Postfix will not send or receive, but sends to queue with flush error.

I have Webmin/Virtualmin running on Cent OS5 and everything seems fine, except that if I send mail (from webmail or POP3 account) it states sent, but is going into the server mailqueue. Have tried flush, but get this error:/usr/sbin/postqueue -c /etc/postfix -f …
postqueue: fatal: Cannot flush mail queue - mail system is down

But the Postfix server is running. Also Dovecot is running.

Also I cannot receiving any email from anyone, even localhost sent from webmail.

telnet localhost 110
Trying 127.0.0.1…
Connected to localhost.secureserver.net (127.0.0.1).
Escape character is ‘^]’.
+OK Dovecot ready.
user xxxx.shoots
+OK
pass XXXXX
+OK Logged in.
quit
+OK Logging out.

telnet shoots.com 110
Trying 208.109.126.xxx…
Connected to shoots.com (208.109.126.xxx).
Escape character is ‘^]’.
+OK Dovecot ready.
user xxxx.shoots
+OK
pass XXXX
+OK Logged in.
quit
+OK Logging out.

Connections look good. I’ve been at this for two days and can’t get anything to work. HELP!!!

Any ideas??

Howdy,

Rather than Dovecot, it’s Postfix that you’ll want to look into.

Postfix is made up of a number of different processes, so if there’s a configuration problem of some kind, it can look like it’s up and running, even when one of it’s processes is down.

What I’d suggest is restarting Postfix (using “/etc/init.d/postfix restart”), then take a peek in the email log to see if any errors show up. The email log is located in /var/log/maillog.

-Eric

Have tried the restart a few times, have stopped everything, restarted and even rebooted the server.

The only thing in the maillog is:

Jun 23 19:41:22 ip-208-109-123-210 postfix/master[22887]: warning: process /usr/libexec/postfix/smtpd pid 30934 exit status 1
Jun 23 19:41:22 ip-208-109-123-210 postfix/master[22887]: warning: process /usr/libexec/postfix/smtpd pid 30935 exit status 1
Jun 23 19:41:22 ip-208-109-123-210 postfix/master[22887]: warning: process /usr/libexec/postfix/smtpd pid 30936 exit status 1
Jun 23 19:41:22 ip-208-109-123-210 postfix/master[22887]: warning: process /usr/libexec/postfix/smtpd pid 30937 exit status 1
Jun 23 19:41:22 ip-208-109-123-210 postfix/master[22887]: warning: process /usr/libexec/postfix/smtpd pid 30938 exit status 1
Jun 23 19:42:20 ip-208-109-123-210 postfix/postfix-script: stopping the Postfix mail system
Jun 23 19:42:20 ip-208-109-123-210 postfix/master[22887]: terminating on signal 15
Jun 23 19:42:20 ip-208-109-123-210 postfix/postfix-script: starting the Postfix mail system
Jun 23 19:42:20 ip-208-109-123-210 postfix/master[30996]: daemon started – version 2.3.3, configuration /etc/postfix
Jun 23 19:42:21 ip-208-109-123-210 postfix/smtpd[30999]: fatal: unknown notify_classes value “david@euro-pacific.com” in “david@euro-pacific.com
Jun 23 19:42:22 ip-208-109-123-210 postfix/master[30996]: warning: process /usr/libexec/postfix/smtpd pid 30999 exit status 1
Jun 23 19:42:22 ip-208-109-123-210 postfix/master[30996]: warning: /usr/libexec/postfix/smtpd: bad command startup – throttling
Jun 23 23:42:53 ip-208-109-123-210 postfix/postqueue[31059]: fatal: Cannot flush mail queue - mail system is down

/etc/rc.d/init.d/postfix status
master (pid 30996) is running…

Hrm, it sounds like the error is saying there’s a “notify_classes” line setup that’s invalid.

If you run “postconf -n”, what output do you receive?

