Traceroute Online (Internet Trace)

How to read your result

Each row of the table is one router on the way from the probe to the website. Hop 1 is next to the probe; the last hop should be the website itself. The times grow as the packets travel further, so a larger number further down the list is normal.

Under the table, a short summary points out what matters. These are the labels you may see:

The Network (ASN) column tells you whose routers you are passing through: your provider, a long-distance carrier, or the website's own network. When the name changes from your provider to the website's network within a few hops, the two networks connect directly, which usually gives a short, fast route.

How to read hops: stars, loss and jumps

A row of * * * (or "Request timed out" in Windows) means that router did not answer the trace. Many routers are set not to. If later hops answer normally, the path is fine. Worry only when every hop after a point shows stars.

MTR runs a traceroute again and again and shows loss and delay for every hop. Loss that starts at one hop and continues to the last hop points to a real problem. Loss at a middle hop only, with a clean last hop, is usually that router ignoring probes.

The same idea applies to delay. A single slow hop whose next neighbours are fast again is a router that answers traces slowly, not a slow route. A jump that every later hop keeps is where the extra time really comes from, and that is the hop our summary marks as "Delay starts".

If the trace stops early with stars until the end, the website or a firewall in front of it may simply ignore trace packets. Check whether the site opens in your browser before you decide it is down.

What affects the route and the delay

• Where the probe is. A probe in Mumbai and a probe in Delhi reach the same website over different routes. Pick the probe closest to you, or your own provider, to see a route like yours.
• Distance. Light in fibre needs time. A route that leaves India for Singapore or Europe adds a clear, lasting step in the round-trip time.
• How networks connect. Some providers hand traffic to a website's network inside India; others carry it abroad first. The ASN column shows where each hand-over happens.
• Which address the website gives. Big sites answer with a nearby server. A trace from Singapore and a trace from Mumbai to google.com can end at different servers.
• Routers that rate-limit replies. Routers answer trace packets with low priority, so a busy router can look slow or lossy while real traffic passes through it without trouble.

Trace back to your own IP address

"Trace back to my IP" runs the trace the other way round: from a probe elsewhere to your own public IP address. It shows the route into your provider's network, which is the direction your downloads, videos and game updates travel.

Compare it with a normal trace. If the path into your network goes abroad while the path out stays in India, a slow download from an Indian website can have a simple explanation. The last hops often stay silent, because home routers and mobile networks usually do not answer trace packets from outside. Your IP address becomes part of the public Globalping result when you use this button.

How to run tracert, traceroute or mtr yourself

A trace from your own computer shows your own Wi-Fi, router and provider, which no website can do for you. Run one of these, then paste the full output into "Check a trace from your own computer" above to get the same hop table and summary.

Windows

Open Command Prompt and type tracert -d google.com. The -d option skips name lookups, so it finishes faster. Windows tries up to 30 hops by default and waits 4 seconds for each reply before it prints a star (Microsoft's tracert documentation).

Mac and Linux

On a Mac or Linux, type traceroute google.com in Terminal. Each line is one router and shows three round-trip times. !H or !N after a time means a router reported the host or network as unreachable (the Linux traceroute manual).

MTR

The mtr manual describes it as a tool that "combines the functionality of the traceroute and ping programs". Install it from your Linux package manager, then run mtr --report google.com. Report mode sends a set number of rounds and prints one summary line per hop, which is the easiest format to paste here.

Stop a running command with Ctrl+C. Before you paste, you can remove anything you do not want to share; the analyser only needs the lines that start with hop numbers.

How this test works

A web page cannot send traceroute packets from your device, so the trace runs on a Globalping probe: a small volunteer-run server in the city or network you choose. The result names the probe's city, network and ASN. Every number in the table is measured from that probe, not from your computer.

"My internet provider" asks Cloudflare which network (ASN) your IP address belongs to and requests a probe on that ASN. If no probe on your network is online, the page tells you and runs the trace from a home or mobile connection in India instead. MTR is the default because it repeats each hop several times and reports the ASN of every router; plain traceroute is one pass.

Globalping's terms say "Measurement Data is public", so the website or IP address you trace can be seen by others. Do not trace private names or addresses. A trace back to your IP address makes your IP part of that public result. Globalping allows 250 free tests per hour for each IP address. Network names come from RIPE NCC RIPEstat. Read what each service sees and the full method.

Pasted output is read by this page in your browser and is never sent to our server. Only if you press Look up network owners are the public router addresses in it sent to RIPEstat to name the networks.