Still trying to tackle this issue. I've referred to the original config from the initial installation of the production server last year, and installed the VM machine in exactly the same way, after configuring splitdns correctly. I believe the ldap issue was related to a "zcsetservername" that was done to change the server name of the original box. So I followed the same steps - installed exactly as per the original, then changed the server name.
The ldaprestore then worked fine - all accounts were restored without error.
Then I got the following:
Code:
zimbra@mail:~$ zmrestoreoffline -sys -a all -c -br
mailboxd is not running.
zimbra@mail:~$ zmmailboxdctl start
Starting mailboxd...done.
zimbra@mail:~$ zmrestoreoffline -sys -a all -c -br
mailboxd is running.
Error: mailboxd still running. Stop mailboxd before running zmrestoreoffline.
Going back a step, I also got this error:
Code:
zimbra@mail:~$ zmconvertctl start
httpd: Could not reliably determine the server's fully qualified domain name, using 10.4.1.23 for ServerName
zimbra@mail:~$ nslookup 10.4.1.23
Server: 127.0.0.1
Address: 127.0.0.1#53
23.1.4.10.in-addr.arpa name = mail.shawnigan.ca.
dig -x 10.4.1.23
; <<>> DiG 9.4.2-P2 <<>> -x 10.4.1.23
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 61154
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 1, ADDITIONAL: 2
;; QUESTION SECTION:
;23.1.4.10.in-addr.arpa. IN PTR
;; ANSWER SECTION:
23.1.4.10.in-addr.arpa. 604800 IN PTR mail.shawnigan.ca.
The system is running bind to resolve A,MX,PTR queries correctly for the local machine, so this error doesn't make sense to me.
I've now opened a support ticket, as we can't wait any longer to sort this out. I need to see this upgrade from 5.0.18 to 6.0.5 complete in a testbed before I attempt it on a live system. I also would feel a *lot* better having some estimate of how long it will take to complete.
I'm waiting for a response to my ticket, and will post any relevant info here in case it helps others.
Thanks.