Reverse proxy deep dive

(medium.com)

82 points | by miggy 4 days ago

8 comments

  • MortyWaves 20 hours ago
    Caddy, Nginx, Traefik seem to be the most popular reverse proxies in the self hosting/homelab communities.

    I definitely prefer Caddy in my experience, so far.

    • ethan_smith 12 hours ago
      HAProxy deserves a mention alongside those - it's particularly strong for high-traffic production environments where its advanced load balancing algorithms and detailed metrics shine.
      • p_ing 9 hours ago
        I would argue this is the best mainstream proxy. Even better when paired with OpenBSD and CARP.
    • somehnguy 8 hours ago
      Caddy has been excellent for me thus far as well. I'm using it on a VPS to reverse proxy to the services I run at home via a Tailscale tunnel. Coming from Nginx in the past Caddy was drop-dead simple to configure.

      The entire config for each vhost is 3 lines, including the domain definition and closing brace - and that includes TLS!

      • MortyWaves 6 hours ago
        Just curious if you have Caddy running in Docker or normal?
    • joshbaptiste 13 hours ago
      Trying out ferron recently as a reverse proxy https://www.ferronweb.org/.. config is super simple
    • lowwave 14 hours ago
      Is there a reverse proxies that can support DTLS support out of box without some kind experimental patch[1]?

      1: https://nginx.org/patches/dtls/

  • tdiff 11 hours ago
    Really looks like an ai-generated overview.
  • raincom 9 hours ago
    What's the difference between Reverse proxy and forward proxy? Is there something like "intermediate proxy"? Is this concept of L7 proxy, similar to DNAT/SNAT or Port forwarding in L3/L4?
  • philwelch 9 hours ago
    Original, Medium-free URL is https://startwithawhy.com/reverseproxy/2024/01/15/ReversePro...

    Meta request: can we change the URL to the original source? This isn’t quite blogspam (since it’s the same author reposting the same piece onto Medium) but Medium is annoying enough that I’d still rather resolve to the original source

  • vojtechrichter 14 hours ago
    Amazing read, I personally find it fascinating to make my own load balancer.
  • jeffbee 12 hours ago
    I would say the bullet points at the top are not strictly correct. The response does not necessarily transit the proxy. Responses can be returned directly to the client (DSR).
    • nyrikki 10 hours ago
      > Note: For simplicity, we’ll focus on Layer 7 (HTTP) reverse proxy.

      Layer 4 proxies are a very specific sometimes food that most people should actively avoid until they need it because of the tradeoffs.

      DSR is layer 4, and not in scope of this post.

      • jeffbee 8 hours ago
        Your comment, to me, only points out that the OSI layer model is nonsense. Envoy in DSR mode routes traffic based on application features, at "layer 7".
        • nyrikki 7 hours ago
          • jeffbee 6 hours ago
            That's fair. Of course that post also calls the OSI model "unfortunate" and "a poor approximation".
        • philwelch 6 hours ago
          The model itself isn’t nonsense because it’s not a model of load balancers; it’s a model of network protocols. Load balancers might handle multiple levels of the stack for the same traffic, but so does any other networked program, eg handling cross-domain redirects.
  • shelajev 4 days ago
    It took me an embarrassingly long time to internalize what the reverse proxy is. My brain got stuck on the fact that it is just proxying requests. What's so reverse about this? Silly.
    • happytoexplain 22 hours ago
      It's one of the classic cases of a thing being named relative to what came before it, rather than being named on its own merit. This makes sense to people working at the time the new thing is introduced, but is confusing to every other learner in the future.
      • raincom 9 hours ago
        What came before "reverse proxies"? Just curious to understand the history.
        • p_ing 9 hours ago
          Forward proxies, proxies where client machines were configured to route all their outbound traffic through (similar to a router). Usually performed caching back in the day when the Internet tube was slow, later on got SSL decryption capabilities and filtering lists to make sure you stay off of your naughty sites and so the proxy admin could decrypt your banking credentials.
      • nosianu 17 hours ago
        Could be worse. All the many things named after people prevalent in some fields more than in others, biology/medicine for example. When you read, for example, "loop of Henle" or "circle of Willis" you don't even know where to begin. You either know the term or not.
        • happytoexplain 11 hours ago
          True, though I think it's often a larger challenge to capture the intrinsic quality of a medicinal compound or physiological feature than a man-made tool.
    • rini17 4 days ago
      Since web proxy was originally used near clients, caching stuff to save precious bandwidth of their kbps-tier connection.
    • azaras 20 hours ago
      Nowadays, "reverse" is suppressed in most ways. I have heard that Nginx is a proxy more often than a reverse proxy.
      • Valodim 18 hours ago
        Except in the configuration where you use the reversep_proxy directive, of course
      • daveguy 13 hours ago
        How about service proxy vs web proxy rather than reverse proxy and proxy? Makes more clear that one is a proxy on the service side and the other is a proxy on the client side. Service proxy and Client proxy might be even better.
  • leptons 15 hours ago
    How does this relate to "AI"? /s