Online traceroute

What the online traceroute measures

An online traceroute shows the path packets take from the IPVIEW origin nearest you to a destination, hop by hop, with the round-trip time to each hop. It runs from the server inside Iran for visitors there and from Germany otherwise, and the result says which. Each hop is probed once with a two-second wait, up to twenty hops; a name is resolved first and the address traced is on the page, and reserved addresses are refused.

The technique is to send packets with a time-to-live of 1, then 2, then 3, and record which router returns the expiry notice at each step. So the path shown is the one those probes took at that moment; routes change, load balancers spread packets across parallel links, and the return path, half of every round-trip time, is not shown at all. A traceroute is a sketch of the route, not a map.

How to read the result

The table has the hop number, the address of the router that answered and its time in milliseconds; an asterisk means that hop sent nothing back within two seconds. Addresses are numbers, not names: Reverse DNS on any of them says whose router it is, and WHOIS names the network. The raw output is under the table.

Three readings go wrong often. An asterisk at a middle hop while later hops answer is a router that does not reply to expired packets, not a break. One probe per hop means one slow reply is noise: a hop with a high time followed by low ones is a router slow to answer about itself, not a slow link. A real problem is a jump in time that persists for every hop after it, or a run of asterisks that never reaches the destination.

When it is useful

A server is slow or unreachable from Iran and the question is where along the way it goes wrong: at the last hop inside the country, at the international gateway, or on the far side. The hop where the times jump, or the last that answers before the asterisks begin, is the answer, and its network is one WHOIS away.

To compare the route from Iran with the route from Germany to the same host, by running it from each side, which shows whether traffic from Iran takes a long detour. For loss or jitter rather than a broken path, MTR repeats the measurement at every hop; for a quick check that the host answers at all, Ping is cheaper.

Common questions

What does the asterisk in a traceroute mean?

That hop sent no reply within two seconds. If later hops answer, the router is simply configured not to respond to expired packets and the path is fine. Only a run of asterisks that never reaches the destination indicates a break.

Why does one hop show a much higher time than the hops after it?

Routers answer about themselves at low priority, so the reply from that router was slow while packets passing through it were not. A slow link shows as a jump that stays for every later hop.

Traceroute or MTR?

Traceroute shows the path once, one probe per hop, and is quick. MTR probes every hop five times and reports loss and variation per hop, which is what you need when the problem is intermittent.

Is the traceroute run from inside Iran?

For visitors inside Iran, yes, from the IPVIEW server there; for visitors elsewhere, from Germany. The vantage point is written under the result and travels with a shared link.