Can you explain these tracert results?

Team Member Joined: 09 Jan 2002 Posts: 147 Location: UK

Hi,

In investigating U.S. based hosts for my next hosting provider I keep on getting long times like:

Code:

Tracing route to www.cagedtornado.com [140.99.17.3]
over a maximum of 30 hops:

1 22 ms 22 ms 21 ms 81-86-192-1.dsl.pipex.com [81.86.192.1]
2 22 ms 23 ms 22 ms 62-241-161-1.dsl.pipex.com [62.241.161.1]
3 22 ms 22 ms 23 ms POS5-0.GW1.LND9.ALTER.NET [146.188.63.129]
4 22 ms 22 ms 23 ms so-3-0-0.xr2.lnd9.alter.net [158.43.150.145]
5 23 ms 22 ms 24 ms so-1-1-0.TR1.LND9.Alter.Net [146.188.15.37]
6 93 ms 94 ms 94 ms so-2-0-0.IR2.NYC12.Alter.Net [146.188.8.178]
7 96 ms 94 ms 93 ms 0.so-0-0-0.IL2.NYC9.ALTER.NET [152.63.23.65]
8 172 ms 173 ms 172 ms 0.so-3-0-0.TL2.LAX2.ALTER.NET [152.63.29.109]
9 172 ms 171 ms 172 ms 0.so-0-2-0.XL2.LAX2.ALTER.NET [152.63.2.73]
10 171 ms 172 ms 172 ms POS4-0.XR2.LAX2.ALTER.NET [152.63.115.230]
11 183 ms 182 ms 182 ms 194.ATM6-0.GW3.PHX2.ALTER.NET [152.63.114.141]
12 184 ms 183 ms 182 ms puregig-oc12-gw.customer.ALTER.NET [157.130.230.146]
13 183 ms 183 ms 183 ms ge1-2-1000M.sw1.core.phx.puregig.net [140.99.96.70]
14 182 ms 183 ms 184 ms puregig-deru-1000M.puregig.net [140.99.96.118]
15 184 ms 185 ms 186 ms www.cagedtornado.com [140.99.17.3]

Trace complete.

So, it looked as if their network was slow. Then I tried Amazon.com:

Code:

Tracing route to amazon.com [207.171.182.16]
over a maximum of 30 hops:

1 23 ms 21 ms 22 ms 81-86-192-1.dsl.pipex.com [81.86.192.1]
2 21 ms 23 ms 22 ms 62-241-161-1.dsl.pipex.com [62.241.161.1]
3 22 ms 24 ms 23 ms POS4-0.GW1.LND9.ALTER.NET [146.188.63.105]
4 21 ms 22 ms 22 ms so-3-0-0.xr1.lnd9.alter.net [158.43.150.141]
5 24 ms 22 ms 22 ms so-0-1-0.TR2.LND9.Alter.Net [146.188.15.41]
6 96 ms 94 ms 94 ms so-2-0-0.IR2.NYC12.Alter.Net [146.188.8.178]
7 95 ms 94 ms 95 ms 0.so-0-0-0.IL2.NYC9.ALTER.NET [152.63.23.65]
8 168 ms 168 ms 168 ms 0.so-0-1-0.TL2.POR3.ALTER.NET [152.63.70.21]
9 174 ms 173 ms 173 ms 0.so-4-0-0.XL2.SEA1.ALTER.NET [152.63.145.234]
10 172 ms 174 ms 173 ms POS5-0.XR2.SEA1.ALTER.NET [152.63.106.234]
11 173 ms 172 ms 172 ms 194.ATM6-0.GW6.SEA1.ALTER.NET [152.63.106.189]
12 * * * Request timed out.
13 * * * Request timed out.
(continued all the way to 30)

So, am I doing something wrong in tracerting? Only very few U.S. hosts give me a quicker response:

Code:

Tracing route to www.fluidhosting.com [66.150.201.71]
over a maximum of 30 hops:

