From: Baety Wayne SrA 18 CS/SCBX (Wayne.Baety@kadena.af.mil)
Date: Tue Oct 01 2002 - 01:24:31 GMT-3
Non 'ip route' method (relies on proxy-arp or auto summarization)
Choose multiple secondary networks that complete the /25.
Put the secondary addresses on an OSPF enabled interface. Make sure it's a
real interface involved with packet forwarding (i.e. not a loopback
interface) otherwise you will kitchen sink packets destined for the more
specific routes.
Turn Proxy-ARP on for all OSPF routers that could route the more specific
/26's and that are directly connected to the advertising OSPF router (Any
router advertising the /25 LSA) (This step can be ignored if you're dealing
with a classful routing protocol that auto summarizes and you use two
address ranges that cross a classful boundary).
Do not insert the /26's into your OSPF domain.
Inject the /26's into your RIP domain.
OSPF (/25)-----Any-------RIP (/26)
/26 OSPF <-- /26
/26 Router <-- /26
There are two problems with this. The first, is that you will have to
choose an ip address that is not being used anywhere else in your network to
use for the secondaries (the normal ip address conflict detection code will
not work). Whatever you choose would need to be heavily documented.
I tested this by using three routers.
Lo1 (/25:1)
|
|
A----------B---------C
/25:2 /26:3
/26:1a
/26:1b
Router A answered ARP packets for addresses within the range of /26:1a and
/26:1b because it had /25:1 in its routing table. Also, Router B summarized
the /26's into a /24 before advertising it to Router C since you will be
crossing a classful boundary. However, this technique will also work for
classless routing protocols even without relying on the auto summarization.
Because /26:1a and /26:1b was in the routing table Router B was able to
advertise them to Router C via RIP.
Router C was able to ping the Loopback on Router A (and get replies) because
/26:3 was in the OSPF domain and /26:1a and /26:1b was in the RIP domain.
The second problem with this method is that Router B will be ARPing for all
addresses in the /26:1a and /26:1b ranges and therefore would need to
increase it's arp cache size if the proxy addresses become significant in
number.
But since you are dealing with a classful protocol that automatically
summarizes subnets when crossing classful boundaries, you don't even need
the secondaries (just choose an addressing scheme that would fit the
results). This would probably be the answer that a testing proctor would
want you to come up with.
WAYNE A. BAETY, SRA, MCSE, USAF
18th Communications Squadron
DSN: 634-9927
Commercial: 011 81 611 734 9927
-----Original Message-----
From: Young K. Bae [mailto:ybae@cisco.com]
Sent: Saturday, September 28, 2002 9:51 AM
To: ccielab@groupstudy.com
Subject: What if OSPF routes have shorter masks than RIP?
Guys,
I'm sure this question has been asked many times before, but can someone
kindly explain to me again? There are some /25 routes in OSPF Area0 that I
need to redistribute into RIP v1. The serial link that connects R1 with R2
has a /26 mask. How can I inject /25 OSPF routes into a classful routing
domain in a case such as this?
R1 --- 172.16.12.0/26 (RIPv1) --- R2 ---- 172.16.100.0/29 (Area1) --- R5
172.16.200.0/25 (Area0)
Thanks in advance,
Young
This archive was generated by hypermail 2.1.4 : Tue Nov 05 2002 - 08:35:36 GMT-3