Kevin,
Thanks for the information, I am aware that the traceroute might be
getting to the last hop, then get dropped, hence my request that people
who can actually see the website submit some traceroute information.
What I'm trying to do is find the point where a lot of people (myself
included) get misrouted, misdirected, or blocked such that any attempt
to view the PLUG website results in a timeout error, even though I can
see the site if I redirect the request through another site. In my
experience, that's usually a routing problem, one direction or another,
or a problem with the destination server. If other people (preferably
people NOT using COX) can put up some traceroute information that can
help me determine some common routing information for the destination,
then perhaps I can track the problem to a router, or to not being a
routing problem, in which case I can look into other possible causes.
What it comes down to is that I want to find out what's blocking some
people from the PLUG website, and either post a workaround, or get it fixed.
I appreciate the additional information you provided, and you're right,
in that it may not be a COX routing problem, I'd still like to see a
non-COX route to help me verify that. After that, I'll start sending
web requests with various bits modified to see if I can get one to go
through, and thus determine if there is a malformed request issue that
may need to be addressed.
==Joseph++
Kevin wrote:
>On Mon, 2005-02-14 at 17:32 -0700, Joseph Sinclair wrote:
>
>
>>As I've mentioned before, it is not possible to access the PLUG website
>>from much of the COX network due to a faulty route somewhere. I'm
>>trying to track down the point where the route goes bad in hopes of
>>getting the COX network staff off their collective hindquarters to
>>actually do something for a change.
>>
>>
>
>Hi Joseph,
>
>I am also using a Cox cable modem, but I have no problems reaching the
>plug website.
>
>I can't speak for the PLUG staff, but it appears to me that
>www.plug.phoenix.az.us resolves to 66.98.190.6 (as you indicated) which
>is IP space owned by a hosting company called "Everyone's Internet" out
>of Houston Texas (http://www.ev1.net). They appear to be hosting the
>website on Apache/Linux (good!) and using vhosts, which means it's
>probably not a dedicated server. Since they are using vhosts you can't
>surf directly to the resolved IP address. Apache is using the hostname
>requested in the URL to determine which site to serve up on the shared
>server. In fact, if you surf directly to http://66.98.190.6 you will be
>redirected to an SSL-encrypted site admin page (bad idea - IMHO!).
>
>As for traceroute falling down, well it's pretty common for hosting
>providers to drop icmp and udp traceroute packets. It's often a matter
>of survival for them. We can thank all the wormdows infestations for
>that. NOTE: your traceroute (and mine) fails at the last hop in
>Houston. Probably at ev1's front door. ;-)
>
>
>This is all guess work on my part, but here is a terminal log to base it
>on...
>
>
><SNIP>
>
>
(text/plain)