[SOLVED] Zimbra send but not receive CentOS 5 VM

[killc]

OK, I have scoured the depths of the forums and net for help but I am just at a loss to get things working. First the setup:

Dynamic DNS (no-ip.org) MX is set
Westell DSL/Router open
Windows Xp Pro "server" running VMware server 2
CentOS 5 VM using NAT (and on the host port forwarding all set for each port)

I can send email fine from Zimbra.
When I run an online test of SMTP from the web this is the result:

RESULT: .no-ip.org
Banner: 220 .no-ip.org ESMTP Postfix
Connect Time: 0 seconds - Good
Transaction Time: 0.375 seconds - Good
Relay Check: OK - This server is not an open relay.
Rev DNS Check: OK - resolves to .pitt.east.verizon.net
GeoCode Info: Geocoding server is unavailable
Session Transcript:
HELO please-read-policy.mxtoolbox.com
250 .no-ip.o [78 ms]
MAIL FROM:
250 2.1.0 [109 ms]
RCPT TO:
554 5.7.1 : Relay access deni [78 ms]
QUIT
221 2.0.0 B [109 ms]

I can telnet to port 25 from the Internet and get a response fine as well.

On the Zimbra Admin panel I can see the emails attempted to be delivered but with "connection refused" as the reason under Mail Queues.

My guess is that I have something local screwed up. Possibly bad info in Zimbra or in CentOS 5 /etc/hosts. I think I have a .no-ip.org somewhere a localhost.localdomain should be or vise-versa or else some local DNS issue.

I'll be happy to supply the configs or any info needed to help me properly.

Thanks!

UPDATE:

I'm not sure why my last reply didn't appear, but here goes again:

When my /etc/hosts file is set to this:

127.0.0.1 server.no-ip.org localhost
#.#.#.# server.no-ip.org mail
::1 localhost6.localdomain6 localhost6

Zimbra starts and I can send and can connect via SMTP but get relay access denied.

when I change it to:

127.0.0.1 localhost.localdomain localhost
#.#.#.# server.no-ip.org mail
::1 localhost6.localdomain6 localhost6

Zimbra does not start:

[zimbra@server ~]$ zmcontrol start
Host server.no-ip.org
Starting ldap...Done.
FAILED
Failed to start slapd. Attempting debug start to determine error.
daemon: bind(7) failed errno=99 (Cannot assign requested address)
slap_open_listener: failed on ldap://server.no-ip.org:389


I believe the problem has to do with the fact that it is a dynamic IP and/or DNS, but I cannot figure out a solution. Any help is greatly appreciated.

Thanks!

[killc]

Can anyone help? To update:

I think I may have an idea of the issue. Since it is a dynamic IP using Dynamic DNS when I do >host #thecurrentIPaddress# I get back the ##.pitt.east.verizon.net instead of #myserver#.no-ip.org. Could this be the possible issue?

When I go into /etc/hosts:

127.0.0.1 localhost.localdomain localhost
#.#.#.# #myserver#.no-ip.org mail
::1 localhost6.localdomain6 localhost6

and then run, I get:

[zimbra@XXXXX ~]$ zmcontrol start
Host #myserver#.no-ip.org
Starting ldap...Done.
FAILED
Failed to start slapd. Attempting debug start to determine error.
daemon: bind(7) failed errno=99 (Cannot assign requested address)
slap_open_listener: failed on ldap://#myserver#.no-ip.org:389

BUT

if I change /etc/hosts to:

127.0.0.1 #myserver#.no-ip.org localhost
#.#.#.# #myserver#.no-ip.org mail
::1 localhost6.localdomain6 localhost6

Then Zimbra starts fine and again I can send but not receive from the Internet.

It appears this is purely a DNS/Resolution issue and I can understand why the second /etc/hosts works to get zimbra to start and also why that will not work... but I do not have a solution. Any help is appreciated.

Thanks!

[killc]

I'm not sure why my last reply didn't appear, but here goes again:

When my /etc/hosts file is set to this:

127.0.0.1 server.no-ip.org localhost
#.#.#.# server.no-ip.org mail
::1 localhost6.localdomain6 localhost6

Zimbra starts and I can send and can connect via SMTP but get relay access denied.

when I change it to:

127.0.0.1 localhost.localdomain localhost
#.#.#.# server.no-ip.org mail
::1 localhost6.localdomain6 localhost6

Zimbra does not start:

[zimbra@server ~]$ zmcontrol start
Host server.no-ip.org
Starting ldap...Done.
FAILED
Failed to start slapd. Attempting debug start to determine error.
daemon: bind(7) failed errno=99 (Cannot assign requested address)
slap_open_listener: failed on ldap://server.no-ip.org:389


I believe the problem has to do with the fact that it is a dynamic IP and/or DNS, but I cannot figure out a solution. Any help is greatly appreciated.

