From: Huan Pham (Huan.Pham@peopletelecom.com.au)
Date: Mon Sep 22 2008 - 03:09:42 ART
Hi Paul,
You're correct about the behaviour of static route and network statement
in EIGRP! I did not know that EIGRP (and RIP as you said) treated Static
Routes that point to a connected interface similar as as Connected
Interfaces! Very interesting behaviour! I labbed up a simple scenario
and the behaviour is exactly as you said (although it is irrelevant to
the example in Doc CD, as the static routes in the Doc CD use next hop
IP address, instead of outgoing interfaces). However, I could not find
this behaviour documented anywhere yet. Could you pls give me any
pointer. Thanks.
Thanks again for bringing this behaviour into my attention. This is what
we benifit from taking part in discussion of the high quality mailing
list such as GroupStudy.
R2#sh ip int brief | ex unass
Interface IP-Address OK? Method Status
Protocol
Serial1/1 23.0.0.2 YES manual up
up
R2#sh run
router eigrp 100
network 10.10.10.0 0.0.0.255
network 23.0.0.2 0.0.0.0
no auto-summary
ip route 10.10.10.0 255.255.255.0 Null0
R3#sh ip route eigrp
10.0.0.0/24 is subnetted, 1 subnets
D 10.10.10.0 [90/2169856] via 23.0.0.2, 00:06:38, Serial1/3
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
paul cosgrove
Sent: Sunday, 21 September 2008 10:20 PM
To: Huan Pham
Cc: Gary Duncanson; Cisco certification; ccie.unnumbered@gmail.com
Subject: Re: Cisco DOCS confusion
Hi Huan,
You mentioned that 'network 0.0.0.0' with a static default route is a
special case, but actually the same behaviour is seen for any static
route pointing to a connected interface. Static routes pointing to
connected interfaces are treated like connected interfaces themselves by
EIGRP and RIP. If there is a matching network statement they will be
advertised. The default route is not treated any differently.
The example config from the cisco site is poorly explained, and has a
few errors. Difficult to know if it was intended to show filtering, or
an auto summarisation problem as Gary suggested. The text suggests to
me that the number of lines in the acl, as well as the wildcard masks on
the acl and network statement are all errors.
The example shows a config for a router which has two interfaces in
different major networks, 192.168.7.0 and 10.10.20.0. If auto
summarisation is enabled on the router being configured (though off by
default these days, it is not specifically disabled), then the
10.10.10.0 route will not be advertised to the 192.168.7.x neighbor,
instead (without the distribute-list) it will receive 10.0.0.0/8.
The distribute-list would stop this, so that only the two 192.168.x.x
redistributed routes are advertised. If you wanted all three advertised
you would have to disable auto summarisation.
Paul.
Huan Pham wrote:
> HI Gary,
>
> I do not have that book in front of me. The no auto-summary command is
to disable automatic summmarization of routes advertized to neighbors
at classfull boundary. This is not relevant to the wildcard mask
specified after the network statement.
>
> The IOS behaviour I see is that, if you do not specify the wildcard,
the router will convert it to the whole classfull network, and enables
EIGRP on all interfaces that belong to that classfull network.
>
> R1(config)#router ei 100
> R1(config-router)#no auto-summary
> R1(config-router)#net 150.1.1.0
> R1(config-router)#
>
> R1#sh run | s router
> router eigrp 100
> network 150.1.0.0
> no auto-summary
>
>
>
>
> --- On Sun, 9/21/08, Gary Duncanson <garyduncanson@btinternet.com>
wrote:
>
>
>> From: Gary Duncanson <garyduncanson@btinternet.com>
>> Subject: Re: Cisco DOCS confusion
>> To: "Huan Pham" <pnhuan@yahoo.com>
>> Cc: "Cisco certification" <ccielab@groupstudy.com>,
>> ccie.unnumbered@gmail.com
>> Date: Sunday, September 21, 2008, 4:53 AM You can also disable
>> automatic summarization in eigrp and get over the network wildcard
>> mask hassles to some extent.
