I’ve been burned by using Docker’s networking directly too many times to count, especially in the context of Docker for Mac (where “the host” sometimes means “your computer”, while other times meaning “the VM Docker runs in”, arbitrarily.)
However, Kubernetes on Docker (microk8s or whatever it’s called) has always been extremely predictable in its (development-time, single-node) networking behaviour for me. Set up the right Deployment + Service + Ingress resources, ask kubectl(1) for the external IP and port to talk to, curl it—just works. Does the same externally-observable thing on your workstation that it does in prod.
Of course, that requires you to learn Kubernetes… which is a much bigger pain than it should be. But once you've got it, it's pretty simple/lightweight to wield Kubernetes at a problem; and the results are much more widely-applicable to everywhere you'd want to deply than e.g. Docker Compose is.
Docker for Mac's builtin Kubernetes cluster keeps getting better. These days (as of a year or so ago?) if you create a LoadBalancer Service it will wire up the port forward to your Mac's localhost, which is really nice. If you set up a local CA cert (minica makes this easy) and add an entry in your hosts file like "127.0.0.1 localhost.myhost.com", you can actually get full HTTPS to your development pod, using a production-like LoadBalancer Service setup.
I haven't tried microk8s, does it do full Service proxying to localhost? I did try Minikube a few years back and the Service proxying wasn't implemented yet.
I'm a big fan of fully replicating the production-like environment (including TLS) in your dev setup, at least for iterating on k8s-layer config changes; taking the cycle time for k8s changes down to seconds makes for a very pleasant development experience.
I've been using Minicube so far. Seems like Microk8s is a bit more lightweight, as it doesn't require a full vm, and you can just install it with a simple sudo snap install microk8s --classic without the need for Docker on your system.
That's very cool, will definitely give it a whirl!
However, Kubernetes on Docker (microk8s or whatever it’s called) has always been extremely predictable in its (development-time, single-node) networking behaviour for me. Set up the right Deployment + Service + Ingress resources, ask kubectl(1) for the external IP and port to talk to, curl it—just works. Does the same externally-observable thing on your workstation that it does in prod.
Of course, that requires you to learn Kubernetes… which is a much bigger pain than it should be. But once you've got it, it's pretty simple/lightweight to wield Kubernetes at a problem; and the results are much more widely-applicable to everywhere you'd want to deply than e.g. Docker Compose is.