Moved from Server Ops Thread - Error in mail.log : NOQUEUE: recipient address reject

And it was all going so well! (sigh)

Today, we had no mail in the boxes, none of the usual spam, falling like snow in the night. Very odd, so I checked /var/log/mail.log , and find this sort of error:

Mar 19 09:00:43 deb-server postfix/smtpd [3169]: NOQUEUE: reject: RCTP from xxx.xxx.xxx.whatever.com [xxx.xxx.xxx.xxx]: 550 <[email protected]> Recipient address rejected: User unknown in local recipient table; from =<[email protected]> ...etc.

This is what I'd been doing the night before:

I was trying to add a virtual host for squirrelmail webmail via apache2, following the suggestions and instructions from Torvald in this thread:

http://www.howtoforge.com/forums/showthread.php?t=4964&page=2

I encountered a difficulty when I accidentally REMOVED/DELETED my host/web from the ISPClient-Site Managment Tab within ISPConfig. (I was following Torvald's first step, attempting to add a new web named mail.domain.com) I quickly located it in the Recycling Bin, and undeleted it.

Then I followed Torvald's steps 1 through 4, made the appropriate changes to my DNS service (I do not use my own DNS server), restarted everything. My virtual host for webmail was supposed to be port 80 on the same ip as my standard web. (I am still unsure if this was the correct thing to do.)

It did not work. If I pointed a browser at https://webmail.mydomain.com, my standard web would open, but not squirrelmail. Furthermore, if I attempted to open a mailbox using squirrelmail or squirreloutlook going directly to webmail (https://www.mydomain.com:81/squirrelmail/]) none of the mailboxes would open, giving me username/password errors.

I "undid" Torvald's instructions, by using ISPConfig to remove the co-domain mail.mydomain.com. I renamed the file /etc/apache2/sites-available/webmail.conf to webmail.conf.save. I removed the symlink between sites-available/webmail.conf and sites-enabled/webmail.conf. And I reset my dns service to its original configuration. Then I called it a day...

Last night pop3 mail seemed to be unaffected by any of this, but this morning I am getting the error shown at the start of this post. I really need help knowing what to look at.

I don't know what I've broken exactly. I've checked and restarted things individually, and also rebooted the server entirely. I have taken the server off-line, so our mail can go to a spooler provided by our dns service (we use dnsexit.com).

Here is the output of netstat -tap (everything is started but all the router ports are closed):

deb-server:~# netstat -tap
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 *:imaps *:* LISTEN 1175/couriertcpd
tcp 0 0 *:9090 *:* LISTEN 1425/perl
tcp 0 0 *op3s *:* LISTEN 1198/couriertcpd
tcp 0 0 localhost.localdom:9092 *:* LISTEN 1489/mysqld
tcp 0 0 *:9000 *:* LISTEN 1425/perl
tcp 0 0 *:mysql *:* LISTEN 1256/mysqld
tcp 0 0 *op3 *:* LISTEN 1184/couriertcpd
tcp 0 0 *:imap2 *:* LISTEN 1161/couriertcpd
tcp 0 0 *:www *:* LISTEN 3070/apache2
tcp 0 0 *:81 *:* LISTEN 1484/ispconfig_http
tcp 0 0 virt-server.swif:domain *:* LISTEN 2655/named
tcp 0 0 deb-server.swift:domain *:* LISTEN 2655/named
tcp 0 0 localhost.locald:domain *:* LISTEN 2655/named
tcp 0 0 *:ftp *:* LISTEN 1780/proftpd: (acce
tcp 0 0 *:ssh *:* LISTEN 1427/sshd
tcp 0 0 localhost.localdoma:953 *:* LISTEN 2655/named
tcp 0 0 *:smtp *:* LISTEN 2302/master
tcp 0 0 *:https *:* LISTEN 3070/apache2
tcp 0 0 *:3483 *:* LISTEN 1425/perl
tcp 0 568 deb-server.swiftpa:3483 192.168.0.100:45284 ESTABLISHED1425/perl
tcp 0 0 localhost.localdo:32768 localhost.localdom:9092 ESTABLISHED1425/perl
tcp 0 2976 deb-server.swiftpas:ssh 192.168.0.101:53024 ESTABLISHED1812/0
tcp 0 0 localhost.localdom:9092 localhost.localdo:32768 ESTABLISHED1489/mysqld


Any help would be deeply appreciated!

Elliott

 

Newbie learning curve. Thanks all!

Got some good help from a private reply, and thanks, too, Till. You're quite right, the issue was mostly in /etc/postfix/virtuserstable .

When a web or domain is deleted using ispconfig, a deal of the pertinant data goes to Recycle, where it has a privileged after-life. It forces new generations of usernames to be generated (web2_xxxxx, web3_xxxxx, etc.) UNLESS you delete them from Recycling.

For a noob like myself, this caused all manner of confusion.


What a great tool, tho! Ispconfig allows the user to pull or reinstate webs with such ease!

The second matter was that of malformed host names, because of my own clumsiness in using ispconfig when setting things up at first. This required checking the form of things in /etc/postfix/local-host-names. This was preventing some mail from finding the server, until I caught on.

If you are new to mail server management, and reading this, please open a terminal and run: tail -f /var/log/mail.log , and just watch what happens as you initiate a mail outgoing, or incoming to the server, or as you check mail from T-bird or using a box set up for pop or imap mail. It is very instructive.

Noob to noob nooz...

Elliott