Thanks!

[uxbod]

Is your server on a private IP behind a firewall/router ? If so you will need to setup a Wiki :: Split DNS. Please post the following

cat /etc/hosts
cat /etc/resolv.conf
dig yourdomain mx
dig yourdomain any
host `hostname` <- note backticks and not single quotes (copy and paste)

[killc]

No, that is why most of what I am finding in searches does not apply... or doesn't directly apply.

It is a public IP, standard 3Mb DSL service from Verizon. I use no-ip.org for dynamic dns handling and MX record.

[root@server ~]# cat /etc/hosts
# Do not remove the following line, or various programs
# that require network functionality will fail.
127.0.0.1               server.no-ip.org localhost
#correctexternalIP#          server.no-ip.org mail
::1             localhost6.localdomain6 localhost6


[root@server ~]# cat /etc/resolv.conf
; generated by /sbin/dhclient-script
search localdomain
nameserver 192.168.164.2
[root@server ~]# dig server.no-ip.org



[root@server ~]# dig server.no-ip.org mx

; <<>> DiG 9.3.4-P1 <<>> server.no-ip.org mx
;; global options:  printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 19359
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;server.no-ip.org.              IN      MX

;; ANSWER SECTION:
server.no-ip.org.       5       IN      MX      5 server.no-ip.org.

;; Query time: 2113 msec
;; SERVER: 192.168.164.2#53(192.168.164.2)
;; WHEN: Fri May 22 08:11:53 2009
;; MSG SIZE  rcvd: 50


[root@server ~]# dig server.no-ip.org any

; <<>> DiG 9.3.4-P1 <<>> server.no-ip.org any
;; global options:  printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 53467
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;server.no-ip.org.              IN      ANY

;; ANSWER SECTION:
server.no-ip.org.       5       IN      A       #correctexternalIP#
server.no-ip.org.       5       IN      MX      5 server.no-ip.org.

;; Query time: 2035 msec
;; SERVER: 192.168.164.2#53(192.168.164.2)
;; WHEN: Fri May 22 08:12:02 2009
;; MSG SIZE  rcvd: 66


[root@server ~]# host `hostname`
server.no-ip.org has address #correctexternalIP#
server.no-ip.org mail is handled by 5 server.no-ip.org.
[root@server ~]#

The other thing is that when I send mail to the server from the internet it does reach it, and does reach Zimbra because I can see it in the mail queue it is just that it shows as connection refused. So it seems like something with the internal routing/DNS that Zimbra uses is either not finding the actual delivery mailbox or something similar.

Thanks!

[uxbod]

Anything else listening on port 389 ? iptables and selinux disabled ?

[killc]

I don't believe so. iptables and selinux disabled. I followed the CentOS5 guide from these forums which has them disabled from the install.

[uxbod]

No close enough then Your /etc/hosts is wrong

127.0.0.1               server.no-ip.org localhost

it should look like

127.0.0.1	localhost.localdomain	localhost
::1	localhost6.localdomain6	localhost6
xxx.xxx.xxx.xxx	server.fq.dn	server

[killc]

I truly appreciate your assistance! But unfortunately this takes us full circle back to my initial post.

When I change /etc/hosts to:

# Do not remove the following line, or various programs
# that require network functionality will fail.
127.0.0.1               localhost.localdomain localhost
::1             localhost6.localdomain6 localhost6
#correctexternalIP#          server.no-ip.org server.no-ip.org

I get:

[zimbra@server ~]$ zmcontrol start
Host server.no-ip.org
        Starting ldap...Done.
FAILED
/opt/zimbra/bin/ldap: line 59: kill: (9369) - No such process
/opt/zimbra/bin/ldap: line 59: kill: (9369) - No such process
/opt/zimbra/bin/ldap: line 59: kill: (9369) - No such process
/opt/zimbra/bin/ldap: line 59: kill: (9369) - No such process
/opt/zimbra/bin/ldap: line 59: kill: (9369) - No such process
/opt/zimbra/bin/ldap: line 59: kill: (9369) - No such process
/opt/zimbra/bin/ldap: line 59: kill: (9369) - No such process
Failed to start slapd.  Attempting debug start to determine error.
daemon: bind(7) failed errno=99 (Cannot assign requested address)
slap_open_listener: failed on ldap://server.no-ip.org:389

It seems to me that this is logical because it is trying to resolve its own FQDN to the external IP address internally which will not work. I need Zimbra to internally resolve to localhost, not the FQDN.

Any help is truly appreciated.
Thanks!

[uxbod]

Is your server sat in a DMZ behind a firewall ? I see from your resolv.conf you are using a private IP for your nameserver. You should really setup a Wiki :: Split DNS architecture and either NAT or port forward on your FW.