From: Arun Arumuganainar (aarumuga@hotmail.com)
Date: Tue Jul 12 2005 - 10:55:32 GMT-3
Yeah .... I confirmed it in my lab . Order does not matter at all !!! IOS
will re-order them in the Most specific way ( Longest match will be put
first automatically )
Arun
----- Original Message -----
From: <Bill.McKenzie@bisys.com>
To: "ccie2be" <ccie2be@nyc.rr.com>
Cc: "'Arun Arumuganainar'" <aarumuga@hotmail.com>; "'Chris Aguillo'"
<ccaguillo1@hotpop.com>; <ccielab@groupstudy.com>; "'ccie_06'"
<ccie_06@att.net>; "'Amit Jain'" <netsteps@rediffmail.com>;
<nobody@groupstudy.com>; "'Godswill Oletu'" <oletu@inbox.lv>
Sent: Tuesday, July 12, 2005 6:36 PM
Subject: RE: OSPF Network Statement
> Tim,
> This was demonstrated by Brian from IE that in the newer IOS's , it
doesn't
> matter which order you put them in because it will re-order them from most
> to least specific anyway.
>
> Bill
>
>
> Bill McKenzie
> Network Support
> BISYS Fund Services
> Work: 614.428.3293
> Cell: 614.425.2370
>
> "Exceeding Client Expectations"
>
>
>
> "ccie2be"
> <ccie2be@nyc.rr.c
> om> To
> Sent by: "'Arun Arumuganainar'"
> nobody@groupstudy <aarumuga@hotmail.com>, "'Amit
> .com Jain'" <netsteps@rediffmail.com>,
> "'Chris Aguillo'"
> <ccaguillo1@hotpop.com>, "'Godswill
> 07/12/2005 08:58 Oletu'" <oletu@inbox.lv>,
> AM "'ccie_06'" <ccie_06@att.net>,
> <ccielab@groupstudy.com>
> cc
> Please respond to
> "ccie2be" Subject
> <ccie2be@nyc.rr.c RE: OSPF Network Statement
> om>
>
>
>
>
>
>
>
>
>
> Arun,
>
> I like the way you demonstrated your point of view.
>
> But, unfortunately, that doesn't address the issue head on.
>
> The issue is NOT exactly which interfaces will be put into which areas
> based
> on the network statements.
>
> The issue IS does the sequence of network statements matter.
>
> In your excellent example, this issue was avoided because your network
> statements went from the most specific to the least specific.
>
> The question is whether the results would be exactly the same if the
> sequence were changed to listing the least specific to most specific?
>
> It used to be that the order of network statements mattered a great deal.
> The question is whether this is still true.
>
> Tim
>
> _____
>
> From: Arun Arumuganainar [mailto:aarumuga@hotmail.com]
> Sent: Tuesday, July 12, 2005 8:08 AM
> To: Amit Jain; ccie2be; 'Chris Aguillo'; 'Godswill Oletu'; 'ccie_06';
> ccielab@groupstudy.com
> Subject: Re: OSPF Network Statement
>
> Hi Amit ,
>
> This is the OSPF behavior from the good old times !!!! You should look at
> how inverse mask works to understand this better Let me explain .
>
> Inverse mask contains series of zeroes followed by series of 1s ( Just
> opposite to subnetmask ) . When match is to made it will just check zero
> portion of it and 1s portion are called don't cares.
>
> i.e "11.0.0.0 0.0.0.255" will try to match for 11.0.0 and last octet
> could be any thing .
>
> how ever "11.0.0.0 0.255.255.255" will match for 11 alone and rest of
> bits
> are dontcares
>
> If all the inverse mask bits are 1 ( 255.255.255.255) then it will
> equivalent to Match any address . In such a case OSPF will be enabled of
> all
> the interfaces which has an IP address configured on the router .
