I’m looking at moving our production workload from AWS ECS + consul to fly.io. One of the things I can’t seem to figure out is how fly deals with internal load balancers. It looks like I can use consul’s DNS but then that forces me to update all of my services to constantly poll for updates so I’m not sending internal requests to unhealthy containers/VMs.
Is there a way to use the existing fly-proxy layer to route internal requests?
I know it’s possible for me to set up an entire Consul Connect service mesh on fly but I’m trying to keep the firecracker VMs as simple as possible (if I can).