(PM) OSPF setup help (longish)

Andrea Campi (andrea@planet.it)
Tue, 9 Feb 1999 18:03:53 +0100

Hi!

I know, I know... from time to time, someone asks here for help on OSPF.
Nevertheless, I don't think you can just point me to the FAQ and dismiss me
easily...
What is different here is:

1- E1 PRI, which means I can't just start recycle email about US setups;
2- mixed Livingston - Cisco equipment

OK, so here is my network. Let's fake out the addresses... not that it would
make any difference...

10.10.0.0/22 is assigned to my ISP.
10.10.0.0/25 is the backbone subnet; on this subnet I have just 2 Cisco
routers, 3 PM3, 1 OR and a Linux box running gated.
10.10.1.0/24 is reserved for "assigned dialup addresses" for my customers.

I am currently running everything with OSPF, and it works, but I don't feel
it's being done in the best (read, most efficient) way. The reason why I say
so is that I initially set it up this way:

pm3-1: assigned 10.10.1.1 pool size 60
pm3-2: assigned 10.10.1.65 pool size 60
pm3-3: assigned 10.10.1.193 pool size 60

forget about the 10.10.1.128/26 for now...

The problem with this setup is that it generates very few aggregates!!! What
I'd like to have is:

pm3-1: assigned 10.10.1.0 pool size 64
pm3-2: assigned 10.10.1.64 pool size 64
pm3-3: assigned 10.10.1.192 pool size 64

But I'm having a few problems... Before going in too many details with my
problems, is there anything eveidently wrong in this? With this setup, is it
ok to use the .0 address? Is it ok to use the .255 address?

A couple of other issues I feel might be related... Every router in the area
is setup with a network range of 10.10.0.0/22... Is this ok, or would it be
better to only configure every router and NAT to only include the addresses
it sees directly? I mean:

pm3-1: 10.10.0.0/25 and 10.10.1.0/26
pm3-2: 10.10.0.0/25 and 10.10.1.64/26
pm3-3: 10.10.0.0/25 and 10.10.1.192/26

And more: I configured the cisco router connected to the backbone to inject
a default route in the OSPF area (it is an external area, 0.0.0.0). The
second cisco sees it and accepts it, while the PM3s don't seem to like it.
Why? To make this clear: my PM3s are now running with the default gateway
configured; I changed it to 0.0.0.0, then tried both reset ospf and
rebooting everything, but nothing happened.

OK, this has become too long already. Thanks a lot, I hope I didn't ask
anything stupid, but I read everything, and the documents on the livingston
website actually confused me more that they helped.

Thanks for any answer, bye,
Andrea

-
To unsubscribe, email 'majordomo@livingston.com' with
'unsubscribe portmaster-users' in the body of the message.
Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>