SLES 11 VNC losing connection

[dsmith1984uk]

I have a Suse Linux Enterprise Server 11.
There are 4 Windows Server 2008 Virtual Machines on it that I need to configure.
The problem is I remote into the Linux Server using UltraVNC then I load up one of the windows Server machines and begin configuring.
Suddenly I will lose connection through UltraVNC and be unable to reconnect.
If I go and physically visit the server I cannot get anything on the screen, no amount of key presses or mouse moving will solve this. It's almost as if the machine is off.

The only way to get it back is to hold in the power button until it shuts down then start it back up again.
One other detail that may or may not be important is that none of the windows server machines can join the domain. I'm setting this Linux server up exactly the same way I've done the previous 4 and none of them had problems joining the domain, I also never got this connection lose and machine freeze up.

Any idea's what might be causing it, it's like a complete machine lock up?
I'm willing to provide any addition information you guys need just tell me how to get it. (I'm a newbie with Linux)

[gogalthorp]

SLES is the commercial version You might want to check on the Novel boards.

Have you checked memory?

There should be a memory check on the install disk.

[dsmith1984uk]

Okay thanks I'll check the memory however I'm doubtful that it'll be the memory only because I can use the Linux machine all day long with no problems. It's only when I load up the virtual machines that I run into a problem.

But anyway I'm now going off to run a memory test.

[dsmith1984uk]

There are no memory errors

[dsmith1984uk]

I've checked the /var/log/messages
The last entry before the system freezes up is:

ifup: No configuratiion found for vif2.0

[gogalthorp]

Maybe a flaky network adapter?

As you said 4 other machines have had no problem and assuming more or less the same hardware. It almost has to be a hardware problem with just this one.

[dsmith1984uk]

The thing is I wouldn't expect a "flaky network adapter" to completely kill the machine. I mean it's as if the machine has been powered off.
You cannot get anything but a black screen the only evidence the machine is still on is the spinning fans and green light.

Also it only happens when I fire up the virtual machines. I can UltraVNC into the linux server and leave it for aslong as I want and when I return it's still connected.
If I turn on one of the virtual machines and go to lunch I know when I return I will have been disconnected and the linux server will be completely unresponsive.
In fact the last 3 times I've turned on virtuals the machine has froze within 5 minutes.

I think it would be best to ignore the subject I have given this thread. I do get disconnected but I actually think that is totally irrelevant to the problem. I'm obviously going to be disconnected if the server has crashed.

[dsmith1984uk]

I have once again resolved my own problem.

With the configuration I have there is a bond0 which contains eth0, eth1, eth2, eth3.
I had one network cable which was suppose to be plugged into eth0 (onboard NIC) but instead I had accidently plugged into eth2 (network card).
There shouldn't be a problem with this but there was and it crashed the whole server everytime I loaded the virtual machines.
I tested it plugged into eth3 and got the same results.

Not sure if it is faulty hardware (NIC card) or it's down to the bond configuration which is set as mode=active-backup miimon=100.