VMs on vDS portgroups migrating to NSX-T overlay segments. How’re you doing it?

Given the scenario of a brownfield deployment and you need to retain the original IPs, also not running vSphere 7 so there’s a dedicated NVDS rather than vDS integration.

I figure there are a few ways, but none seem “graceful” since at some point the default gateway will need to move. That is of course unless there’s another method I haven’t thought of, which is likely the case.

1) Most direct path: Create overlay segments, T1, T0, peer with ToR, advertise routes, move over VMs using powerCLI, shutdown the SVIs on the physical ToR and hope the VMs arp for the new Mac of their default gateway.

2) Intermediary Setup: create equivalent VLAN backed segments in NSX-T, migrate VMs over to get them on the fabric and remove reliance on vDS (this is assuming not using vDS 7.0), then similar process, create overlay segments, T1s etc, move VMs and shutdown physical SVI

3) Slow migration: Build a Bridge Transport Node and bridge the overlay segments into the needed physical vlan. Migrate VMs over, then move gateway.

4) best case? Move VMs over to vm backed segment, slowly re-IP onto overlay segments as needed. But obviously doesn’t fall in line with IP retention.

Any other methods you’ve used?

View Reddit by TarzzanaView Source

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *