syepes/network_exporter — reverse-engineered prompt

Build me a small self hosted network monitoring exporter that Prometheus can scrape over HTTP. I want to define a list of targets in a config file and have it continuously check them using ping, traceroute style path checks, TCP port checks, and simple HTTP GET checks. It should work with both IPv4 and IPv6, let me add labels to targets, and automatically pick up config changes or DNS changes without interrupting everything already running.

Please make it practical for real operations use, with sensible logging, optional JSON logs, configurable DNS server and source IP, and controls so I can limit concurrency when monitoring lots of hosts. It should expose clear Prometheus metrics for status, latency, packet loss, hops, and connection results. A firewall friendly traceroute option over TCP would be great too, along with startup jitter so many instances do not all hit at once.

If it helps, include example config files and a simple way to run it locally or in Docker. You can check current Prometheus exporter docs online if needed.