Upgrade M1 -> M2? .... Yes!!

[jholder]

Are you using nslookup '(insert_hostname)'
?

[agnOstos]

nope, just 'hostname'

[jholder]

Try running the following:
nslookup '(insert your hostname, not ip)'

It will lookup your hostname. You want to try and use DNS more than the hosts file, since that is what postfix prefers.

[agnOstos]

Ok, nslookup resolves for me so I'm guessing that's not one of my problems. I'm getting really frustrated. I had M1 installed and working with only a couple hitches. I tried the M1->M2 upgrade and it failed miserably, I wish I knew what was going on with ldap. Now my M2 install is having ldap errors too. What do I have to do to get Zimbra working again?

[jholder]

can you telnet to the port?
telnet localhost 389

If yes, then: Ldap is starting...and when zimbra is trying to start it, it's going to fail because it was running the whole time

If no, then: It's probably a PW issue.

[tim brennan]

Marc said to run the "/usr/bin/host" command with the output of the "/bin/hostname" command as a parameter.

As in:

% host `hostname`


...I am not enough of a linux person to tell you if host can ever give a different answer than nslookup -- but maybe it does, so I thought I'd point out the misread.

[marcmac]

Is ldap running - make sure it's not (via ps) - make sure there's no iptables stuff left around (as root, iptables -t nat -L ) this should show no port maps. (If it does, iptables -t nat -F will clear it).

If this is all clear, and telnet localhost 389 returns connection refused (NOT a timeout, but connection refused - if you get a timeout, you're probably running the redhat firewall)

su - zimbra
ldap start - did it start?
If no - edit conf/slapd.conf, remove the loglevel 0 line, restart, check for errors in zimbra.log.

If it did start - /opt/zimbra/libexec/zmldapinit - any errors?

[agnOstos]

Originally Posted by marcmac

Is ldap running - make sure it's not (via ps) - make sure there's no iptables stuff left around (as root, iptables -t nat -L ) this should show no port maps. (If it does, iptables -t nat -F will clear it).

If this is all clear, and telnet localhost 389 returns connection refused (NOT a timeout, but connection refused - if you get a timeout, you're probably running the redhat firewall)

su - zimbra
ldap start - did it start?
If no - edit conf/slapd.conf, remove the loglevel 0 line, restart, check for errors in zimbra.log.

If it did start - /opt/zimbra/libexec/zmldapinit - any errors?

Ok, I uninstalled everything except FC3.(no firewall running as I've had M1 working on it just fine) trying to run 'iptables' as root returns "command not found"
ran yum update
reinstalled M2 and set my host as 'zimbra' and domain as 'standaleinteriors.com.
set my admin password.
let the config fly and got even more of the errors than I had before.

ldap is not running and wasn't running before I reinstalled. I could not telnet to 389. I got connection refused.

I also checked to make sure I had no rogue zimbra processes running.

I'm attaching my install log.

I tried to start ldap manually and after running ps it doesn't show so I'm assuming it never starts. Not sure where zimbra.log is at the moment or I'd show you the results of that.

[marcmac]

zimbra.log is in /var/log

Does the hostname "zimbra" resolve to the local host?

[agnOstos]

'host zimbra' returns my wan ip