
We've just installed our ESXi server, and first port of call was to move the Zimbra server over. However, the network doesn't seem to be getting picked up within Ubuntu.
Any ideas? I'm sure some of you have done this. BTW, it wasn't a physical box - it used to be a VMWare Server machine that I have just converted to ESXi

Just tried this and didn't work.
I've just converted a Windows machine and this worked as soon as I powered it on the first time, which tells me I have the networking set up correctly in ESXi.
I've checked /etc/network/interfaces and it's configured as it should be, but an ifconfig just show the local loopback or something?

@Hightower:
Did you build the Ubuntu server first as a VM then install Zimbra on this VM then import all the users mailboxes and accounts?
Have a word with Webman and see if he can help you on this score![]()

Ok, I've managed to get the machine replying to a ping - but I have an even bigger problem in that Zimbra won't start properly. ZMControl Status shows that mailbox is not starting since converting this machine to ESXi Infrastructure.

Looking at a few VMWare forum posts, they recommend using the VMWare Converter tool to import VMs from Server into ESXi. Give that a go (choose your ESXi host as both source and destination) and see what happens.

That's what I tried initially webman. VMWare Converter with source set to our VMWare Server (where the VM image is now) and destination set to the new ESXi server.
This took about an hour to complete but power on worked great. I then discovered that the new network port had been defined as eth1 instead of eth0 so I reset this (deleting the persistent file). This put the new network port to eth0 and I started getting a reply. But Zimbra still isn't working.

Ah right
Anything meaningful in the logs?
/var/log/messages, /opt/zimbra/log/mailbox.log or /opt/zimbra/log/zmmailboxd.out are good places to look.
You need dns and an mx record working before zimbra will start.

That's not the problem - Zimbra has been setup and working for the past 6 months. It's the conversion/moving to ESXi that's created the problem.
@Webman: I'm just in the process of converting again to make sure it wasn't a one off problem. This is gonna take a while so will check the logs in the morning if I have the same problems again.
Cheers,

The static address is fine and DNS isn't the problem as we can resolve the server address, and get a reply. Flip side and the server can resolve other hosts too and get a reply.
I think the problem lies within zimbra as I can SSH in no problem - it's just the mailbox service in zimbra which is failing to start.

Ok, got this working. I converted the machine from VMWare Server to ESXi using the VM conversion tool, and selecting the same options as I did the first time.
Then before I did anything else I powered on the machine and deleted /etc/udev/rules.d/70-persistent-net.rules
I then restarted the box immediately and logged in as Zimbra to check on status. All services up and I can log into webmail!
Chuffed!![]()

Whoops - powered the machine down again, and upped the available RAM to 8GB (was set to 1GB) and this stopped the Zimbra mailbox service from working (and I did this the first time too).
How can I up the memory to 8GB without breaking the box?
I've just changed it back to 1GB and it works again, but need to up it to 8 ideally.
There are currently 1 users browsing this thread. (0 members and 1 guests)