1    <1 ms    <1 ms     3 ms
  2    75 ms    71 ms    73 ms  reserve.cableplus.com.cn []
  3    61 ms   133 ms   140 ms

What does 75 ms 71 ms 73 ms mean there?

  • there are many such explanations on the web. see exit109.com/~jeremy/news/providers/traceroute.html#reading for one. – Anonymous Feb 3 '10 at 10:10
  • Isn't that site for server configurations? – Anonymous Feb 3 '10 at 10:10
  • I've provided an answer, but I think this question doesn't really belong here, it's really a sysadmin question, as traceroute is an network / system administrator tool NOT an API or developer tool. – Iain Collins Feb 3 '10 at 10:26

From the man page of traceroute:

This program attempts to trace the route an IP packet would follow to some internet host by launching UDP probe packets with a small ttl (time to live) then listening for an ICMP "time exceeded" reply from a gateway. We start our probes with a ttl of one and increase by one until we get an ICMP "port unreachable" (which means we got to "host") or hit a max (which defaults to 30 hops & can be changed with the -m flag). Three probes (change with -q flag) are sent at each ttl setting and a line is printed showing the ttl, address of the gateway and round trip time of each probe.

So, 75 ms 71 ms 73 ms stands for the round trip time it took for each different probe.

| improve this answer | |

Each entry is the RTT (round-trip time) for a particular probe to a particular host. I.E., the RTT times for the second hop host (reserve.cableplus.com.cn) were 75 ms, 71 ms, and 73 ms for the first, second, and third probes respectively.

On Linux, you can change the number of queries/probes to each host with -q.

| improve this answer | |

They are each the result of three separate estimates for ICMP packet Round Trip Time (RTT) between your computer and the host listed. (ICMP packets are just a specialised type of IP packet, specifically for this sort of thing).

The default in most traceroute implementations is three separate tests, but you can typically change this via a command line flag.

To give some reference, your local gateway (or system on your local network) is typically going to have a value of between 1ms and 5ms. If you are on cable or DSL you can expect ping times anywhere between 15-30 ms, and to hosts on the internet from a DSL/cable connection typically a total time of between between 30-90 ms. The further away a host is, the higher the RTT value (from the US to the EU it's likely to be around 90ms or higher, for example).

To give some context: For online gaming, for example, you would want to have a value of < 90ms, ideally <60ms if you are playing a fast paced action game (and <30 ms if you can get it), because for games you would want the network to be as responsive as possible. For things like web browsing and watching videos on line, things like over all bandwidth and connection quality (e.g. little or no packet loss) are more important.

It can also be high if the line is congested (e.g. being utilised heavily, which causes IP packets to be queued on routers) or if one of the devices at either end is for some other reason having trouble passing your packets along. Satellite connections have extremely high values (> 200 ms) because of the distance data has to travel (e.g. they can be high bandwith, so they transfer data quickly, but they are not very "responsive" - they have high RTT times).

In your example, the ping time to your provider cableplus.com.cn - which the first result after (your local router/gateway) - looks a bit on the high side.

| improve this answer | |
  • You are correct about RTT. It should also be pointed out that the overall traceroute can be longer for person A even if the end server is physically closer to person A than to person B. Routers are going to make these choices based on route cost and router settings. – Patrick R Feb 3 '10 at 13:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy