[BlueOnyx:18984] Re: Email issues

[Michael Stauber]

Hi Gregg,

> I don't know if this has anything to do with the issues we are having
> lately, but today I noticed that on one  of my servers, I'm not able to get
> any email.  The configurations seem to be all wrong.
> It's looking for /etc/aliases, and I know that the aliases are at
> /etc/mail/aliases
>
> I also noticed something strange when logging on to the system.  Usually
> after you SU to root, you get: root at servername admin
> I'm getting just bash-4.1
> In my email logs I get these
> Jan 13 15:30:02 main sendmail[7524]: NOQUEUE: SYSERR(root): QueueDirectory
> (Q) option must be set

this looks like you did run out of space on the / partition during a
reboot. As a result several config files got hosed.

Let's start with the simple stuff:

Edit /root/.bashrc and put this in it (minus the -------- lines):

----------------
# Source global definitions
if [ -f /etc/bashrc ]; then
        . /etc/bashrc
fi
----------------

That will show the the box name at the bash prompt again.

Sendmail:

Yes, some CentOS versions do have an /etc/aliases, but we use
/etc/mail/aliases. Your sendmail.mc and sendmail.cf are probably hosed.
That's why you get all sorts of weird email issues.

A good copy of sendmail.mc is located in this directory:
/usr/sausalito/configs/sendmail

[root at 5208r sendmail]# ls -k1 /usr/sausalito/configs/sendmail
aliases
popauth.m4
sendmail.mc
sendmail.mc.bxold
sendmail.mc.mailman
sendmail.mc.majordomo
sendmail.pam

What you'll need for anything from 5106R up to 5209R is named
sendmail.mc.mailman

So copy that to /etc/mail:

cp /usr/sausalito/configs/sendmail/sendmail.mc.mailman /etc/mail/sendmail.mc

Then (just to be sure) delete your wrecked sendmail.cf:

rm /etc/mail/sendmail.cf

Then rebuild a good sendmail.cf from the sendmail.mc:

cd /etc/mail
make all

Restart Sendmail and see if it works again:

killall -9 sendmail
/sbin/service sendmail restart
tail -f /var/log/maillog

> Jan 13 15:30:07 main dovecot: pop3-login: Fatal: Couldn't parse private
> ssl_key: error:0906D06C:PEM routines:PEM_read_bio:no start line: Expecting:
> ANY PRIVATE KEY

That looks like your SSL certificates for Dovecot are damaged, too. They
usually reside here: /etc/pki/dovecot/certs/

Easiest fix: Go to the BlueOnyx GUI. Click on "Server Management" /
"Security" / "SSL" and create a new SSL certificate. That will create
one for the GUI, Dovecot and Sendmail. You might as well try the new
"Let's Encrypt" certs while you're at it.

But chances are: When you try to login to the GUI you might see weird
errors related to PHP. If that is the case, then your
/etc/admserv/php.ini is wrecked, too.

Fix for that:
yum reinstall base-admserv-glue
/sbin/service admserv restart

Let me know if there is anything else.

-- 
With best regards

Michael Stauber