>>
>> no auto-summary
>>
>> Doyle Vol I pp315-316
>>
>> I have seen a number of variants in practice lab examples.
>>
>>
>>
>> ----- Original Message -----
>> From: "Huan Pham" <pnhuan@yahoo.com>
>> To: <ccielab@groupstudy.com>; "CCIE
>> unnumbered" <ccie.unnumbered@gmail.com>
>> Sent: Saturday, September 20, 2008 11:12 AM
>> Subject: Re: Cisco DOCS confusion
>>
>>
>>
>>> This is a terrible example in the DOC CD. I cannot
>>>
>> understand why it
>>
>>> exists in the DOC CD!!! It is just a terriblly wrong
>>>
>> example, in my
>>
>>> opinion.
>>>
>>>
>>>> (1) I thought all three routes were advertised by "network" command
>>>> under router eigrp 1
>>>>
>>> No, you are not correct. All three routes are
>>>
>> advertized by the
>>
>>> redistribute static command.
>>>
>>> In the EIGRP, similar to all IGP, the network command
>>>
>> is to specify what
>>
>>> interface(s) take part in the routing protocol. It is
>>>
>> not to advertize the
>>
>>> subnets specify by that statement.
>>>
>>> EIGRP also requires you to put the wildcard mask. If
>>>
>> you do not, then
>>
>>> EIGRP a classfull network is specified.
>>>
>>> In the example
>>>
>>> router eigrp 1
>>> network 10.10.10.0
>>>
>>> is the same as
>>>
>>> router eigrp 1
>>> network 10.0.0.0
>>>
>>> or
>>>
>>> router eigrp 1
>>> network 10.0.0.0 0.255.255.255
>>>
>>> In EIGRP there's one special case where the
>>>
>> network statement also
>>
>>> advertizes the route, is when you do network 0.0.0.0
>>>
>> and you have a static
>>
>>> default route. In this case, the network 0.0.0.0
>>>
>> statement also enables
>>
>>> EIGRP on all interfaces as well.
>>>
>>> router eigrp 1
>>> network 0.0.0.0
>>> ip route 0.0.0.0 0.0.0.0 null0
>>>
>>> In the DOC CD examle, these network statments for
>>>
>> 192.168.0.0 and
>>
>>> 10.10.10.0 have nothing to do with the usage of
>>>
>> distribute-list to control
>>
>>> redistribution from Static to EIGRP!
>>>
>>>
>>>> (2) the access-list 3 permit 192.168.x.0 should
>>>>
>> have
>>
>>>> wildcard mask 0.0.0.255?
>>>>
>>> You are right. Doc CD is wrong.
>>>
>>>
>>>> (3) I thought all three routes were redistributed
>>>>
>> by the
>>
>>>> *redistribute
>>>> static* router configuration?
>>>>
>>> You are right. I do not know what they meant to
>>>
>> blocked here ;) Maybe they
>>
>>> mean "deny 10.10.10.0 0.0.0.255" to block
>>>
>> the last one (???!!!)
>>
>>>> (4) what does the *distribute-list 3 out static*
>>>>
>> command do
>>
>>>> in this case?
>>>>
>>> Nothing, unless they change the ACL 3 to block one of
>>>
>> the three subnets.
>>
>>> In that case, the one blocked by the ACL will not be
>>>
>> advertized into
>>
>>> EIGRP.
