We’re having a few issues and I want to make sure that it’s not me mis-understanding what should be. version is 2.6.0 beta 2.
- We have 2 region controllers in 2 different racks, call them (regiond001 regiond002)
- We have 2 rack controllers in 2 different racks and different DNS domains (rackd001.cluster1 rackd002.cluster2)
Everything has a single interface, and no VLANs are in use so it’s all untagged.
The rack controllers control a single, unrelated rack, call it cluster3.example.com.
I created a fabric called cluster3-rack1, vlan untagged, subnet as 192.168.0.0/24.
The switch in rack1 has DHCP helpers pointed to both rack controllers.
- It doesn’t seem possible to assign HA DHCP servers on 2 different subnets with DHCP relay.
- On the VLAN page, the “rack controllers” section is empty and I can’t add a rack controller.
- The generated DHCP for the subnet is missing the next-server and doesn’t get sent to the client (confirmed with tcpdump)
So, on relay dhcp subnets, one rack controller is a single point of failure. I’m not sure, since the terminology of relaying DHCP requests from a VLAN to a VLAN doesn’t really make sense. I’ve only known DHCP helpers to point to a single IP, so it should be a rack controller (or set of rack controllers), not a VLAN.