without knowing your configs, I can't say for sure, but in fooling with this
a bit after reading this post, I believe you may have run into a classful
issue on the default-network.

in a simple linear setup R5----R1-----R3 observe the R3 routing table:


 *   100.0.0.0/24 is subnetted, 1 subnets
D       100.1.25.0 [90/304128] via 172.10.15.5, 00:09:32, TokenRing0/0
     172.10.0.0/29 is subnetted, 1 subnets
C       172.10.15.0 is directly connected, TokenRing0/0
D*   199.1.1.0/24 [90/304128] via 172.10.15.5, 00:09:32, TokenRing0/0
     181.37.0.0/27 is subnetted, 1 subnets
C       181.37.13.0 is directly connected, Serial1/1
r1#

note that the 100.0.0.0 classfull network is flagged as a candidate default,
as is the classful 199.1.1.1 network. but the EIGRP route that has been
learned is a subnet of the 100 net - 100.1.25.0/24 to be precise.

now check the R3 routing table:


     100.0.0.0/24 is subnetted, 1 subnets
D       100.1.25.0 [90/20656128] via 181.37.13.1, 00:10:45, Serial1/1
     172.10.0.0/29 is subnetted, 1 subnets
D       172.10.15.0 [90/20528128] via 181.37.13.1, 00:10:45, Serial1/1
D*   199.1.1.0/24 [90/20656128] via 181.37.13.1, 00:10:45, Serial1/1
     181.37.0.0/27 is subnetted, 1 subnets
C       181.37.13.0 is directly connected, Serial1/1

note that the 100 network is not flagged, but the 199.1.1.0/24 network is.

as a sanity check, I added another router into the line, set up two
loopbacks - one using 135.35.1.1/24 and the other using 155.55.1.1/16
not from the following that there are two candidate defaults on R2: ( plus
the 199 advertised from the other end of the line. note the similar issue
with the classful versus non classfull breakdown of the two candidates.

     100.0.0.0/24 is subnetted, 1 subnets
D       100.1.25.0 [90/21168128] via 192.200.23.3, 00:02:09, Serial1/1
C*   155.55.0.0/16 is directly connected, Loopback102
     172.10.0.0/29 is subnetted, 1 subnets
D       172.10.15.0 [90/21040128] via 192.200.23.3, 00:02:09, Serial1/1
D*   199.1.1.0/24 [90/21168128] via 192.200.23.3, 00:02:09, Serial1/1
C    192.200.23.0/24 is directly connected, Serial1/1
     181.37.0.0/27 is subnetted, 1 subnets
D       181.37.13.0 [90/21024000] via 192.200.23.3, 00:02:09, Serial1/1
 *   135.35.0.0/24 is subnetted, 1 subnets
C       135.35.1.0 is directly connected, Loopback101
r2#


now check the R3 table again - note that the classful default only shows up:

     100.0.0.0/24 is subnetted, 1 subnets
D       100.1.25.0 [90/20656128] via 181.37.13.1, 00:42:38, Serial1/1
D*   155.55.0.0/16 [90/20640000] via 192.200.23.2, 00:03:40, Serial1/2
     172.10.0.0/29 is subnetted, 1 subnets
D       172.10.15.0 [90/20528128] via 181.37.13.1, 00:42:38, Serial1/1
D*   199.1.1.0/24 [90/20656128] via 181.37.13.1, 00:42:38, Serial1/1
C    192.200.23.0/24 is directly connected, Serial1/2
     181.37.0.0/27 is subnetted, 1 subnets
C       181.37.13.0 is directly connected, Serial1/1
     135.35.0.0/24 is subnetted, 1 subnets
D       135.35.1.0 [90/20640000] via 192.200.23.2, 00:04:05, Serial1/2
r3#

So to summarize ( so to speak ) EIGRP does indeed advertise the default
network throughout the domain. But there are still things to be aware of.

Chuck



""s vermill""  wrote in message
[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> Sorry to bring this up again but apparently, for at least some of us, it
> needs to be.  I chimed in the other day and offered a way to get EIGRP to
> distribute a default route.  That basically amounted to simply defining a
> static to 0.0.0.0 and redistributing static into EIGRP (whoever corrected
me
> by pointing out that the 'network 0.0.0.0' command isn't necessary, I
thank
> you (it was in BSCN)).
>
> Wayne jumped in and explained some problems with EIGRP and the 'ip
> default-network' command.  I thought it all sounded quite reasonable.  So
I
> tried this in the lab for several hours today (no "get a life" remarks
> please).  I must be really dense, because nothing I try works quite the
way
> (that I thought) it was described.  It seems that the only way to
> sucessfully use the default-network is to configure it on every single
> router in the AS (I tried this with RIP just as a sanity check and it
worked
> just fine).
>
> If that is the case, can a good argument be made in favor of this approach
> over redistributing static?  Or summarizing to 0.0.0.0 0.0.0.0 on an
> interface?  Both of those, at least, propogate throughout the AS after
> configuration on just one router.
>
> Thanks in advance,
>
> Scott




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=31814&t=31799
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to