Cloning in Parallels Gives Bad ETH0

So, I finally got to the point where I was pretty happy with my Ubuntu Server 8.1 image up to the point where I was going to start firing up Python virtenvs and add stuff to them.

The idea is to have a canonical ref, set up to just before the point of doing the actual Python stuff.

First step, obviously, is to clone off the reference VM to a new one and start mucking about with that.

So I cloned, and checked ifconfig and there was no ‘eth0′ interface. Which meant I couldn’t actually do anything with the clone.

Restarted, checked Parallels configuration etc. ad naseum, searched Google a hundred times, etc. etc.

Finally, I found the magic search term:

parallels cloned vm no eth0

First article mentioned exactly the problem I was experiencing. So I went and looked at my /etc/udev/rules.d/70-persistent-net.rules and sure as shit, the entry for eth0 on both VMs were identical. IOW, they were using the same MAC address and everything.

It’s a Debian thing. Here are the relevant articles:

http://forum.parallels.com/showthread.php?t=22204
http://communities.vmware.com/thread/83063

I edited the persistent-net.rules file a couple of times following the instructions but couldn’t get it to work. Since it’s a generated file anyway, I just deleted it, restarted the VM, and it picked up a new MAC address etc. and all was well.

Since I can’t do this easily in a startup file because it only has to be done the first time a VM is booted, it’ll just have to go into the checklist of things to do when cloning off a new VM until I either file a detailed bug report with Parallels and they fix it (never) or I script around it (equally unlikely).