Ask HN: Are you using akka,actors,akka-cluster,akka-remote with HIGH scale?

How is it going with production high scale high availability, ease of troubleshooting, ease of other people such as devops, technical support understanding and troubleshooting the system? maintenance? is it just fun technology to work with or you actually have a large deployment for example with akka-cluster with real HIGH traffic volume, replacing a like 150 microservices frameworks? is it going smooth? any specific issues? i'm trying to understand if it's fun technology but for real high scale projects I should use standard bullet proof systems i used in past like microservices request-response with combination of load balancers, apps as microservices, kafka when I need replicated log message bus, and the rest of the well known standard stack. baking in my usual recovery and retry and backoff, Future/Promise, when almost every programmer knows and if not, it's not a whole shift in paradigm. one thing I can think of, with standard request-response model i can very easily troubleshoot my systems, for example when I have no logs (you never know what would happen at friday 2am when you have a problem in our case a problem and no logs to help you out because either someone forgot or your framework does not log for whatever reason) with request response - then any support engineer with http-text based request-response can just sniff the network and very easily understand the http protocol, i know which is the requesting microservices hostname (I can easily find out etcl..). with actors its a different story. This is just an example when I talk about supportability, production, high scale, just one.