undelivered local mail sitting in /var/spool/clientmqueue

[Almost-retired]

So I am not getting the emails that various functions send me

I have added a line
amanda:  gene
in /etc/aliases, and ran newaliases (which truncates the mailman aliases filename in the final :include: line by trimming the last s and then complains about the name being illegal).

So I have several messages intended for gene@FQDN that are stuck in /var/spool/clientmqueue.

Some have a line that they were refused by 127.0.0.1 if that is a useful clue.

I have tried sending mail to myself, and while its sent quickly enough, it never arrives, not even in /var/spool/clientmqueue
Here is a paste of the last one from the user amanda:

$_amanda@localhost
${daemon_flags}c u
Samanda
Aamanda@coyote.coyote.den
MDeferred: Connection refused by [127.0.0.1]
C:gene@coyote.coyote.den
rRFC822; gene@coyote.coyote.den
RPFD:gene@coyote.coyote.den
H?P?Return-Path: <�g>
H??Received: (from amanda@localhost)
        by coyote.coyote.den (8.14.4/8.14.4/Submit) id p3L72iGH029822
        for gene@coyote.coyote.den; Thu, 21 Apr 2011 03:02:44 -0400
H?F?From: Advanced Maryland Automatic Newtwork Disk Archiver <amanda>
H?x?Full-Name: Advanced Maryland Automatic Newtwork Disk Archiver
H?M?Message-Id: <201104210702.p3L72iGH029822@coyote.coyote.den>
H??Date: Thu, 21 Apr 2011 03:02:44 -0400
H??To: gene@coyote.coyote.den
H??Subject: The Coyote Den AMANDA MAIL REPORT FOR April 21, 2011
H??User-Agent: Heirloom mailx 12.4 7/29/08
H??MIME-Version: 1.0
H??Content-Type: text/plain; charset=us-ascii
H??Content-Transfer-Encoding: 7bit

Can anyone detect what I need to do here?

Thanks & Cheers, Gene

[Almost-retired]

Thanks to a 2004 message I found on google, it turns out that sendmail was not running.

So now my question changes to why not?  It is set to run at startup in run-level=5 but a restart got this:

[root@coyote init.d]# service sendmail restart
Shutting down sendmail:                                                                                                                                  [FAILED]
Starting sendmail:                                                                                                                                       [  OK  ]
Starting sm-client:                                                                                                                                      [  OK  ]

I have not taken any actions that would have killed sendmail, so why was it not running?

I found these 3 lines in the messages log, corresponding to sendmail's restart:

Apr 21 11:43:10 coyote sendmail[10657]: auxpropfunc error invalid parameter supplied
Apr 21 11:43:10 coyote sendmail[10657]: sql_select option missing
Apr 21 11:43:10 coyote sendmail[10657]: auxpropfunc error no mechanism available

This auxpropfunc does not grep in /etc/mail/sendmail.*, empty return.

What is this /Swahili/ trying to tell me?

Thanks for any clues.

Cheers, Gene

[muungwana]

do the emails get sent now that sendmail is running?

Are you sure it is set to auto run as startup? You can add that line in "/etc/rc.local" script file and the command will be run on every reboot.

[Almost-retired]

Yes, it is set to autorun.  However since this is a crash recovery situation, I probably should not lay my hand too close to a bible when I say that. :-)

By crash, I updated the firmware in my boot drive, and that changed the blkid's, and shuffled the labels.  By the time I had deduced that, the system was not recoverable.

Using Seagates own drive firmware updater iso I had downloaded on April 12th.  If you use it, be prepared for an extended session of finding menu.lst or grub.conf, and editing the blkid's therein and if labels are in use (I prefer them but they were scrambled too, the names were all ok, but were now connected to the wrong partitions after the firmware update)

It updated 2 of my drives, but only scrambled the 'a' drive.  sdb was being a problem, spamming the logs with reset msgs, the firmware fixed that, but its write speeds are in the <1meg/sec range so I'll warranty it anyway.

Things are slowly falling into place and I should reboot just to see what else is munged yet.

Putting it in rc.local might be ok too, but the screen has been replaced by the X startup by the time local runs and you can't see the success/fails then, unless I go back to a 3 in inittab and startx after I log in.  I did do it that way for a decade or so.

Thanks for replying, I appreciate it.

Cheers, Gene