Hi,
I was almost sure I understood RIP a little. Until today, grrrrrr ;-)
My goal was to inject 172.16.200.0/25 OSPF routes into 172.16.12.0/26 RIP
domain
and I tried to use secondary addresses to extend /25 to RIP domain.
We have two routers connected via serial link on the edge of a network:
R1 (s0/0) ---- (s0/1) R2 (s0/0) ---- the rest of a network.
R1 is doing OSPF/RIP redistribution and R2 only in RIP domain:
router ospf 666
log-adjacency-changes
redistribute rip subnets
network 172.16.100.0 0.0.0.7 area 1
!
router rip
redistribute ospf 666 metric 6
passive-interface Serial0/0
network 172.16.0.0
distribute-list 1 out Serial0/1
!
access-list 101 permit ip 172.16.200.0 0.0.0.7 any
R1:
interface Serial0/1
ip address 172.16.66.1 255.255.255.128 secondary
ip address 172.16.12.2 255.255.255.192
no ip split-horizon
172.16.0.0/16 is variably subnetted, 4 subnets, 3 masks
O IA 172.16.200.0/25 [110/65] via 172.16.100.2, 02:29:30, Serial0/0
C 172.16.12.0/26 is directly connected, Serial0/1
C 172.16.100.0/29 is directly connected, Serial0/0
C 172.16.66.0/25 is directly connected, Serial0/1
R2:
interface Serial0/0
ip address 172.16.77.3 255.255.255.248 secondary
ip address 172.16.66.1 255.255.255.128 secondary
ip address 172.16.12.1 255.255.255.192
clockrate 64000
172.16.0.0/16 is variably subnetted, 3 subnets, 3 masks
C 172.16.12.0/26 is directly connected, Serial0/0
C 172.16.77.0/29 is directly connected, Serial0/0
C 172.16.66.0/25 is directly connected, Serial0/0
Now a RIPv1 update is received:
Mar 1 01:12:44.149: RIP: received v1 update from 172.16.66.1 on Serial0/0
*Mar 1 01:12:44.149: 172.16.200.0 in 6 hops
and the routing table looks now:
C 205.2.3.0/24 is directly connected, Loopback10
172.16.0.0/16 is variably subnetted, 4 subnets, 3 masks
R 172.16.200.0/29 [120/6] via 172.16.66.1, 00:00:15, Serial0/0
C 172.16.12.0/26 is directly connected, Serial0/0
C 172.16.77.0/29 is directly connected, Serial0/0
C 172.16.66.0/25 is directly connected, Serial0/0
We see that an update about 172.16.200.0 was received from 172.16.66.1
(secondary of serial interface of R1) and installed in route table. But the
netmask was chosen not as I expected: not /25 subnet locally configured on
s0/0
(172.16.66.2/25) has been chosen but longest-match rule was applied and /29
mask
configured on one of subinterfaces won.
This behaviour get me confused. Doyle vol.1 doesn't even mention of choosing
masks on receive (this is a great book but lacks of little-funny-details by
the
way), and even more detailed and full of algorithms Zinin's book did not
clear
this (p.325): "The route mask is determined as follows. If the network
reported
is the route belongs to the same major network as one of the interfaces
assigned
subnets - primary or secondary - the route mask is the same as the subnet
mask
of the interface's address. Otherwise, the classfull default address mask is
used."
I expected that the router would choose netmask belonging to the subnet of
update source (172.16.66.1/25 in this case). This would make more sense - or
maybe I am wrong??
Could anybody clarify this?
robert,
--
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=57049&t=57049
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
