Traceroute Meets Ping
A single traceroute is a snapshot, and a single ping watches one point. The MTR tool combines them, repeatedly probing every hop on the path and updating a live table of loss and latency per hop.
Why The Combination Helps
Watching all hops over time reveals patterns a snapshot misses.
- It shows loss percentage at each hop, not just the destination.
- It shows average and worst latency per hop as samples accumulate.
- Because it runs continuously, it catches intermittent problems a one shot misses.
This makes it far easier to see where a path degrades rather than just that it does.
Reading Loss Carefully
A tricky part is that loss at a middle hop may be misleading. Some routers deprioritize the control replies that MTR depends on, so they report loss while still forwarding real traffic fine. The signal that matters is loss that persists to the destination and to every hop after a given point.
Key idea
MTR merges traceroute and ping into a live per hop table of loss and latency, and the meaningful signal is loss that persists to the destination rather than loss at a single middle hop that merely deprioritizes replies.