Re: Hard-coding Router-id

From: Paul Cosgrove (paul.cosgrove@heanet.ie)
Date: Fri May 09 2008 - 19:44:16 ART


Using Anycast RP might require you to manually specify router ids.

Paul.

Shine wrote:
> H Folks,
>
> I think, Huan (the original poster) knows the advantages of using the
> router-id on OSPF. His query was why anyone would/wouldn't use router-ids on
> BGP and EIGRP.
>
> -Shine
>
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> Ronnie Angello
> Sent: Saturday, 10 May 2008 12:21 AM
> To: Jason Madsen
> Cc: Amit Oberoi; Shine; Huan Pham; ccielab@groupstudy.com
> Subject: Re: Hard-coding Router-id
>
> If it doesn't say that you can't, then you can! It is definitely good
> practice to do so...
>
> On Fri, May 9, 2008 at 10:06 AM, Jason Madsen <madsen.jason@gmail.com>
> wrote:
>
>> Amit, great point about the addition of loopback interfaces. That could
>> most certainly screw up (change) router IDs. Does anyone know whether or
>> not we are allowed to manually specify router IDs if the lab instructions
>> don't specifically state to do so?
>>
>> I personally prefer to hardcode them, but I'm wondering if doing so when
>>
> not
>
>> specified would be a harmless "over-config" or if they'd consider it a
>> misconfiguration or something and possibly deduct points?
>>
>> Anyone with any experiences that can be talked about?
>>
>> Jason
>>
>> On Fri, May 9, 2008 at 7:55 AM, Amit Oberoi <amiobero@cisco.com> wrote:
>>
>>
>>> Using them you get more control over what you are trying to achieve.
>>> Virtual
>>> Links is a good example. If you analyze and practice the topologies you
>>>
> ll
>
>>> gradually realize it's saving and not wasting time coz you know what the
>>> router id on the neighbor would be instead of finding it out. They might
>>> change as well if you add lop back addresses and screw up the whole thing
>>>
>>>
>>> Cisco Systems, Inc.
>>> Amit Oberoi
>>> CISCO TAC Engineer-VPN
>>> Cisco Systems, Inc.
>>> Mon - Fri: 5.00 AM to 12.00 PM (PST)
>>> amiobero@cisco.com
>>> 1-212-329-2092 Ext. - 2127
>>>
>>> -----Original Message-----
>>> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
>>> Shine
>>> Sent: Friday, May 09, 2008 3:01 PM
>>> To: 'Huan Pham'; ccielab@groupstudy.com
>>> Subject: RE: Hard-coding Router-id
>>>
>>> BGP synchronization rule says the OSPF and BGP router IDs must match- not
>>> relevant if the synchronization is turned off.
>>>
>>> EIGRP router IDs are useful when redistributing.
>>>
>>> Other than these, I can't actually think of any other purpose of having
>>> router IDs, for these protocols.
>>>
>>> -Shine
>>>
>>> -----Original Message-----
>>> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
>>> Huan
>>> Pham
>>> Sent: Friday, 9 May 2008 7:19 PM
>>> To: ccielab@groupstudy.com
>>> Subject: Hard-coding Router-id
>>>
>>> GS,
>>>
>>> The workbook I use recommends to hardcode router-id for every routing
>>> protocol (OSPF, EIGRP, BGP), unless the scenario does not allow it. This
>>> practice is time consuming (if applied for all routers, and all routing
>>> protocols), and also error-prone as well, so I am reluctant to do it,
>>> unless
>>> I really need to.
>>>
>>> I do see the need for OSPF as virtual-links use router ID. If router-id
>>> change (e.g. after router reload) virtual-links will fail.
>>>
>>> However, I do not see any need for BGP and EIGRP. Can anyone tell me a
>>> common BGP or EIGRP scenario where we need to fix router-id? Thanks.
>>>
>>> Huan
>>>
>>>
>>> _______________________________________________________________________
>>> Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
>>>
>>>
>>> _______________________________________________________________________
>>> Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
>>>
>>>
>>> _______________________________________________________________________
>>> Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
>>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html



This archive was generated by hypermail 2.1.4 : Mon Jun 02 2008 - 06:59:16 ART