How to propagate cable pull to VMs?

[thomastn24]

Hi All,

We have a multi-server esxi 6.7 U3 setup with 3 physical switches corresponding to 3 virtual switches on each server. If an entire physical LAN goes down, we are expecting this status to propagate to the VMs. However they are looping back on the vSwitch such that only VMs on the same server can talk to each other over said down physical network. We would like to be able to propagate this network failure so that the VMs isolated on each server cannot talk to each other when the physical network is not available. VMs communicating with each other on isolated systems is a security concern and may cause VMs to get out of sync, and we would like to avoid this possibility

Is there an easy way to bring a vNIC / VM interface down when an uplink failure is detected? Is there any way to 'turn off' isolated vSwitch functionality? Note, we do not have vCenter and use the VMware Host Client and esxcli instead

[Kamikadz3]

1. How many uplinks is connected to vSwitch? I guess more than one - correct?

2. If there are multiple uplinks - are they connected to ONE or MORE upstream switch?

[scott28tt]

VMs on the same port group and host talk directly with one another through the vSwitch whether the physical uplink NIC is active or not - their traffic never traverses the physical network.

So, you would need to have a network topology and configuration such that their traffic always has to traverse the physical network.

[scott28tt]

Your duplicate post has been reported to the moderators.

[thomastn24]

More than one uplink for each vSwitch yes, but never more than one upstream switch allocated to a vSwitch

[thomastn24]

Is there documentation  / logic for guaranteeing that network traffic over the physical network? Or do you mean via infrastructure and passing NICs through to VMs?

[thomastn24]

I'm sorry, I was not sure which section to post this question in. Should I delete one of the posts?