What is traceroute actually telling you?
The path your packets take toward a destination, hop by hop — and it is genuinely useful for diagnosis, provided you know which parts of it to ignore.
How it works — a clever trick. Every IP packet carries a TTL (time to live) that each router decrements by one. When it reaches zero, the router discards the packet and sends back an ICMP Time Exceeded message revealing its own address.
Traceroute exploits this deliberately: send a packet with TTL 1, and the first router replies. Send TTL 2, and the second replies. Increment until the destination itself answers. Each response identifies one hop along the path.
How to read the output. Each line is one hop, usually showing three timing samples. Rising times with distance are normal.
What people misread, and this is most of the value:
Asterisks do not mean packet loss. Many routers are configured not to reply to TTL-expired packets, or to rate-limit those replies heavily. A row of asterisks in the middle of an otherwise working path is almost always a router declining to respond, not a failure.
High latency at one hop does not mean a problem there. Generating an ICMP reply is a low-priority task handled by a router's control plane, not its fast forwarding path. A busy router can be slow to reply while forwarding traffic perfectly. Only latency that persists for every subsequent hop indicates a real problem — a single spike that recovers is an artefact.
Loss at an intermediate hop is meaningless unless it continues downstream, for the same reason.
Return paths can differ. Traceroute measures round trips, and the return path may take an entirely different route, which is invisible in the output.
Load balancing can send successive probes along different paths, producing inconsistent hop lists.
The destination not responding does not mean it is down; many hosts drop ICMP.
What it is genuinely good for: seeing where a path leaves your ISP and enters transit; identifying geographically absurd routing; and — with MTR, which runs continuously — spotting sustained loss or latency that begins at a specific hop and persists to the end.