1 23 ms 21 ms 20 ms 81-86-192-1.dsl.pipex.com [81.86.192.1]
2 23 ms 22 ms 22 ms 62-241-161-1.dsl.pipex.com [62.241.161.1]
3 24 ms 22 ms 22 ms POS4-0.GW1.LND9.ALTER.NET [146.188.63.105]
4 21 ms 22 ms 21 ms so-3-0-0.xr2.lnd9.alter.net [158.43.150.145]
5 21 ms 29 ms 22 ms so-1-1-0.TR2.LND9.Alter.Net [146.188.15.45]
6 94 ms 95 ms 94 ms so-6-0-0.IR1.NYC12.Alter.Net [146.188.15.50]
7 95 ms 94 ms 93 ms 0.so-0-0-0.IL1.NYC9.ALTER.NET [152.63.23.57]
8 94 ms 94 ms 94 ms 0.so-3-0-0.TL1.NYC9.ALTER.NET [152.63.9.246]
9 101 ms 99 ms 102 ms 0.so-5-0-0.XL1.BOS4.ALTER.NET [152.63.16.125]
10 99 ms 100 ms 99 ms 0.so-7-0-0.XR1.BOS4.ALTER.NET [152.63.16.122]
11 99 ms 100 ms 100 ms 179.ATM6-0.GW5.BOS4.ALTER.NET [152.63.25.57]
12 101 ms 102 ms 102 ms internap-gw.customer.alter.net [65.194.79.74]
13 101 ms 101 ms 100 ms border7.ge1-2-bbnet2.bsn.pnap.net [63.251.128.76]
14 101 ms 102 ms 102 ms fluidhosting-1.border7-8.bsn.pnap.net [66.150.201.250]
15 102 ms 102 ms 102 ms www.fluidhosting.com [66.150.201.71]

Trace complete.

Ebay.com gave a response like Amazon.com, but Yahoo.com was like fluidhosting.com, with a max time of 105ms.

Can someone explain please?

Thanks,
Peter

Posted: Sun Sep 22, 2002 1:43 pm (21 years, 7 months ago)

Quote:

5 23 ms 22 ms 24 ms so-1-1-0.TR1.LND9.Alter.Net [146.188.15.37]
6 93 ms 94 ms 94 ms so-2-0-0.IR2.NYC12.Alter.Net [146.188.8.178]

The jump from 23ms to 93ms is the jump across the ocean. so-1-1-0.TR1.LND9.Alter.Net from its name looks to be located in London and so-2-0-0.IR2.NYC12.Alter.Net in New York. A traceroute just measures the time it takes for a router to reply to an ICMP packet so the futhur the router is away, the longer it takes to receive the responce.

Id imagine the increase in ping time between 0.so-0-0-0.IL2.NYC9.ALTER.NET and 0.so-3-0-0.TL2.LAX2.ALTER.NET can be explained in the same way, in that newyork is probably miles away from LA.

In general a traceroute isnt really much use in selecting a host as unless the route really is terrible then you wont generally notice much, server loads and other stuff affect things a lot more.

Also routing is an odd thing and while it may take a good route one day, a few hours later it might take a round the world trip to arrive at its destination.

As for the Amazon trace

Quote:

11 173 ms 172 ms 172 ms 194.ATM6-0.GW6.SEA1.ALTER.NET [152.63.106.189]
12 * * * Request timed out.
13 * * * Request timed out.
(continued all the way to 30)

If the site was up, this can be explained by a router being configured to reject ICMP(ping) traffic. Microsoft also reject ICMP so pinging/trace routing them will result it a load of timed outs.

Team Member Joined: 09 Jan 2002 Posts: 147 Location: UK

Thanks Ben.

So what's the best way to test a host (there must be some way to choose between them!)

Peter

Team Member Joined: 06 Jan 2002 Posts: 2564

Personally, I just ping them Mr. Green

Posted: Sun Sep 22, 2002 5:59 pm (21 years, 7 months ago)

There isn't really a way other than experience Sad

. Or search WHT (mod note: http://www.webhostingtalk.com/) for other peoples experience.