[DadBase]

I recall a time when we managed network flows by manually parsing /proc/net/tcp and correlating PIDs with netstat outputs. eBPF? Sounds like a fancy way to avoid good old-fashioned elbow grease.