postfix dropping SMTP connection when receiving mail

Glenn Steen glenn.steen at gmail.com
Mon May 14 10:28:28 IST 2007


On 11/05/07, Jonathan Nalley <jnalley at fgp.com> wrote:
> Hi, I'm running MailScanner 4.58.9-1 using postfix 2.3.3 on Centos 5.
>
> I've set our M$ Exchange Server 2007 to use our MailScanner/Postfix box as a "smart host" such that all outbound mail would be sent through the MailScanner/postfix box.  The problem is that when the Exchange box is trying to pass along an e-mail to postfix, the connection and SMTP session gets dropped or something really bizarre is going on.
>
> Below is a snippet of /var/log/maillog (with postfix's debug_peer_level=1000).  For the purposes of these pasted log files. 192.168.0.207 is the exchange box and 192.168.0.208 is the Postfix box.
>
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: match_hostname: exchange.myinternaldomain.com ~? 192.168.0.0/24
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: match_hostaddr: 192.168.0.207 ~? 192.168.0.0/24
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 220 mailscanner.myinternaldomain.com ESMTP Postfix
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: watchdog_pat: 0x99266a0
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: vstream_fflush_some: fd 9 flush 83
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: vstream_buf_get_ready: fd 9 got 29
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: < exchange.myinternaldomain.com[192.168.0.207]: EHLO exchange.myinternaldomain.com
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-mailscanner.myinternaldomain.com
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-PIPELINING
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-SIZE 10240000
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-VRFY
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-ETRN
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-AUTH LOGIN PLAIN
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: match_list_match: exchange.myinternaldomain.com: no match
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: match_list_match: 192.168.0.207: no match
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-AUTH=LOGIN PLAIN
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-ENHANCEDSTATUSCODES
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250-8BITMIME
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: > exchange.myinternaldomain.com[192.168.0.207]: 250 DSN
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: watchdog_pat: 0x99266a0
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: vstream_fflush_some: fd 9 flush 178
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: smtp_get: EOF
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: match_hostname: exchange.myinternaldomain.com ~? 192.168.0.0/24
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: match_hostaddr: 192.168.0.207 ~? 192.168.0.0/24
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: lost connection after EHLO from exchange.myinternaldomain.com[192.168.0.207]
> May 11 09:40:23 mailscanner postfix/smtpd[18829]: disconnect from exchange.myinternaldomain.com[192.168.0.207]
>
>
> When enabling verbose session logging on the M$ Exchange box, below is the relevant SMTP session snippet:
>
> 2007-05-11T18:06:53.714Z,PostFixSendConnector,08C9612DB0BAB77B,0,,192.168.0.208:25,*,,attempting to connect
> 2007-05-11T18:06:53.714Z,PostFixSendConnector,08C9612DB0BAB77B,1,192.168.0.207:11256,192.168.0.208:25,+,,
> 2007-05-11T18:06:53.714Z,PostFixSendConnector,08C9612DB0BAB77B,2,192.168.0.207:11256,192.168.0.208:25,<,220 postfixbox.myinternaldomain.com ESMTP Postfix,
> 2007-05-11T18:06:53.714Z,PostFixSendConnector,08C9612DB0BAB77B,3,192.168.0.207:11256,192.168.0.208:25,>,EHLO exchange.myinternaldomain.com,
> 2007-05-11T18:06:53.714Z,PostFixSendConnector,08C9612DB0BAB77B,4,192.168.0.207:11256,192.168.0.208:25,<,
> 2007-05-11T18:06:53.714Z,PostFixSendConnector,08C9612DB0BAB77B,5,192.168.0.207:11256,192.168.0.208:25,-,,Local
>
> The second-to-last line actually has 65 spaces and then the closing comma but i couldn't get my mailer to display that in a desirable fashion.
>
> I know it's tempting to say or to think that the problem might be with the M$ Exchange box, but i can tell Exchange to use other Postfix/MailScanner installations (admittedly running MailScanner version 4.46.2-2 and postfix version 2.1.5) as the smart host and everything works smoothly.
>
> I've seen other posts in the mailing list with people having "smtp_get: EOF" in their logs and other similar entries but many were related to TLS, SSL, SASL and authentication and all that stuff but i'm not trying to do any of that here.  Any thoughts, wisdom, insights, fixes, and solutions greatly appreciated. Thanks In Advance.
>
>
> Below is postconf -n:
>
> alias_database = hash:/etc/aliases
> alias_maps = hash:/etc/aliases
> broken_sasl_auth_clients = yes
> command_directory = /usr/sbin
> config_directory = /etc/postfix
> daemon_directory = /usr/libexec/postfix
> debug_peer_level = 1000
> debug_peer_list = 192.168.0.207
> header_checks = regexp:/etc/postfix/header_checks
> html_directory = no
> inet_interfaces = all
> mail_owner = postfix
> mailq_path = /usr/bin/mailq.postfix
> manpage_directory = /usr/share/man
> mydestination = localhost.$mydomain, localhost, $mydomain
> mydomain = myinternaldomain.com
> myhostname = mailscanner.myinternaldomain.com
> mynetworks = 192.168.0.0/24, 127.0.0.0/8
> myorigin = $mydomain
> newaliases_path = /usr/bin/newaliases.postfix
> queue_directory = /var/spool/postfix
> readme_directory = /usr/share/doc/postfix-2.3.3/README_FILES
> relay_domains = $mydestination
> sample_directory = /usr/share/doc/postfix-2.3.3/samples
> sendmail_path = /usr/sbin/sendmail.postfix
> setgid_group = postdrop
> smtpd_banner = $myhostname ESMTP $mail_name
> smtpd_recipient_restrictions = permit_mynetworks permit_sasl_authenticated reject_unauth_destination
> smtpd_sasl_auth_enable = yes
> smtpd_sasl_path = smtpd
> transport_maps = hash:/etc/postfix/transport
> unknown_local_recipient_reject_code = 550
>

Not really an MailScanner problem this... If you telnet to it and
"play MTA", does it work then?
If you remove the SASL thing from recipient_restrictions, still the same?

-- 
-- Glenn
email: glenn < dot > steen < at > gmail < dot > com
work: glenn < dot > steen < at > ap1 < dot > se


More information about the MailScanner mailing list