Traceroute Online
Trace the network path to any domain or IP. See every router hop, latency, and where packet loss occurs — in real time.
How to Read Traceroute Results
A traceroute result is a map of every router your traffic passes through to reach a destination. Each numbered line is one router hop — with its IP address, hostname (if resolvable), and round-trip time in milliseconds. Together, they paint a picture of both the path your packets take and where latency accumulates along the way.
Key things to look for:
- Latency spikes: A sudden jump between two consecutive hops — like 5ms to 180ms — indicates congestion or a slow link between those two routers. If subsequent hops return to normal latency, the spike may be artificial (the router deprioritizing ICMP replies) rather than real forwarding delay.
- Timeouts (*): A single timeout in the middle of an otherwise complete trace is usually a firewall dropping ICMP probes — not a real problem. A string of timeouts at the end means the destination is unreachable from that point on.
- Unexpected geographic hops: Traffic that crosses a continent before coming back indicates inefficient BGP routing or a misconfigured peering agreement between ISPs.
- Asymmetric routes: Traceroute only shows the forward path. The return path may be completely different, so a high-latency intermediate hop doesn't always mean round-trip latency for end users is that high.
Reading the output
Each line is one hop along the path, showing hop number, router hostname or IP, and three RTT samples. Packet loss is implied when you see * * * at a hop. Rising latency at a single hop and falling back to normal afterward usually means that router is ICMP-rate-limiting, not that the path is broken.
- Steady latency ramp: normal — each hop adds some transit time.
- Loss at the final hop only: the destination firewall is dropping ICMP; the service itself may still be fine over TCP.
- ASN transition: where latency usually jumps — the handoff between your ISP and the next carrier.
Common issues and fixes
- Intermediate
* * *with a working path: ICMP rate limiting, not a real outage — confirm with TCP traceroute on port 443. - High RTT at the ISP handoff: congested peering; escalate to your ISP with the specific hop IP and timestamp.
- Asymmetric routing: the forward path is not the return path; pair with a mtr from the other end before blaming a single hop.
- MPLS-hidden hops: providers sometimes tunnel hops invisibly; hop count can jump by several with no per-hop detail — this is expected.
- Unexpected geographic detour: a BGP misroute; report to your provider with the traceroute attached.
What to do next
Identify the hop where latency or loss first appears and share the traceroute with the responsible network (your ISP, the CDN, or the destination operator). For users far from your origin, consider a CDN POP closer to them.
For continuous monitoring, FlashMon.app tracks changes automatically and alerts you the moment something breaks.
Monitor latency continuously
Set up automated monitoring to get alerted the moment your site's response time degrades — before users notice.
Frequently Asked Questions
Traceroute is a network diagnostic command that maps the route packets take from a source to a destination. It works by sending packets with incrementally increasing TTL (Time to Live) values — each router along the path decrements the TTL by 1 and returns an ICMP Time Exceeded message when TTL reaches zero, revealing its IP address and latency. The result is an ordered list of every router (hop) between you and the destination, along with the round-trip time to each one. This makes it invaluable for pinpointing exactly where in the network chain delays or failures occur.
A timeout (shown as * * *) means the router at that hop either didn't respond within the time limit or is configured to drop ICMP packets — a common firewall policy for intermediate routers that don't want to expose their infrastructure. A single timeout hop in the middle of a route doesn't necessarily mean a problem: if subsequent hops respond normally, traffic is still flowing through that router, it's just silently dropping diagnostic probes. Only worry if you see timeouts from a specific hop onward, especially if the destination becomes unreachable.
This tool highlights hops with round-trip times above 100ms in amber. As a rough guide: hops within the same country are typically under 20ms, hops crossing a continent are 50–150ms, and trans-oceanic hops (e.g., US to Europe) are 80–150ms. A dramatic latency jump between two consecutive hops — say, 10ms to 200ms — points to congestion or a slow link between those routers. However, some routers artificially inflate ICMP response times as a low-priority operation, so a high-latency intermediate hop doesn't always mean the actual data path is slow.
Several reasons: (1) The destination firewall blocks ICMP, making it appear unreachable even though HTTP/TCP traffic gets through fine. (2) The route hits a maximum hop count (30 by default) before reaching the target — unusual but happens on misconfigured networks. (3) There is genuine packet loss or a routing loop somewhere along the path. (4) Load balancers or ECMP routing can cause asymmetric paths where return probes take a different route and appear lost. To confirm connectivity, compare the traceroute result with an HTTP or ping check.
Ping sends ICMP Echo Request packets directly to the destination and reports round-trip time — it only tells you if the endpoint is reachable and how long the full round-trip takes. Traceroute sends packets to each intermediate router sequentially, building a hop-by-hop map of the entire path. Use ping for a quick connectivity check. Use traceroute when you need to identify where in the network path a problem is occurring — whether it's a slow ISP backbone, a congested peering point, or an unresponsive server. FlashMon can monitor both automatically.
Modern networks use ECMP (Equal-Cost Multi-Path) routing, meaning different packets may take different physical paths to the same destination. Traceroute probes are individual packets, so they can each take a different route. This causes apparent out-of-order hops, or hops that seem geographically "backwards." Some routers also prioritize forwarding over replying to ICMP probes, so the displayed latency reflects probe handling priority, not actual forwarding delay. For consistent latency monitoring, FlashMon's continuous synthetic monitoring tracks real user-path performance over time rather than a single probe.