From: steve.skinner@uk.pwc.com
Date: Mon Nov 01 2004 - 12:38:06 GMT-3
Hi ,
Many thanks for all your input
Adjacenies are required NOT to form across the link ..
as correctly stated ab Adebola a redistribute connected awith an distrib
list was the correct way to do this
regards
Stephen Skinner
AdebolaA@mtnnigeria
.net To: alexander.arsenyev@ericsson.com, Steve Skinner/UK/GTS/PwC@EMEA-UK,
Sent by: ccielab@groupstudy.com
nobody@groupstudy.c cc:
om Subject: RE: EIGRP without passive interface
26/10/2004 12:00
Please respond to
AdebolaA
Steve,
You would do well to clarify if there is a need within your lab/logical
topology for adjacency to be formed over this interface.
-----Original Message-----
From: Alexander Arsenyev (GU/ETL) [mailto:alexander.arsenyev@ericsson.com]
Sent: 26 October 2004 11:25
To: Adebola Adegbonmire [ MTN - UBA ]; steve.skinner@uk.pwc.com;
ccielab@groupstudy.com
Subject: RE: EIGRP without passive interface
The trick I thought of is:
1. Do "network x.x.x.x" under "router eigrp YY" on 3725 to include 3725' E0
into the process, to meet the req.
2. On 3725 drop locally-originated EIGRP packets having 3725 E0' source
address by appropriate "ip local policy route-map"
(send EIGRP packets having source address of 3725 E0 to Null0).
3. Build GRE tunnel between 3725 E0 and EIGRP neighbor.
4. Include GRE tunnell into the EIGRP process on both sides.
I don't have the rack on hand to test the above, so I leave it to someone
else.
HTH,
Cheers
Alex
-----Original Message-----
From: AdebolaA@mtnnigeria.net [mailto:AdebolaA@mtnnigeria.net]
Sent: 26 October 2004 11:10
To: Alexander Arsenyev (GU/ETL); steve.skinner@uk.pwc.com;
ccielab@groupstudy.com
Subject: RE: EIGRP without passive interface
Little catch is in Steve's post I see no requirement for forming
adjacencies
over the E0 interface. Just include the E0 interface subnet in the EIGRP
process. I.E it should be in the topology. I believe that is what the
requirement is asking for. Any thots.
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
Alexander Arsenyev (GU/ETL)
Sent: 26 October 2004 10:58
To: 'steve.skinner@uk.pwc.com'; Group Study
Subject: RE: EIGRP without passive interface
Hello,
How about EIGRP inside GRE tunnell?
Technically the packet wouldn't be an EIGRP (IP protocol 88) but rather GRE
packet (IP protocol 47).
Of course, You need to do this on both sides.
HTH,
Cheers
Alex
CCIE#13405/JNCIE#0107
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
steve.skinner@uk.pwc.com
Sent: 26 October 2004 10:31
To: Group Study
Subject: EIGRP without passive interface
Include 3725 E0 interface in the EIGRP process. Do not send out any EIGRP
hello packets out the Ethernet interface.
Accomplish this without using any passive-interface command.
.... iam sorry i don`t understand how this is possible ..
in order for eigrp to sync it needs to send hello packets to it`s
neigbhor...
i can use the neighbor command ,which will unicast ....but it STILL sends
hello`s...
if i nat ...all i do is change the type of packet .....it`s still a hello
packet
again i could turn it into a stub ...but it still has to send heelo`s to
communicate with it`s neigbhor...
AFAIK ....you CANT communicate between eigrp host WITHOUT hello`s
can someone please clarify...
question was posed by my SE....
many thanks in advance
Stephen Skinner
This archive was generated by hypermail 2.1.4 : Thu Dec 02 2004 - 06:57:37 GMT-3