Mail Archives: cygwin/2005/12/11/13:22:36
Robert Body wrote:
> 9 32 ms 33 ms 33 ms 12.122.81.137
> 10 35 ms 33 ms 33 ms p5-0.core01.sjc03.atlas.cogentco.com
> [154.54.11.237]
> 11 35 ms 34 ms 35 ms p3-0.core01.sjc01.atlas.cogentco.com
> [154.54.1.29]
> 12 80 ms 78 ms 79 ms p14-0.core01.iah01.atlas.cogentco.com
> [66.28.4.237]
> 13 * 486 ms 517 ms p4-0.core01.dca01.atlas.cogentco.com
> [66.28.4.89]
> 14 643 ms * 487 ms p4-0.core01.phl01.atlas.cogentco.com
> [66.28.4.18]
> ...
> if 486ms starts at entry 13, but is still at 20, can I say that 20 is
> playing a part in the horrible delay?
From that output it would appear that there is some serious congestion
somewhere in Cogent's network that is causing your high latency. From
traceroute it's not always possible to tell exactly where or what is
going on, but it does seem clear that there is probably an overloaded or
misconfigured router, or a saturated link, etc.
One thing that you should also be aware of is that the windows version
of traceroute uses ICMP echo - the same kind of packet used by ping. In
fact it's really just doing a number of sequential pings, each with an
increased TTL. Anyway, ping is not always a good tool to judge network
congestion because it is not processed in the same way by routers as
normal IP flows. More here:
<http://www.iwl.com/Resources/Papers/icmp-echo_print.html>.
Finally, this isn't Cygwin specific so if you want to dig deeper you
should find a newsgroup or mailing list devoted to networking.
Brian
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -