VMware Cloud Community
kpc
Contributor
Contributor

accessing different network

hi there, pretty new to esx and having following problem with migrating over existing VM's from another network, e.g

1) New ESX server installed with 2 NICs at present using IP address 10.192.5.50 plugged into physical switch on new 10.192.5 network. From the new server I can ping all 10.192.5 and 10.192.4 address's.

2) The old VM has an IP address 10.192.4.50 when I copy it over and import it into ESX the network no longer works using the 10.192.4 IP address, if i change this to something on the same IP range as the ESX server e.g. 10.192.5.51 then it works fine

I'm guessing I have to configure my vSwitch so it can also see 10.192.4 but can't seem to find where I do it. at present both adapters use (10.192.5.1-10.192.5.254) what I don't understand is why can I ping both ranges from ESX console.

Many Thanks for listening...

0 Kudos
7 Replies
wobbly1
Expert
Expert

The ESX console probably has a default gateway that it can access that allows it to ping the 10.192.4.x subnet whereas the VM will have a default gateway on the 4.x network that the physical nic can not access. You might need the switch port changing to a trunk port.

If possible change the VM's ip to a 5.x address and see if that then works. This may prove the routing problem. Also, try a trace route with the current address to see where it is failing

kpc
Contributor
Contributor

thanks wobbly, i did already try it with a 5.x address and that worked fine, and the traceroute didn't come back with anything. Maybe I need to assign a physical NIC to the 4.x address? Not sure.

0 Kudos
mbrkic
Hot Shot
Hot Shot

The routing in the service console and the connectivity of VMs are not related. Can you post output of esxcfg-vswitch -l so we can see what exactly you have going?

0 Kudos
wobbly1
Expert
Expert

I am assuming here that your physical nics connect to different switches. Get you network people to check the router that your vm's physical nic connects to to ensure that there is a valid route from the 5.x network to the 4.x network and vice versa.

0 Kudos
kpc
Contributor
Contributor

here's the output:

Switch Name Num Ports Used Ports Configured Ports Uplinks

vSwitch0 32 5 32 vmnic1,vmnic0

PortGroup Name Internal ID VLAN ID Used Ports Uplinks

VM Network portgroup1 0 1 vmnic0,vmnic1

Service Console portgroup0 0 1 vmnic0,vmnic1

Both NICs are in the same physical 5.x switch

ta

0 Kudos
mbrkic
Hot Shot
Hot Shot

With this configuration you are running both the 4.x and the 5.x IP subnets on the same physical segment. Since your SC works on 5.x that implies that the phycal segment is OK for that subnet.

How is the 4.x subnet connected to the 5.x subnet? Is it a separate VLAN connected through a router? If it is, then you will need to assign appropriate VLAN numbers on both your portgroups, and set-up the physical switch port to trunk both those VLANs.

The bottom line there has to be physical connectivity to any segment that you want to assign IP addresses from.

The thing to keep in mind is taht the routing in your SC has no bearing on the connectivity of the VMs. They have to be able to connect on their own, and have IPs that correspond to the segment that their portgroup is connected to.

Hope this helps. Let us know if you need more details/quetions.

0 Kudos
Mayur_Patel
Enthusiast
Enthusiast

SO, you will have to either put a physical nic on the 4.x subnet or set up the port at the physical switch as a trunk for the vlans for both the 5.x & 4.x subnets.

If you go the trunk route, you have to create additional port groups in your virtual switch for each vlan.

Mayur

0 Kudos