How to use
- Enter a domain (
example.com) or a public IP address under Address to trace. If you paste a URL, only the host is used. - Tick one or more regions. South Korea is selected by default, and each region you add gets its own table.
- Press Start traceroute. A real measurement probe in each region runs traceroute toward the target. It takes from a few seconds to a minute, and you can Stop it.
- In the hop table, read the hop number, address, hostname and the round-trip times for that hop. Hops that did not answer show
*, and a hop whose time rose by 50 ms or more over the previous answering hop is marked as a “big jump”. - The summary at the top shows path length, whether the target was reached, how many hops were silent and the biggest latency rise. Copy result gives you the text.
It does what tracert on Windows or traceroute on macOS and Linux does, except it runs from data centers in other countries instead of your own machine.
What traceroute shows
Every packet carries a TTL value that drops by one at each router. When it reaches zero, that router sends back a “time exceeded” reply. Traceroute sends packets with TTL 1, 2, 3 and so on, collects each router’s reply and draws the path. This tool follows up to 20 hops.
A real sample (from a South Korea probe to example.com):
| Hop | Address | Round trips (ms) |
|---|---|---|
| 1 | 140.91.230.6 (_gateway) | 0.171 / 0.143 |
| 2 | 139.150.100.233 | 2.369 / 2.361 |
| 3 | * | no reply |
| 4 | 141.101.82.9 | 3.376 / 3.408 |
| 5 | 104.20.23.154 (example.com) | 2.951 / 3.260 |
The target was reached in 5 hops. Hop 3 stayed silent but hops 4 and 5 answered, so nothing is wrong. The last hop being faster than hop 4 is not odd either, because intermediate routers often answer traceroute slowly.
Reading * and big jumps
A * is not necessarily an outage. Many routers are set not to answer traceroute, or answer with low priority. Use these rules:
- If hops after the
*answer and the target is reached, the path is fine. Tracing www.ucla.edu from Korea, hops 4 to 10 were all*, yet the target answered at hop 11. - If the last hops keep showing
*and the target is never reached, a firewall near the target probably blocks ICMP. Check whether its web server is open with the HTTP Header Viewer.
A big jump is a hop whose average time is at least 50 ms above the previous answering hop. In the www.ucla.edu example the time went from 2.57 ms at hop 3 to 114.3 ms at hop 11, a rise of about 111.7 ms. Since that path crosses from Korea to the United States, the physical length of undersea cables is the likely reason. A sudden rise inside one country would instead point to congestion or a detour on that segment. The 50 ms threshold is this tool’s own rule of thumb.
What is sent, and the limits
- The address you enter is sent to the measurement service Globalping, and the probe IP can be seen by the target and intermediate routers. See the privacy policy for details. Results are not stored.
- Private and internal addresses are not measured. There is an hourly request limit.
- This traces one address once per region. It is not meant for generating load or scanning networks, and it never repeats automatically.
- Paths change with time and load balancing, so two runs may differ. Hostnames of intermediate hops appear only when reverse DNS exists.
- For just the round-trip time to the destination, the Ping Test is quicker.
Frequently asked questions
Traceroute keeps showing * * *. Is my internet down?
No. Routers commonly block traceroute replies. If later hops answer or the target is reached at the end, the path is normal. If only * continues to the end, the target or a firewall on the way may be blocking it.
How do I see where it gets slow?
Compare the time of each hop and look for a sudden rise. It is a real bottleneck only if every hop after it is slow too. If one hop is slow and the next is fast again, that router is just answering traceroute slowly.
Can the hostname tell me which network it is?
Sometimes the hostname contains a carrier or city name, which is a clue. To be sure, look up the address owner with the WHOIS Lookup or the IP Geolocation tool. Locations are approximate and can be wrong.
Why does it differ from my Windows tracert?
Your machine and the probe start from different places and carriers, so the paths differ. Equipment may also apply different reply policies.
Where do I check DNS problems?
Use the DNS Lookup for records. Traceroute only shows the road after the name has been resolved.
The address you enter is sent to an outside measurement service (Globalping) and is never stored on this site’s servers.