Need help figuring out why router configuration in a VM env only partially working


#1

Hi there…
Trying to figure out why the above setup is only partially working.

  • All of the above reside on a single physical switch.
  • ESXi 6.5 is hypervisor for the VM Host(s).
  • VyOS-1, VyOS-2 and VyOS-3 are VMs running ver 1.1.8 (installed from .iso)
  • All non-VyOS VMs are Windows 10 (from the same .iso)
  • The hosts within each VLAN (both physical and VM) are able to ping and RDP each other.

It seemed to me like a typical router (NAT) problem…

  • Two VLANs (100, 200) containing hosts with the same IP addresses (192.168.1.0/24) trying to access the a common host in the 3rd VLAN (700) - 192.168.100.0/24.
  • Created Router VMs with NAT (source NAT, Port Forwarding, Reflect) for each VLAN, and used 701 as the “WAN”.

Here’s what I mean by “partially working”:

When Vyos-2 is hosted in VM Host 2 as shown in picture, it cannot ping any public addresses (10.16.11.0) on VM Host 1. However, when I move it to be hosted in VM Host 1, all the three public IP are able to access each other.
I would like to understand this behavior.

I’m a newbie to networking, VyOS and also ESXi… so triple hex on me.
Appreciate your help and comments, eh!


#2

Thinks i would check:

  1. is the switch tagged/untagged for vlan 701 for the interface that VM HOST is using to connect to the switch. If it’s tagged then you’d need to set the 701 tag in the hyper-v configuration to match. If you untag that port, then you’d just leave the vlan out for that interface. I’d check that both vm hosts 1 and 2 have identical hyper-v switches configured with the correct vlans and that both ports on the switch are configured the same way.

#3

That was exactly the fix…
I forgot to make that change for VLAN 701 on the physical switch (HP Procurve).

Thank you!!


#4

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.