Major issues ** Now Resolved **

Essendon

I can ping 192.168.100.1 from both machines.

server3 doesnt have a firewall running.

I have changed the preferred DNS to 192.168.100.129 (server1) for both computers.

Still no connectivity. One thing I have noticed is that, server3 is veryyyyy slow to do anything. Even right clicking something takes about two minutes for the options to open up. Wonder why this is.

I wasnt really ragging on VMware, just a lil fed up of it not working. I know I am doing something wrong. Sure enough, I know VMware is powerful and is fairly well used in my company.

dynamik

Actually, you'd want to check Server 1 for the firewall since that's the one that's not responding to pings, etc.

Have you given Server 3 adequate memory? If so, you may have other configuration issues outside of it's network settings that are causing problems. If you can, create a new VM and do a clean install and see how that goes.

Essendon

server3 has 384MB RAM, should be enough for a VM. There's hardly any traffic passing between the two servers.

there are no firewalls on either machine.

seems like another VM might be the easy ( ) way out.

Sie

MobilOne,

Are Server 1 and Server 3 hosted on seperate Virtual Machines??

If so then you'll need top bridge the network cards and ensure you connected the two with a routing device.

If they are on the same machine just put both on VMNet3 or something for the Network Interface so they share the same Virtual Network.

Another thing to bear in mind is if they are on different machines is that VMWare has its own DHCP service so if you have your own and this one it will cause all sorts of issues.

Personally I would:

Setup the Network interfaces as above

Trash the DNS Zone as with only two servers it wont take long to recreate, flush both the client DNS and Server DNS (dnscmd /clearcache).

Recreate the Primary Zone in Server 1

Recreate the Secondary Zone in Server 2

Essendon

Sie,

server1 is the actual physical host machine. server3 is the virtual machine.

Yupp, I am aware that VMware has its own DHCP service. I have disabled it for VMnet7.

Sie, please have a look the current configs for both servers and lemme know what should be modified (something is wrong here). I could waste both servers' DNS zones and recreate it, but there really isnt much point doing it before connectivity between the two is restored.

Sie

Hi MobilOne,

Looks like you not using the best VMnetx for your machine.

Off the top of my head VMnet0 is already setup for bridged connection on your physical network device. Have you tried using this?

Run vmnetcfg.exe and check under Host Virtual Network Mapping tab.

You say earlier your using VMnet7, whats that bridged to? It needs to be bridged to you current physical network card.

Found you this link if you want more info:

http://www.vmware.com/support/ws55/doc/ws_net_configurations_changing_bridged_windows.html

Its 2am here and im at work so I havent had time to really read each post word for word but from the sounds of it, its more your VM setup thats the issue rather than your DNS settings.

You could also convert the Physical Machine to a Virtual Machine by following the steps here:

http://www.petri.co.il/virtual_convert_physical_machines_to_virtual_machines_with_vmware_converter.htm

If you wanted to cut out the problems with bridging the Virtual & Physical NIC's.



Final Thought: if Server 1 or Server 3 isnt running DHCP you could turn the VM DHCP service back on or assign Static IPs.

dynamik

He doesn't have to bridge (but he can -- I don't think it would solve his problems though).

Host-only networking creates a connection between the host (physical) machine and the VM, so there shouldn't be any reason he can't communicate between machines because of that, unless they are on different physical machines.

Sie

Yes... you are right. May bad.

As i sais havent had time to read it all.

Top section here then:

http://www.vmware.com/support/ws55/doc/ws_net_advanced_ipaddress.html

What Network cards are being used on Server 1 (Physical) ?? Is it just the VMnet7??

Just thinking if theres more and routing isnt right etc the packets may be dropping off the wrong NIC to nowhere....

So make sure your VM NIC is host-only and network config and Physical Machine are using the same Virtual NIC.

I havent got it here but I think under properties of the Network Connections there a check box for Virtual Machine Network Services?

On the Host machine if ONLY this is ticked the host wont use the connection only Virtual Network will. If everything BUT this is ticked the host will use it but the Virtual Network will ignore it. Just a FYI.

Edit: Dont forget to flush all the DNS caches, then register each again to be on the safe side.

Essendon

Network cards being used on server1 are the physical card and VMnet7.

I have made sure that server1 and the VM are using the same virtual adapter i.e. VMnet7.

Sie, you pointed out the Virtual Bridge protocol thingy when you right-click on any network adapter on server1. I have set the physical adapter and the VM NIC to use the same virtual adapter, VMnet7.

Yesterday, there was absolutely no connectivity between the two servers, but after I did the above steps, server1 can ping server3 any way it wants. But, the other way still around doesnt work.

On the Host machine if ONLY this is ticked the host wont use the connection only Virtual Network will. If everything BUT this is ticked the host will use it but the Virtual Network will ignore it. Just a FYI.

