> You need to play with traceroute. The routing on your LAN isn't working
> properly. You might have RIP on some box accidentally, or the OSPF group
> isn't communicating fully, or something like that.
>
> Tracerouting to and from the customer machine should show you where the
> routing breaks down.
>
The routing breaks down on the PMx (PM2/PM3) *ONLY* when trying to
ping/traceroute from one /24 to another /24 ON THE SAME BOX. The first
/24 is part of the pool, while the second /24 is only used for
ISDN/dedicated customers with multiple machines on a lan (subnet
customers). The routing dies AT the PMx (I get one hop in the
traceroute). IF I telnet to the PM3 (for instance) which has an Ethernet
address in the same /24 as the pool (but NOT the same subnet as the pool),
I can ping the other /24 just fine. When I set up OSPF on the PM3, I
specifically made sure that area range commands covered each /24 that the
PM3 would see locally. Everything is in area 0.
I just ran some more tests and discovered that the problem exists ONLY
with machines in the first /24 (206.168.144.X) trying to ping/trace
machines in the second /24 (207.174.113.X). If I assign myself an address
from the second /24 (or any of the other /24's we have) I can ping/trace
all IP's fine. Since the "problem" /24 is the one with the dynamic pool
AND the Ether0 interface I suspect a foobar in the way I set up
"areas"...the PM3 must do something "special" with the /24 used by its
Ether0 (even though, like I said, I included specific "area 0 range ..."
commands on the PM3 for each /24).
> Once upon a time John-David Childs shaped the electrons to say...
> >4 /24's ("class C's"), two contiguous, two non-contiguous.
> >OSPF on all Portmasters and Cisco border router, no RIP redistribution.
> >
> >PM3 running OSPF with a dynamic poolsize of 46 from one /24 ("class C"),
> >also serving subnet customers (of many different sizes) using a different
> >/24.
> >
> >If a customer logs into PM3 with an IP from the pool, they cannot connect
> >to a subnet customer on the other /24. However, if the same customer logs
> >into ANY OTHER PM (getting a dynamic IP from the same /24 that PM3 hands
> >out dynamically), they can reach the subnet customer(s).
>
> >So, the question of the day is: do I have to enable RIP on ether0 of the
> >PM3 and then enable RIP broadcasts to the customers? I intended that each
>
> No. This is not a customer issue, and you really DO NOT want to run
> routing over a dial link.
I didn't think so...but I am grasping at thin, cold, smoggy air :-)
>
> -MZ
> --
> Lucent Remote Access Division - Chair, Department of Interstitial Affairs
> Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com
> For support requests: support@livingston.com <http://www.livingston.com/>
> Snail mail: 4464 Willow Road, Pleasanton, CA 94588
> -
> To unsubscribe, email 'majordomo@livingston.com' with
> 'unsubscribe portmaster-users' in the body of the message.
-- John-David Childs (JC612) Enterprise Internet Solutions System Administrator @denver.net/Internet-Coach/@ronan.net & Network Engineer 1039 S. Parker Rd. #I-8 Denver, CO 80231 As of this^H^H^H^H next week, passwords will be entered in Morse code. - To unsubscribe, email 'majordomo@livingston.com' with 'unsubscribe portmaster-users' in the body of the message.