From: frank wells (fwells12@xxxxxxxxxxx)
Date: Mon Jan 22 2001 - 17:58:43 GMT-3
<html><DIV>
<P>Exactly. My thought was when addressing your network, you should think
about how you want to address loopback interfaces before ever applying any ip
addresses to any interfaces. OSPF's rule that uses highest loopback in th
e event there is one, should make us use the highest ip address we can from the
block of addresses we decided to use on that router. By assigning i
t to a loopback, rather than physical interface, we can be assured of
our RID's and they will not change with a reboot etc. It
would not be wise to arbitrarily put ip addresses on loopbacks without thinking
about their role in summarization too. </P>
<P>I just realized why assigning loopbacks first is the way to go though:
The neighbor relationships will be made using those addresses, and would not c
hange with the later introduction of physical interface addressing, and reboots
etc. This is opposite to my initial thinking on this. </P>
<P>Ronnie was right on the money.</P>
<P> </P>
<P> </P></DIV>
<DIV></DIV>
<DIV></DIV>>From: Kyle Galusha <KGALUSHA@CISCO.COM>
<DIV></DIV>>To: "frank wells" <FWELLS12@HOTMAIL.COM>
<DIV></DIV>>Subject: RE: IP allocation matrix question?
<DIV></DIV>>Date: Mon, 22 Jan 2001 12:57:01 -0600
<DIV></DIV>>
<DIV></DIV>>Frank the highest loopback will be the rid regardless of the int
erface
<DIV></DIV>>address. If there are no loops then the highest address wins.
<DIV></DIV>>Kyle
<DIV></DIV>>
<DIV></DIV>>
<DIV></DIV>>At 10:34 AM 1/22/2001 -0800, you wrote:
<DIV></DIV>>
<DIV></DIV>> >
<DIV></DIV>> > Explain your reasoning behind allocating looback addresses
first please. I
<DIV></DIV>> > thought it would be better to allocate them last so that t
hey are the highest
<DIV></DIV>> > addresses on the routers. That way your RID's wouln't chan
ge if you reboot
<DIV></DIV>> > etc.
<DIV></DIV>> >
<DIV></DIV>> > >From: Ronnie Royston
<DIV></DIV>> > >Reply-To: Ronnie Royston
<DIV></DIV>> > >To: "'Dave Martin'" , ccielab@groupstudy.com
<DIV></DIV>> > >Subject: RE: IP allocation matrix question?
<DIV></DIV>> > >Date: Mon, 22 Jan 2001 11:19:43 -0600
<DIV></DIV>> > >
<DIV></DIV>> > >Yea, I do.
<DIV></DIV>> > >
<DIV></DIV>> > >Don't try and get all scientific about it. Even though
you've spent alot of
<DIV></DIV>> > >time mastering the VLSM concept, it's easier to read r
outing tables if you
<DIV></DIV>> > >keep things as simple as possible - using incrementall
y higher numbers in
<DIV></DIV>> > >the 3rd octet for example even though you may have /24
/25 /26 /27, etc. As
<DIV></DIV>> > >long as the network portion of your addressing is diff
erent for each subnet,
<DIV></DIV>> >
<DIV></DIV>> > >you won't have any overlapping issues. Also, before yo
u start typing or
<DIV></DIV>> > >writing, take into account if you are going to use loo
pbacks and allocate
<DIV></DIV>> > >those subnet numbers first. Hope that helps.
<DIV></DIV>> > >
<DIV></DIV>> > >-----Original Message-----
<DIV></DIV>> > >From: Dave Martin [mailto:ciscoguy2000@hotmail.com]
<DIV></DIV>> > >Sent: Monday, January 22, 2001 10:59 PM
<DIV></DIV>> > >To: ccielab@groupstudy.com
<DIV></DIV>> > >Subject: IP allocation matrix question?
<DIV></DIV>> > >
<DIV></DIV>> > >
<DIV></DIV>> > >Does anyone have any good tips for an IP allocation ma
trix for the lab?
<DIV></DIV>> > >
<DIV></DIV>> > >Thanks,
<DIV></DIV>> > >Dave
<DIV></DIV>> > >
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:27:39 GMT-3