routing process of virtual lab's proxy appliance

[mcz]

Hi everybody,

I have created a proxy appliance by setting up a veeam virtual lab. I use this proxy appliance for additional testing purposes which I do "outside" of a normal SureBackup-Job.
For example if I wannt to test a new setting for our DC's I will do a Instant Recovery of our DC's, connect them to the test-Network (which is also assigned to the proxy appliance) and connect via RDP to those vm's using proxy appliance in between. Because of the setup made when I creating the virtual lab, the proxy appliance will redirect all traffic from my subnet 192.168.254.0 to the "internal" subnet 192.168.0.0 and of course, it will do a NATing.

Everything is working as expected but now it's getting excited: Sometimes there is an existing vm (lets say with ip 192.168.0.33) in my testlab, which will be needed in the future and has been suspended before. When I do a IR of the same source-vm and start it in my lab, proxy appliance won't route the traffic of 192.168.254.33 to 192.168.0.33. What I have to do then is to shut down the previous existing vm (which has been suspended before) and if I start the "new" vm, proxy appliance will route as expected.

Now the question is why proxy appliance acts the way it does. I suspect that this problem is related to the different mac-addresses both vm's have and proxy appliance does not update it's routing table between IP and MAC provided by ARP and therefore sends the layer-2-frame to the wrong recepient. Or is it maybe a configuration issue within vmware's virtual switch?

Thank you for your thoughts.

[PTide]

Hi,

So, you put the VM (192.168.0.33) from the isolated network into a suspended state, spin up (Instant Recovery) the very same VM from the backup into the same isolated network and expect ping to reach the new VM inside the isolated network by the same address (192.168.254.33) which you used to reach the suspended VM, and it does not work. However, if you shutdown the first VM and spin up another one then, everything works fine, is that correct?

Thanks

[mcz]

Yep, you understood that correctly.

[PTide]

Have you tried to ping the 192.168.254.33 after suspend but before spinning up the second VM? Also do both VMs have the same MAC?

Thanks

[mcz]

why should ping work when vm is suspended, what is your thought behind this idea? the vm's have different MAC's...

[PTide]

That was rather a glitch than an actual thought, sorry for confusion. What's the IP address type of the VM - DHCP or static?

Thanks

[mcz]

...it's a static IP...

[PTide]

At this point I'd recommend you to contact out support team directly as the described behaviour doesn't look as expected.

Thanks