NSX Edge - HA vs FT

[marinov]

Hello, what will be the pros and cons of these two scenarios:

Scenario1: NSX Edge configured HA disabled (1 appliance) and on that 1 appliance is activated "Fault Tolerance". They are on different datastores and different hosts.

Scenario2: NSX Edge configured HA enabled (2 appliances). They are on different datastores and different hosts.

I cannot find info for scenario 1 and this makes me worried.

I tested the two scenarios. They both work, the 1 is working without interruption of the outside traffic which is good, but why nobody uses it?

Best regards

[Sreec · Accepted solution]

Scenario1: NSX Edge configured HA disabled (1 appliance) and on that 1 appliance is activated "Fault Tolerance". They are on different datastores and different hosts.

This is unsupported configuration. FT on Edges is not supported, you should ECMP enabled Edges for such use cases.

Scenario2: NSX Edge configured HA enabled (2 appliances). They are on different datastores and different hosts.

This is the supported way of achieving Active-Standby HA for NSX Edges.

[marinov]

Hello, thanks for your answer. Can you be more detailed about "unsupported"? What will go wrong?

I`ll take a look at the ECMP with two edges. Any good "how-to" for that?

[Sreec]

I'm sorry, i can't comment on what will go wrong when we are leveraging unsupported features/deployments . FT for workload VM's , don't do FT on Edges . Read more on Edge redundancy in the blog ->Talking Tech Series: VMware NSX Edge Scale Out with Equal-Cost Multi-Path Routing | Network Virtuali...