>>>
>>>
>>>
>>>
>>>
>>> --- On Sat, 9/20/08, CCIE unnumbered
>>>
>> <ccie.unnumbered@gmail.com> wrote:
>>
>>>> From: CCIE unnumbered
>>>>
>> <ccie.unnumbered@gmail.com>
>>
>>>> Subject: Cisco DOCS confusion
>>>> To: ccielab@groupstudy.com
>>>> Date: Saturday, September 20, 2008, 3:20 PM Dear Groupstudy
>>>> Experts:
>>>>
>>>> Please find a digest from
>>>>
>>>>
>> http://www.cisco.com/en/US/docs/ios/iproute/configuration/guide/irp_i
>> p_prot_indep_ps6350_TSD_Products_Configuration_Guide_Chapter.html#wp1
>> 057083
>>
>>>> in the end of this message, and help me explain
>>>>
>> the
>>
>>>> description of the
>>>> section. (1) I thought all three routes were
>>>>
>> advertised by
>>
>>>> "network"
>>>> command under router eigrp 1
>>>> (2) the access-list 3 permit 192.168.x.0 should
>>>>
>> have
>>
>>>> wildcard mask 0.0.0.255?
>>>> (3) I thought all three routes were redistributed
>>>>
>> by the
>>
>>>> *redistribute
>>>> static* router configuration?
>>>> (4) what does the *distribute-list 3 out static*
>>>>
>> command do
>>
>>>> in this case?
>>>>
>>>> Static Routing Redistribution Example
>>>>
>>>> In the example that follows, three static routes
>>>>
>> are
>>
>>>> specified, two of which
>>>> are to be advertised. The static routes are
>>>>
>> created by
>>
>>>> specifying the
>>>> *redistribute
>>>> static* router configuration command and then
>>>>
>> specifying an
>>
>>>> access list that
>>>> allows only those two networks to be passed to the
>>>>
>> EIGRP
>>
>>>> process. Any
>>>> redistributed static routes should be sourced by a
>>>>
>> single
>>
>>>> router to minimize
>>>> the likelihood of creating a routing loop.
>>>>
>>>> Router(config)# *ip route 192.168.2.0
>>>>
>> 255.255.255.0
>>
>>>> 192.168.7.65
>>>> *
>>>>
>>>> Router(config)# *ip route 192.168.5.0
>>>>
>> 255.255.255.0
>>
>>>> 192.168.7.65
>>>> *
>>>>
>>>> Router(config)# *ip route 10.10.10.0
>>>>
>> 255.255.255.0
>>
>>>> 10.20.1.2
>>>> *
>>>>
>>>> Router(config)# *!
>>>> *
>>>>
>>>> Router(config)# *access-list 3 permit 192.168.2.0
>>>> 0.0.255.255
>>>> *
>>>>
>>>> Router(config)# *access-list 3 permit 192.168.5.0
>>>> 0.0.255.255
>>>> *
>>>>
>>>> Router(config)# *access-list 3 permit 10.10.10.0
>>>>
>> 0.0.0.255
>>
>>>> *
>>>>
>>>> Router(config)# *!
>>>> *
>>>>
>>>> Router(config)# *router eigrp 1
>>>> *
>>>>
>>>> Router(config-router)# *network 192.168.0.0
>>>> *
>>>>
>>>> Router(config-router)# *network 10.10.10.0
>>>> *
>>>>
>>>> Router(config-router)# *redistribute static
>>>>
>> metric 10000
>>
>>>> 100 255 1 1500
>>>> *
>>>>
>>>> Router(config-router)# *distribute-list 3 out
>>>>
>> static
>>
>>>> *
>>>>
>>>>
>>>> Thanks,
>>>> --
>>>> CCIE unnumbered
>>>>
>>>>
>>>> Blogs and organic groups at http://www.ccie.net
>>>>
>>>>
>>>>
>> _____________________________________________________________________
>> __
>>
>>>> Subscription information may be found at:
>>>> http://www.groupstudy.com/list/CCIELab.html
>>>>
>>> Blogs and organic groups at http://www.ccie.net
>>>
>>>
>>>
>> _____________________________________________________________________
>> __
>>
>>> Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
>>>
>
>
> Blogs and organic groups at http://www.ccie.net
>
> ______________________________________________________________________
> _ Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Sat Oct 04 2008 - 09:26:19 ART