sendmail takes care of creating some necessary directories, one of them is /mail/spool/mqueue. After clean installation it should be: drwx------ 2 root root 4096 Nov 12 19:49 mqueue But instead of that, it is: drwxr-xr-x 2 smmsp root 4096 Nov 12 19:39 mqueue Sendmail refuses (or can not) write emails and exits with error message: 421 4.3.0 collect: Cannot write ./dfnACK6CaU032612 (bfcommit, uid=0, gid=209): Permission denied Reproducible: Always Steps to Reproduce: emerge -C ssmtp emerge sendmail telnet-bsd useradd -m -g users -s /bin/false testuser /etc/init.d/sendmail start telnet localhost 25 Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. 220 test ESMTP Sendmail 8.14.0/8.14.0; Thu, 12 Nov 2009 20:06:12 GMT HELO localhost 250 test Hello [localhost], pleased to meet you MAIL FROM:root@some.mydomain.com 250 2.1.0 root@some.mydomain.com... Sender ok RCPT TO:testuser@localhost 250 2.1.5 testuser@localhost... Recipient ok DATA 354 Enter mail, end with "." on a line by itself Test message. . 421 4.3.0 collect: Cannot write ./dfnACK6CaU032612 (bfcommit, uid=0, gid=209): Permission denied Connection closed by foreign host. Actual Results: Message is not delivered, and stays in /var/spool/clientmqueue Expected Results: Message should be delivered to $HOME/.maildir/new Info from Sendmail Installation And Operation Guide: ... 1.3.7. /var/spool/mqueue The directory /var/spool/mqueue should be created to hold the mail queue. This directory should be mode 0700 and owned by root. ... So the fix is very simple: chown root /var/spool/mqueue chmod 700 /var/spool/mqueue And it really works! :-)
Damned, I screwed the first line and even summary! It should have been: /var/spool/mqueue :-(
Is this still a problem? I have drwx------ 2 root root 60 Jun 14 17:52 mqueue after a fresh install with sendmail-8.14.5.
(In reply to comment #2) > Is this still a problem? I have > > drwx------ 2 root root 60 Jun 14 17:52 mqueue > > after a fresh install with sendmail-8.14.5. AFAIR yes. Not long ago I've find similiar issue. See #425626