-Eric

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 = 2
home_mailbox = Maildir/
html_directory = no
mailbox_command = /usr/bin/procmail-wrapper -o -a $DOMAIN -d $LOGNAME
mailq_path = /usr/bin/mailq.postfix
manpage_directory = /usr/share/man
mydestination = $myhostname, localhost.$mydomain, localhost, $mydomain, ip-208-109-123-210.ip.secureserver.net, shoots.com
mynetworks_style = host
newaliases_path = /usr/bin/newaliases.postfix
readme_directory = /usr/share/doc/postfix-2.3.3/README_FILES
relayhost = p3smtpout.secureserver.net
sample_directory = /usr/share/doc/postfix-2.3.3/samples
sender_bcc_maps = hash:/etc/postfix/bcc
sendmail_path = /usr/sbin/sendmail.postfix
setgid_group = postdrop
smtpd_delay_reject = no
smtpd_recipient_restrictions = permit_mynetworks permit_sasl_authenticated reject_unauth_destination
smtpd_tls_cert_file = /etc/postfix/postfix.cert.pem
smtpd_tls_key_file = /etc/postfix/postfix.key.pem
smtpd_tls_security_level = may
soft_bounce = yes
unknown_local_recipient_reject_code = 550
virtual_alias_maps = hash:/etc/postfix/virtual

NB: I have soft bounce selected for testing

That doesn’t strike me as particularly odd… what about if you run this command, do you get any output:

grep notify_classes /etc/postfix/*

Perhaps one of the other config files is using that.

-Eric

/etc/postfix/main.cf.default:notify_classes = resource, software

I was just looking at the maillogs - the mail is being processed, but it isn’t going anywhere. Nothing in webmail account or via POP3 but lots in the queue. Look at the Dovecot reference.

Jun 23 23:52:31 ip-208-109-123-210 dovecot: pop3-login: Login: user=<dave.shoots>, method=PLAIN, rip=::ffff:68.236.219.119, lip=::ffff:208.109.126.180
Jun 23 23:52:31 ip-208-109-123-210 dovecot: POP3(dave.shoots): Effective uid=506, gid=504
Jun 23 23:52:31 ip-208-109-123-210 dovecot: POP3(dave.shoots): maildir: root exists (/home/shoots/homes/dave/Maildir)
Jun 23 23:52:31 ip-208-109-123-210 dovecot: POP3(dave.shoots): maildir: root=/home/shoots/homes/dave/Maildir, index=/home/shoots/homes/dave/Maildir, control=, inbox=
Jun 23 23:52:31 ip-208-109-123-210 dovecot: pop3-login: Login: user=, method=PLAIN, rip=::ffff:68.236.219.119, lip=::ffff:208.109.126.180
Jun 23 23:52:31 ip-208-109-123-210 dovecot: POP3(shoots): Effective uid=504, gid=504
Jun 23 23:52:31 ip-208-109-123-210 dovecot: POP3(shoots): maildir: root exists (/home/shoots/Maildir)
Jun 23 23:52:31 ip-208-109-123-210 dovecot: POP3(shoots): maildir: root=/home/shoots/Maildir, index=/home/shoots/Maildir, control=, inbox=
Jun 23 23:52:33 ip-208-109-123-210 dovecot: POP3(dave.shoots): Disconnected: Logged out top=0/0, retr=0/0, del=0/0, size=0
Jun 23 23:52:33 ip-208-109-123-210 dovecot: POP3(shoots): Disconnected: Logged out top=0/0, retr=0/0, del=0/0, size=0

Have found something strange - any ideas?

I have created 3 mail user accounts, but there are no physical mailboxes created in the MailDir directory in the $HOME/Maildir/ or in /var/mail - but the mail users exist in Virtualmin. That would explain why all of the mail (in and out) is accumulating in the postfix mail queue.

If anyone has some brilliant idea as to what is causing this error - I’d love to hear as I’m in email limbo. Have tried changing MailDir settings, nothing. Could be a permissions problem, but where. This is a fresh install from install.sh.

I tried turning off postfix, uninstalled and did a fresh install. Still can’t get the stupid POstfix to work. Driving me up the wall. The online documentation is really bad and does not refer to Centos5 versions.

I set everything to default:

BIND DNS server is installed, and the system is configured to use it.
A problem was found with your Postfix virtual maps : No map sources were found in the Postfix configuration

… your system is not ready for use by Virtualmin.