this box?

Does mailq -Ac give you anything?

Sascha.

[aaron@mmsm.com]

#mailq -Ac
/var/spool/clientmque is empty
Total requests : 0

Sorry took a couple of days off for the holiday.

[jonny]

*bump*

Same problem here - we have scripts on numerous other servers which attempt to sendmail to mailboxes.

Running a sendmail -v -q on the "sending machine" produces:

Code: Select all

[root@sending_machine mail]# sendmail -v -q
Running /var/spool/mqueue/k74A4lXw031774 (sequence 1 of 1)
<john.reeves@ourdomain.co.uk>... Connecting to mail.ourdomain.net. via esmtp...
<john.reeves@ourdomain.co.uk>... Deferred: Connection timed out with mail.ourdomain.net.


SMTPd.conf on the Scalix Box (mail.ourdomain.net)

Code: Select all

# Spam Assassin
SMTPFILTER=TRUE
RELAY accept 127.0.0.1
RELAY accept .ourdomainl.net
RELAY accept 10.0.2.12
RELAY Log_Reject ALL
# extra rules added to prevent open relay usage
RECIPIENT Log_Reject *@*@*
RECIPIENT Log_Reject *%*
RECIPIENT Log_Reject *!*
RECIPIENT Log_Reject *#*@*


The Sending Machine can resolve mail.ourdomain.net and I can successfully telnet onto mail.ourdomain.net 25 and send mail that way.

Thanks
John

[btisdall]

Perhaps this is relevant to your issue:

http://tinyurl.com/hoy9a

[jonny]

Hi Ben

Thanks for the thread - I had a good read through it but didn't see how it related to my situation. Why is Scalix not accepting an incoming SMTP relay from another machine on our network even though I have specified it in /var/opt/scalix/smtpd.cfg?!

[btisdall]

I don't think it's a question of Scalix smtpd rejecting the delivery, it seems as if the connection is never established.

Ttry turning up the sendmail debug level on one of the sending machines & tailing the logs.

For good measure on your Scalix box add DEBUG_LOG=TRUE to your smtpd.cfg, restart smtpd & tail /var/opt/scalix/tmp/smtpd.log (remembering to reverse the steps when done).