Edit: Dont forget to flush all the DNS caches, then register each again to be on the safe side.

I have made sure of both things. I cleared the caches first thing when I first noticed this problem.

At work, my team-leader is a VCP, have talked him into having a look at my machine. I will have this sorted this thing tomorrow.

Any other suggestions, Sie, dynamik, others?

dynamik

MobilOne wrote:

server1 is the actual physical host machine. server3 is the virtual machine.

Um... That's kind of an important piece of information

In that case, all you'd have to do is set the NIC on the VM to the host-only option. That would be it. You shouldn't have to change anything on the host/physical machine aside from configuring an IP address, subnet mask, and primary DNS server. You don't have to deal with the virtual networks at all.

I still say install another Windows machine. That way you'll know whether it is something wrong with that specific VM or your network settings in VMware. Something like this is difficult to troubleshoot this way. It'd be nice if we could eliminate as many points of failure as possible.

Sie

dynamik wrote:

I still say install another Windows machine.

As I said above I would go with this, easiest way. Theres too many if's / but's and maybes to take into account.

I posted a link above on how to convert the physical machine to a virtual mahcine if you wanted or to save any agro with that start fresh.

Any specific reason your using the host machine as part of the lab? Low RAM?

Just a last thought on your current setup, you cant ping from 3 to 1 by name or ip at all??

Is the Host running Windows 2003? Sure you have no firewalls?? You havent setup and RRAS on this box either?

None the less I still say create a new VM for Server 1.

Essendon

Sie wrote:

dynamik wrote:

I still say install another Windows machine.

As I said above I would go with this, easiest way. Theres too many if's / but's and maybes to take into account.

I posted a link above on how to convert the physical machine to a virtual mahcine if you wanted or to save any agro with that start fresh.

Any specific reason your using the host machine as part of the lab? Low RAM?

Just a last thought on your current setup, you cant ping from 3 to 1 by name or ip at all??

Is the Host running Windows 2003? Sure you have no firewalls?? You havent setup and RRAS on this box either?

None the less I still say create a new VM for Server 1.

No specific reason for using the host machine as part of the lab. Plenty of RAM, 2GB. Dual core processor 2.66Ghz. Good enough for the home lab.

server 3 cannot ping server1 by name or by IP at all. server1 can ping server3 by name and by IP.

Host is running server 2003. No firewalls, I had spybot but that's swtiched off for now.

Never setup RRAS on this box ever, unless someone did some black magic on this!

Essendon

Well, I could not take the computer to be looked at by me team leader, there was too much on at work today.

I have tried a few things myself after having read a fair bit about other people's gripes about VMware on the internet.

Now, server1 (host machine) can ping server3 (VM) by name and IP. It also can now ping the default gateway at 192.168.100.1

server3 cannot ping server1 either by name/IP but it can now ping the default gateway at 192.168.100.1

The VM is bound to VMnet7. What I did was on server1, went to Network Connections > Local Area Connection > Properties > VMware Bridge Protocol. Set that to VMnet7, this was set to VMnet0 originally. In the VMware console, I chose Ethernet > Custom > VMnet7 (host-only).

Doing the above steps has let server1 ping server3. But the other way around still doesnt click.

Someone have any ideas why server3 and server1 both can ping the gateway but server3 cant ping server1. No firewalls operating (including windows firewall/ICS).

Essendon

I thought I might share how this was resolved, so that people running into similar problems are able to solve them quicker or perhaps avoid them altogether.

No offense to anyone here, but since I was following all of your advice and information from other forums/dicussions, I mixed up everything resulting in a mess that even my team-leader (who's a VCP, by the way) took over an hour to figure out and fix up!

I dont connect the server (DC) to the internet at all. When VMWare is setup it creates two virtual adapters by default, VMnet1 and VMnet8 on 192.168.121.0 and 192.168.2.0 subnets respectively. The DC's physical NIC had the address 192.168.100.129 assigned to it and the member server's virtual NIC had 192.168.100.130 assigned to it. Theoretically the two were on the same subnet and should communicate. But the member server's NIC was dropping off packets at the wrong NIC. So here's what was done to correct this.

On server1 (DC) > Network Connections > Local Area Connection was assigned the private address 172.16.192.1 on a completely different subnet (so that packets were not dropped off at the wrong NIC)
Again on server1 > Network Connections > VMnet1 > was assigned the address 192.168.121.1 with DHCP enabled on VMnet1

On server3 (member server) > Network Connections > VMnet1 was assigned 192.168.121.128 (which was getting its address from the inbuilt DHCP)

And hey, they can ping each other now. Zone transfers are going on, DNS's working. Everything's back to normal.

Here a bit of advice, follow it and you wont run into much trouble. Try and have an XP host and have the DC's, the member servers as VM's. Much easier configuration as VMWare takes care of everything itself. Just put the machines on the same VMnet and you should be good.

Thank you all for your posts!