Results 1 to 4 of 4

Thread: No matter what I do, postfix won't start

  1. #1
    bmarkovic is offline Junior Member
    Join Date
    Jul 2009
    Location
    Bosnia Herzegovina
    Posts
    8
    Rep Power
    6

    Default No matter what I do, postfix won't start

    This started this morning. For some reason I can't send any mail.

    I have split DNS and a MTA relay. The relay was setup properly and has worked for years and numerous upgrades (both ZCS and Ubuntu dist-upgrades).

    To move the obviuos out of the way:

    /etc/hosts
    Code:
    127.0.0.1       localhost.localdomain localhost
    192.168.0.254   server.domain.tld server
    Off course, server.domain.tld is the name entered for the server in server settings (domain.tld is obviously masking the real domain name).

    Also, on another machine inside the split DNS (i.e. inside firewall):

    nslookup -query=mx domain.tld
    Code:
    Server:         192.168.0.254
    Address:        192.168.0.254#53
    
    domain.tld        mail exchanger = 10 server.domain.tld.
    From outside MX is the MTA relay host and Zimbra grabs user mail through POP. As I said, this was working fine for 2 years.

    dig domain.tld any
    Code:
    ; <<>> DiG 9.2.4 <<>> domain.tld any
    ;; global options:  printcmd
    ;; Got answer:
    ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 12656
    ;; flags: qr aa rd ra; QUERY: 1, ANSWER: 4, AUTHORITY: 0, ADDITIONAL: 1
    
    ;; QUESTION SECTION:
    ;domain.tld.                      IN      ANY
    
    ;; ANSWER SECTION:
    domain.tld.               86400   IN      SOA     server.domain.tld. admin.domain.tld. 24 604800 86400 2419200 86400
    domain.tld.               86400   IN      NS      server.domain.tld.
    domain.tld.               86400   IN      MX      10 server.domain.tld.
    domain.tld.               86400   IN      A       NNN.NNN.NNN.NNN
    
    ;; ADDITIONAL SECTION:
    server.domain.tld.        86400   IN      A       192.168.0.254
    
    ;; Query time: 3 msec
    ;; SERVER: 127.0.0.1#53(127.0.0.1)
    ;; WHEN: Mon Sep 24 23:00:09 2012
    ;; MSG SIZE  rcvd: 137

    Nothing runs on port 25, not even postfix unfortunately.

    So yes, no exim, sendmail or any other MTA clogging port 25.

    Logs are, as always, very helpful (not!):

    tail -n 20 /var/log/zimbra.log
    Code:
    Sep 24 22:16:44 server zmmailboxdmgr[16908]: status requested
    Sep 24 22:16:44 server zmmailboxdmgr[16908]: status OK
    Sep 24 22:16:44 server zmmailboxdmgr[16916]: status requested
    Sep 24 22:16:44 server zmmailboxdmgr[16916]: status OK
    Sep 24 22:17:03 server postfix/postqueue[17055]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:17:33 server postfix/postqueue[17073]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:17:47 server zmmailboxdmgr[17132]: status requested
    Sep 24 22:17:47 server zmmailboxdmgr[17132]: status OK
    Sep 24 22:17:47 server zmmailboxdmgr[17140]: status requested
    Sep 24 22:17:47 server zmmailboxdmgr[17140]: status OK
    Sep 24 22:18:03 server postfix/postqueue[17384]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:18:07 server zmmailboxdmgr[17492]: status requested
    Sep 24 22:18:07 server zmmailboxdmgr[17492]: status OK
    Sep 24 22:18:33 server postfix/postqueue[17604]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:18:50 server zmmailboxdmgr[17662]: status requested
    Sep 24 22:18:50 server zmmailboxdmgr[17662]: status OK
    Sep 24 22:18:50 server zmmailboxdmgr[17670]: status requested
    Sep 24 22:18:50 server zmmailboxdmgr[17670]: status OK
    Sep 24 22:19:03 server postfix/postqueue[17785]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:19:33 server postfix/postqueue[17797]: fatal: Queue report unavailable - mail system is down
    tail -n 20 /var/log/mail.log
    Code:
    Sep 24 22:24:04 server zmmailboxdmgr[19499]: status requested
    Sep 24 22:24:04 server zmmailboxdmgr[19499]: status OK
    Sep 24 22:24:04 server zmmailboxdmgr[19507]: status requested
    Sep 24 22:24:04 server zmmailboxdmgr[19507]: status OK
    Sep 24 22:24:09 server zmmailboxdmgr[19692]: status requested
    Sep 24 22:24:09 server zmmailboxdmgr[19692]: status OK
    Sep 24 22:24:33 server postfix/postqueue[19822]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:25:03 server postfix/postqueue[19845]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:25:08 server zmmailboxdmgr[19896]: status requested
    Sep 24 22:25:08 server zmmailboxdmgr[19896]: status OK
    Sep 24 22:25:08 server zmmailboxdmgr[19904]: status requested
    Sep 24 22:25:08 server zmmailboxdmgr[19904]: status OK
    Sep 24 22:25:33 server postfix/postqueue[20029]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:26:03 server postfix/postqueue[20151]: fatal: Queue report unavailable - mail system is down
    Sep 24 22:26:06 server zmmailboxdmgr[20224]: status requested
    Sep 24 22:26:06 server zmmailboxdmgr[20224]: status OK
    Sep 24 22:26:11 server zmmailboxdmgr[20378]: status requested
    Sep 24 22:26:11 server zmmailboxdmgr[20378]: status OK
    Sep 24 22:26:11 server zmmailboxdmgr[20386]: status requested
    Sep 24 22:26:11 server zmmailboxdmgr[20386]: status OK
    This is an example off course. The log is full of this, over the whole weekend.