>
> Router Traces to illustrate the behavior
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> Router#sh runn | b ospf
> router ospf 100
> log-adjacency-changes
> ! >>> Configured OSPF with wild card mask( All 1s inverse mask )
> network 0.0.0.0 255.255.255.255 area 0
> !>>>Further traces are omitted
>
> Router#sh ip ospf int brie
> Interface PID Area IP Address/Mask Cost State Nbrs F/C
> Se1/0 100 0 10.1.1.2/24 64 P2P 0/0
> Et0/0 100 0 11.0.0.1/8 10 DR 0/0
>
> Note : all the interfaces in the above example are assigned to area 0
> Again in case there is more specific match is available then that network
> statement will over-ride all the other network statement in the ospf
> configuration .
>
> Router#sh runn | b ospf
> router ospf 100
> log-adjacency-changes
> ! >>> Note : Area 4 is the most specific Match
> network 11.0.0.0 0.0.0.255 area 4
> network 11.0.0.0 0.0.255.255 area 3
> network 11.0.0.0 0.255.255.255 area 2
> network 0.0.0.0 255.255.255.255 area 0
> !
> Note : E0/0 has been assigned the area with most specific match
> Router#sh ip ospf int brie
> Interface PID Area IP Address/Mask Cost State Nbrs F/C
> Se1/0 100 0 10.1.1.1/24 64 P2P 0/0
> Et0/0 100 4 11.0.0.1/8 10 WAIT 0/0
>
> Hope this helps .
>
> Thanks and Regards
> Arun
>
> ----- Original Message -----
> From: "Amit Jain" < <mailto:netsteps@rediffmail.com>
> netsteps@rediffmail.com>
> To: "ccie2be" < <mailto:ccie2be@nyc.rr.com> ccie2be@nyc.rr.com>; "'Chris
> Aguillo'" < <mailto:ccaguillo1@hotpop.com> ccaguillo1@hotpop.com>;
> "'Godswill Oletu'" < <mailto:oletu@inbox.lv> oletu@inbox.lv>; "'ccie_06'"
<
> <mailto:ccie_06@att.net> ccie_06@att.net>; <
<mailto:ccielab@groupstudy.com
> >
> ccielab@groupstudy.com>
> Sent: Tuesday, July 12, 2005 2:42 PM
> Subject: Re: OSPF Network Statement
>
> > I guess you are r right Tim. IOS has gotten smarter. I am sure that I
> read
> > it somewhere that new IOS will not consider the sequence now.
> > Though I also have not tested also but new IOS works that way.
> >
> > Amit
> > ----- Original Message -----
> > From: "ccie2be" < <mailto:ccie2be@nyc.rr.com> ccie2be@nyc.rr.com>
> > To: "'Chris Aguillo'" < <mailto:ccaguillo1@hotpop.com>
> ccaguillo1@hotpop.com>; "'Godswill Oletu'"
> > < <mailto:oletu@inbox.lv> oletu@inbox.lv>; "'ccie_06'" <
> <mailto:ccie_06@att.net> ccie_06@att.net>; <
<mailto:ccielab@groupstudy.com
> >
> ccielab@groupstudy.com>
> > Sent: Tuesday, July 12, 2005 11:28 AM
> > Subject: RE: OSPF Network Statement
> >
> >
> > > Chris,
> > >
> > > Has IOS gotten smarter?
> > >
> > > It used to be that if the first network statement included 0.0.0.0
> > > 255.255.255.255, that was it. No other statements mattered. I haven't
> > > tested this myself but it used to be that the sequence of net
> statements
> > > mattered.
> > >
> > > Tim
> > >
> > > -----Original Message-----
> > > From: <mailto:nobody@groupstudy.com> nobody@groupstudy.com
> [mailto:nobody@groupstudy.com] On Behalf Of
> > > Chris Aguillo
> > > Sent: Tuesday, July 12, 2005 12:55 AM
> > > To: Godswill Oletu; ccie_06; <mailto:ccielab@groupstudy.com>
> ccielab@groupstudy.com
> > > Subject: OSPF Network Statement
> > >
> > > Hi ccie_06,
> > >
> > > upon you start the OSPF process, the algorithm will check which
> > > interface is part of an OSPF area or not.
