Common NAT address on 2 VMs: 10.0.5.15

[socratis]

OK, from the manual, we have this:

In NAT mode, the guest network interface is assigned to the IPv4 range 10.0.x.0/24 by default where x corresponds to the instance of the NAT interface +2. So x is 2 when there is only one NAT instance active. In that case the guest is assigned to the address 10.0.2.15, the gateway is set to 10.0.2.2 and the name server can be found at 10.0.2.3.

Now, I'm not 100% sure what the "NAT instance active" means. I take it that if you have "x" active net cards (on one or more VMs) then "x" increases by every active card. So, 1st NAT => 10.0.3.z, 2nd NAT => 10.0.4.z, 3rd NAT => 10.0.5.z and so on.

In my situation, I have several VMs, but each and every instance of every machine that has a NAT interface gets assigned the same address: 10.0.5.15. Eg. from a WinXP machine and a ScientificLinux running in parallel, I get in Windows:

Ethernet adapter NAT:
        Connection-specific DNS Suffix  . : Home
        Description . . . . . . . . . . . : AMD PCNET Family PCI Ethernet Adapter #4
        Physical Address. . . . . . . . . : 08-00-27-E6-3A-35
        Dhcp Enabled. . . . . . . . . . . : Yes
        Autoconfiguration Enabled . . . . : Yes
        IP Address. . . . . . . . . . . . : 10.0.5.15
        Subnet Mask . . . . . . . . . . . : 255.255.255.0
        Default Gateway . . . . . . . . . : 10.0.5.2
        DHCP Server . . . . . . . . . . . : 10.0.5.2
        DNS Servers . . . . . . . . . . . : 10.0.0.1
        Lease Obtained. . . . . . . . . . : Tuesday, 05 November, 2013 18:27:20
        Lease Expires . . . . . . . . . . : Wednesday, 06 November, 2013 18:27:20

and in Linux:

eth3      Link encap:Ethernet  HWaddr 08:00:27:7E:DD:04  
          inet addr:10.0.5.15  Bcast:10.0.5.255  Mask:255.255.255.0
          inet6 addr: fe80::a00fe7e:dd04/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:40 errors:0 dropped:0 overruns:0 frame:0
          TX packets:54 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:12897 (12.5 KiB)  TX bytes:4686 (4.5 KiB)

What's going on? Are they supposed to have the same IP? Am I reading the manual wrong? Have I done something? I can post the machines config files, the VB config and any logs that anyone might ask.

The only reference I found on this was a reply by a user "vbox4me2" on 9/Aug/2010, but it's not very enlightening...

[BillG]

That is what I would expect. The original NAT in VirtualBox was designed to allow the client access to the host NIC and the network it is connected to by sharing the connection. It was not designed to allow the guest machines to have different IP addresses and be able to communicate with each other. That feature has now been added in 4.3 and it is called NAT network. This works more like a standard NAT router.

The NAT instance mentioned in your post relates to more than one NAT running on the same host (ie all guests are not using the same NAT interface on the host).

[socratis]

I said:

I take it that if you have "x" active net cards (on one or more VMs) then "x" increases by every active card.

and you complemented:

The NAT instance mentioned in your post relates to more than one NAT running on the same host (ie all guests are not using the same NAT interface on the host).

which makes sense. Note that the emphasis on my quote was my misunderstanding. When I setup a VM, I always assign 4 network cards, 1) HostOnly, 2) Internal, 3) Bridged, 4)NAT. That's why ALL of my VMs are assigned the 10.0.5.15 address. I did a test and assigned all 4 to NAT. Understandably and predictably they were 1) 10.0.2.15, 2) 10.0.3.15, 3) 10.0.4.15, 4) 10.0.5.15. It all makes sense now. I think I need to check the "NAT Network" option. Thanks again!