Michael C. Toren on 20 Feb 2009 14:11:04 -0800 |
On Fri, Feb 20, 2009 at 07:19:36AM -0500, gabriel rosenkoetter wrote: > traceroute to 66.33.195.161 (66.33.195.161), 64 hops max, 40 byte packets [..] > 9 border21.po1-20g-bbnet1.lax.pnap.net (216.52.255.38) 223.945 ms 215.612 ms 206.996 ms > 10 * * newdream-8.border21.lax.pnap.net (216.52.220.146) 107.524 ms !X > 11 * * * > From a VPS I lease, 72.249.104.222, the routing gets directly to pnap > (rather than through alter.net), but has none of the packet loss that > popped up at Dreamhost's border routers in pnap for my home connection: You're right, the "!X" indicates that your traceroute probe packets are triggering an access-list filter on what appears to be Dreamhost's border router. If you've already ruled out that this isn't specific to the type of probe packet your traceroute program is using (e.g., by trying to telnet to 66.33.195.161 on port 80, or by using tcptraceroute), I don't have any good suggestions for you other than dropping them a line and asking why the ACL is in place. If you're a customer, or know a Dreamhost customer who can proxy your request, using the ticketing system rather than email would probably get their attention faster. I don't see any indications of packet loss on that link, though. Well behaved routers will rate-limit the number of ICMP responses they send in response to inbound packets. It looks like your traceroute program, combined with other activity on the network from other programs and sources, was just tripping this limit. -mct ___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug
|
|