I am going to throw this out here to see if anyone has seen this before and what the wise might think is a cause. First the symptom: User connectswith no difficulty. Connection established, user gets host errors. User cannot ping, bring up web pages, nada. Look up user's ip using list ip networks and find user's ip. Look at backbone router and the route is present for that IP. Ping from backbone router results in timeout. List ip routes on HyperARC shows route to ip as local and the same slot/mod as list ip networks. List ip routes on other HyperARC shows route as RIP with the correct router cards ip. Do a ping from the HyperARC that the modem is assigned to (local) and get timeout. Do a traceroute and get an UNKNOWN 0 response from the HyperARC Verified that this can happen on either HyperArc, on any card, at any time, with different users, any ip pool. Here is the confusing part. Rebooting the backbone router (flushing the route table) solves the problem for a few hours to a couple of days. This points to some kind of router problem or at least a routing protocol problem. The reason I am confused is that the route is *in* the backbone router. The route is in all the HyperARC's, the HyperARC's know what interface the IP is assigned to, at least by all the tables. But it is the HyperARC that can't ping. So how can it be a backbone router problem? a tcpdump on the backbone router shows the RIP requests comming in and the RIP responses going out. I haven't set up OSPF before and it has been a long time since I used static (although I can't see how that would be difficult). I am willing to abandon RIP and go either back to static or to OSPF if there is any thought this might help resolve the situation. And, as an important note, the backbone router software version, HyperARC version, and DSP version have all been unchanged since well before this problem occurred. And lastly, PLEASE HELP!! OK, I am sounding needy and not just a little whiney. I have worked this problem for about 4 days and can't seem to get my mind wrapped around it. Any ideas, questions, comments are very welcome. -- Lewis Bergman Texas Communications 4309 Maple St. Abilene, TX 79602-8044 325-691-3301 800-299-6962
First thing I would say is....create me a temp account to take a look. Most likely something very simple that you overlooked. :-) When this situation occurs....what happends when you try and trace from the backbone router? Does it make it to the arc and then die? If they are able to connect up then it was routing properly for them as they had to authenticate. Something changes after they connect up. Initially I was going to say it might be something stupid on the customer's end like they had some firewall software installed but since it happends to multiple customers I doubt that's it. I was also thinking something to do with MTU not being standard. Easier to troubleshoot when you can jump in and look at a few things. Todd ----- Original Message ----- From: "Lewis Bergman" <lbergman@wtxs.net> To: "'Discussion relating to the 3Com/US Robotics Total Control modemsystems.'" <usr-tc@mailman.xmission.com> Sent: Friday, October 29, 2004 10:09 AM Subject: [USR-TC] can't ping IP from HyperARC
I am going to throw this out here to see if anyone has seen this before and what the wise might think is a cause.
First the symptom: User connectswith no difficulty. Connection established, user gets host errors. User cannot ping, bring up web pages, nada.
Look up user's ip using list ip networks and find user's ip. Look at backbone router and the route is present for that IP. Ping from backbone router results in timeout. List ip routes on HyperARC shows route to ip as local and the same slot/mod as list ip networks. List ip routes on other HyperARC shows route as RIP with the correct router cards ip. Do a ping from the HyperARC that the modem is assigned to (local) and get timeout. Do a traceroute and get an UNKNOWN 0 response from the HyperARC
Verified that this can happen on either HyperArc, on any card, at any time, with different users, any ip pool.
Here is the confusing part. Rebooting the backbone router (flushing the route table) solves the problem for a few hours to a couple of days. This points to some kind of router problem or at least a routing protocol problem. The reason I am confused is that the route is *in* the backbone router. The route is in all the HyperARC's, the HyperARC's know what interface the IP is assigned to, at least by all the tables. But it is the HyperARC that can't ping. So how can it be a backbone router problem? a tcpdump on the backbone router shows the RIP requests comming in and the RIP responses going out.
I haven't set up OSPF before and it has been a long time since I used static (although I can't see how that would be difficult). I am willing to abandon RIP and go either back to static or to OSPF if there is any thought this might help resolve the situation.
And, as an important note, the backbone router software version, HyperARC version, and DSP version have all been unchanged since well before this problem occurred.
And lastly, PLEASE HELP!! OK, I am sounding needy and not just a little whiney. I have worked this problem for about 4 days and can't seem to get my mind wrapped around it. Any ideas, questions, comments are very welcome.
-- Lewis Bergman Texas Communications 4309 Maple St. Abilene, TX 79602-8044 325-691-3301 800-299-6962
_______________________________________________ USR-TC mailing list USR-TC@mailman.xmission.com http://mailman.xmission.com/cgi-bin/mailman/listinfo/usr-tc
participants (2)
-
Lewis Bergman -
Todd Bertolozzi