> > >
> > > Each NETWORK command is treated sequentially as it is written under
the
> > > ROUTER OSPF command.
> > >
> > > Also in each every NETWORK ip_address inverse_mask command, you will
> > > treat it as it is an ACL.
> > >
> > > taking out from your original script:
> > >
> > > network 0.0.0.0 255.255.255.255 area 0
> > > >>> 0.0.0.0 255.255.255.255 means all IP-Address will match
> > > >>> any interface that have a valid IP address and will be placed in
> > Area0
> > >
> > > network 1.2.3.0 0.0.0.0 area 1
> > > >>> The IP address 1.2.3.0 is a host address as per inverse_mask
> > > >>> The Interface that have the IP address of 1.2.3.0 is pulled out
> > > from area 0 and placed into area 1.
> > >
> > >
> > > network 1.2.3.4 0.0.0.0 area 2
> > > >>> The IP address 1.2.3.4 is a host address as per inverse_mask
> > > >>> The Interface that have the IP address of 1.2.3.4 is pulled out
> > > from area 0 and placed into area 2
> > >
> > > As a result, all router interfaces are in Area 0 except 1.2.3.0 and
> > > 1.2.3.4 which are in Area 1 and Area 2 respectively.
> > >
> > > Moreover, reversing the Network statement and placing as the last
> > > statement the command
> > >
> > > NETWORK 0.0.0.0 255.255.255.255 AREA 0
> > > >>> this will remove all interfaces assigned in other Areas (earlier
> > > NETWORK command) and will bring other interfaces not yet assigned for
> > > any Area and all be brought into AREA 0.
> > >
> > > >>> This is in effect because the Network command is treated
> > sequencially.
> > >
> > > I do not see why network 1.2.3.0 0.0.0.0 will not match any interface
> > > and no interface as a result will be in AREA 1 as per Godswill below?
> > > >>> as long as there is a valid IP_address configured in the router
as
> > > 1.2.3.0, that only interface will be placed in AREA 1.
> > >
> > > I do not agree with the explanation of Godswill for:
> > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> > > network 0.0.0.0 255.255.255.255 area 0 -> Will match ALL remaining
> > > interfaces not previously matched and place them into area 0.
> > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> > >
> > > The network command is placed as the first item in the series of
> Network
> > > command, thus it will make all interface in the area 0 (whether or not
> > > matched in other network command, take note this network is the first
> > > one). The way I analyzed the words <<<all remaining>>> and <<<not
> > > previously mathed>>> is probably he thinks that Network Command is not
> > > trated sequentially.
> > >
> > > HTH....chris
> > >
> > >
> > > Godswill Oletu wrote:
> > > > The interfaces will be assigned like this:
> > > > network 1.2.3.0 0.0.0.0 area 1 -> Will not match any interface, no
> > > > interface will be in area 1
> > > >
> > > > network 1.2.3.4 0.0.0.0 area 2 -> Will match only interface 1.2.3.4
> and
> > > > put it into area 2
> > > >
> > > > network 0.0.0.0 255.255.255.255 area 0 -> Will match ALL remaining
> > > > interfaces not previously matched and place them into area 0.
> > > >
> > > > Thanks.
> > > > Godswill Oletu
> > > >
> > > > ----- Original Message ----- From: "ccie_06" <
> <mailto:ccie_06@att.net> ccie_06@att.net>
> > > > To: < <mailto:ccielab@groupstudy.com> ccielab@groupstudy.com>
> > > > Sent: Sunday, July 10, 2005 12:29 PM
> > > > Subject: Ospf Network statement
> > > >
> > > >
> > > >> Can any one explain following statement under ospf process ( I
mean
> > > >> which interface will participate in which area.?)
> > > >>
> > > >> router ospf 1
> > > >> network 0.0.0.0 255.255.255.255 area 0
> > > >> network 1.2.3.0 0.0.0.0 area 1
> > > >> network 1.2.3.4 0.0.0.0 area 2
> > >
> > >
This archive was generated by hypermail 2.1.4 : Sun Sep 04 2005 - 17:00:29 GMT-3