From: Alvarez, Rolando [NCSUS] (RAlvare5@NCSUS.JNJ.COM)
Date: Tue Aug 05 2003 - 18:36:08 GMT-3
Raj,
Clear the OSPF process.
RA
-----Original Message-----
From: ccie2be [mailto:ccie2be@nyc.rr.com]
Sent: Tuesday, August 05, 2003 5:30 PM
To: Group Study
Subject: iBGP peers over frame relay & ospf DR elections
Hi,
First of all, I'd like to thank everybody that responded to my earlier post
regarding the above issue. It's been very helpful.
It also made me take a closer look at OSPF over F/R where something quite
surprising occurred.
It's the same basic topology as before:
R1 ------- R2 ------- R3
spoke hub spoke
Each router starts out with their loopback addressed as follows:
192.168.x.x
where x= router #.
I left all the interfaces (both the loopbacks and serial interfaces) at
their
default ospf values for priority, ospf network interface type, cost, etc.
Because of the constraints in the practice lab I was doing, I was only
allowed
one frame relay map on R1 and R3 and I was only allowed to used the
specified
f/r dlci's which meant I had to disable inverse-arp on all the f/r serial
interfaces.
I wanted to see which loopback interfaces would show up in each router's
route table and how they would appear (as a host route and/or subnet).
The intial config's were like this:
R1 (R3's config is the same except that the addresses and dlci's were change
as appropriate.)
int s0
encap frame-relay
ip addr 172.16.100.1 255.255.255.0
fram map ip 172.16.100.2 112 broad
no frame inverse-arp
router ospf 1
net 192.168.1.0 0.0.0.255 area 0
net 172.16.100.0 0.0.0.255 area 0
R2
int s0
encap frame relay
ip addr 172.16.100.2 255.255.255.0
fram map ip 172.16.100.1 211 broad
fram map ip 172.16.100.3 233 broad
no fram inverse-arp
router ospf 1
net 172.16.100.0 0.0.0.255 area 0
net 192.168.2.0 0.0.0.255 area 0
The result:
R2 can ping both R1 and R3 but R1 can't ping R3 or vice versa.
The loopback interfaces of all 3 routers are in the route tables of all 3
routers but R1 and R3 can't ping each others loopback addresses.
On R2, the output of show ip os nei is
nei id Pri State
192.168.3.3 1 Full/Dr
192.168.1.1 1 Full/DRother
This result wasn't unexpected but I wondered what would happen if I forced
R2
to be the DR. So, I changed R2's loopback int to be 192.168.22.2/24 and
added
that subnet under the ospf process and cleared the ospf process.
Here's the surprise!!!
The output of show ip os nei stayed the same. Then I checked the output of
show ip os int s0 and it confirmed that R3 was still the designated router.
I
also tried using the router-id command under the ospf process on R2 and
cleared the ospf process again. (BTW, it takes over 3 minutes for the new
ospf adjacencies to form) Still, R3 remained the DR.
I also checked in Doyle book Routing TCP/IP vol 1 how the DR is elected.
And,
it confirmed that, essentially, the router with the router with the highest
router id becomes the DR if all the priorities are equal.
I also shut down int s0 on R2 then turned it back on. Still the same result
-
R3 remains the DR.
Why can't I force R2 to be the DR? And, why does R3 considers R2 the BDR
but
R1 considers R2 the DR.
Sorry for such a long post. Raj
This archive was generated by hypermail 2.1.4 : Tue Sep 02 2003 - 18:53:53 GMT-3