    Postfix check only warns about ownership of conf files, which as I understand from various forum posts on the subjec, is normal.

    zimmbra@server: postfix check
    Code:
    postfix/postfix-script: warning: not owned by root: /opt/zimbra/postfix-2.7.4.2z/conf/main.cf
    postfix/postfix-script: warning: not owned by root: /opt/zimbra/postfix-2.7.4.2z/conf/master.cf
    postfix/postfix-script: warning: not owned by root: /opt/zimbra/postfix-2.7.4.2z/conf/master.cf.in
    postfix/postfix-script: warning: not owned by postfix: /opt/zimbra/data/postfix/data/./master.lock
    postfix/postfix-script: warning: not owned by postfix: /opt/zimbra/data/postfix/data/./prng_exch
    I have tried the following from various threads:

    • Deleted pid file: /opt/zimbra/data/postfix/spool/pid/master.pid that was left after postfix fails.
    • Tried starting postfix manually (postfix reload). Same.
    • Added "POSTCONF inet_protocols ipv4" to the end of mta section of /opt/zimbra/conf/zmmta.cf just before RESTART mta
    • Reinstalling Zimbra using ./install.sh (as this removes packages and reinstalls them so hopefully it should rewrite some wrecked configs, but didnt').


    So, anyone?
    Last edited by bmarkovic; 09-24-2012 at 11:21 PM.

  2. #2
    phoenix is online now Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,581
    Rep Power
    57

    Default

    What about checking if your certificates are still valid or if they've expired?
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

  3. #3
    bmarkovic is offline Junior Member
    Join Date
    Jul 2009
    Location
    Bosnia Herzegovina
    Posts
    8
    Rep Power
    6

    Default

    Quote Originally Posted by phoenix View Post
    What about checking if your certificates are still valid or if they've expired?
    No they're fine.

    They expire somewhere in 2020 (yeah, I know, but this Zimbra is used mainly behind firewall). Also if I remember correctly, if you fuxor your certificates some services (including mailbox) won't start. In my case I have:

    zmcontrol status
    Code:
    Host server.domain.tld
    	antispam                Running
    	ldap                    Running
    	logger                  Running
    	mailbox                 Running
    	memcached               Running
    	mta                     Stopped
    		postfix is not running
    	snmp                    Running
    	spell                   Running
    	stats                   Running
    	zmconfigd               Running
    And the users are able to receive mail and LMTP service on 7025 is running. However neither web-client or thunderbird users are able to send mail (nothing is listening on 25 so clients report errors).
    Last edited by bmarkovic; 09-24-2012 at 11:37 PM. Reason: Clarification

  4. #4
    bmarkovic is offline Junior Member
    Join Date
    Jul 2009
    Location
    Bosnia Herzegovina
    Posts
    8
    Rep Power
    6

    Default

    I'll add this:

    postfix -vvvvv reload results in attached output (forum won't let me paste it inside CODE tag).

    At least I've formed an opinion on postfix -- it's POS. No serious server software should die without logging why. And postfix is the first unix software with such behaviour in my 10 years of IT career.

    Also, it's odd that despite it's logging that it's dead, zmcontrol status reported MTA as running for most of yesterday. It's just when I've deleted the PID file that I've got postfix not running message in there.

    Also I've tried this too:

    /opt/zimbra/libexec/zmfixperms -verbose -extended

    Didn't do it.

    I guess I'll be looking into backing up the databases and message store and reinstalling, right?
    Attached Files Attached Files
    Last edited by bmarkovic; 09-24-2012 at 11:41 PM. Reason: More info

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. mail server name? Does it matter?
    By jim.thornton in forum Installation
    Replies: 12
    Last Post: 10-02-2012, 07:54 AM
  2. Postfix won't start
    By ksoviero in forum Administrators
    Replies: 2
    Last Post: 07-10-2011, 05:21 PM
  3. Postfix won't start
    By stace in forum Installation
    Replies: 1
    Last Post: 05-26-2009, 03:19 AM
  4. Replies: 2
    Last Post: 02-20-2007, 08:53 AM
  5. zmcontrol start - postfix still won't start
    By crp in forum Administrators
    Replies: 6
    Last Post: 10-26-2006, 03:49 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •