From: Connary, Julie Ann (jconnary@xxxxxxxxx)
Date: Mon Jan 15 2001 - 15:41:09 GMT-3
HI,
I looked this up and it is only for changing metrics and tag values of
external routes before they enter into the routing table.
My route-map kind of sort of works. The problem is that now on
Router 4 anything destined for the 170.100.200.0 network (a locally
connected network out Ethernet 0) it is routed out the serial interface to
R5 and then back across the serial interface before going out Ethernet 0.
Not exactly optimal.
I've been playing around with variations but have not come up with one that
allows me to match on next-hop = 0.0.0.0 or null 0 to get around this.
The documentation says under the route-map that I have the following
commands to work with:
Policy Routing
Another purpose of route maps is to enable policy-routing. Use the ip
policy route-map command, in addition to the route-map command, and the match
and set commands to define the conditions for policy-routing packets.
The related match and set commands are listed in the section "Related
Commands for
Policy Routing." The match commands specify the conditions under
which policy routing occurs. The set commands specify the routing actions
to perform if
the criteria enforced by the match commands are met. You might want
to policy-route packets some way other than the obvious shortest path.
The sequence-number works as follows:
1. If no entry is defined with the supplied tag, an entry is created
with sequence-number set to 10.
2. If only one entry is defined with the supplied tag, that entry
becomes the default entry for the following route-map command. The
sequence-number of this
entry is unchanged.
3. If more than one entry is defined with the supplied tag, an error
message is printed to indicate that sequence-number is required.
Related Commands for Policy Routing
You can use the master indexes or search online to find documentation of
related commands.
ip policy route-map
match ip address
match length
set default interface
set interface
set ip default next-hop
set ip next-hop
the only other thing I can think of is to explicitly map the network I want
to reach. Which is also not optimal, because then everytime
a new non/24 bit subnet is added, the route-map for the policy routing must
be updated.
yuck - there must be a better way - just do not know what it is right now.
No statics. A default does not work because
of the 170.0.0.0 via 0.0.0.0 to null-0 route. Can't get 170.0.0.0 out of
the routing table or BGP won't advertise it to it;s EBGP neighbors.
That leaves policy routing with the only match being length or ip address.
Any other ideas?
Julie Ann
At 08:04 PM 1/15/2001 +0200, Lykourgiotis Paraskevas wrote:
>I think another solution coulb be the use of table-map command,under bgp,
>which is used to process bgp routes before entering into the routing table.
>
>-----Original Message-----
>From: Connary, Julie Ann [mailto:jconnary@cisco.com]
>Sent: Monday, January 15, 2001 6:37 PM
>To: ccielab@groupstudy.com
>Subject: Re: BGP aggregates and IGRP default networks - policy routing is
>the answer
>
>
>Hi All,
>
>solved my own problem - policy routing!!!!
>
>I set up a route-map that anything that matched 170.100.0.0 would be policy
>routed to
>next hop ip 170.100.1.5. I then enabled local policy routing - so now
>everything destined for 170.100.0.0 is policy routed before it hits the
>route-table and my
>BGP aggregate route is no longer a problem. Don't forget to policy route
>any interfaces so if you have
>to ping "through" this router that it works.
>
>route-map subnets, permit, sequence 10
> Match clauses:
> ip address (access-lists): 5
> Set clauses:
> ip next-hop 170.100.1.5
> Policy routing matches: 93 packets, 6479 bytes
>Standard IP access list 5
> permit 170.100.0.0, wildcard bits 0.0.255.255
>
>ip local policy route-map subnets
>
>
>Now if I turn on debug ip policy and ping 170.100.42.241 I get:
>
>1d22h: IP: route map subnets, item 10, permit
>1d22h: IP: s=170.100.1.4 (local), d=170.100.42.241 (Serial0), len 100,
>policy ro
>uted
>1d22h: IP: local to Serial0 170.100.1.5
>1d22h: IP: s=170.100.1.4 (local), d=170.100.42.241, len 100, policy match
>1d22h: IP: route map subnets, item 10, permit
>1d22h: IP: s=170.100.1.4 (local), d=170.100.42.241 (Serial0), len 100,
>policy ro
>uted
>1d22h: IP: local to Serial0 170.100.1.5
>1d22h: IP: s=170.100.1.4 (local), d=170.100.42.241, len 100, policy match
>1d22h: IP: route map subnets, item 10, permit
>1d22h: IP: s=170.100.1.4 (local), d=170.100.42.241 (Serial0), len 100,
>policy ro
>uted
>1d22h: IP: local to Serial0 170.100.1.5
>1d22h: IP: s=170.100.1.4 (local), d=255.255.255.255, len 46, policy match
>1d22h: IP: route map subnets, item 10, permit
>1d22h: IP: s=170.100.1.4 (local), d=255.255.255.255 (Serial0), len 46,
>policy ro
>uted
>1d22h: IP: local to Serial0 170.100.1.5
>
>Julie Ann
>
>
>
>
>
>At 10:50 AM 1/15/2001 -0500, Connary, Julie Ann wrote:
> >Hi All,
> >
> >
> >I have one I cannot figure out. I have an IGRP router that is using a
> >default-network to
> >overcome subnet mask length differences into my OSPF network. Works great
> >until I
> >did the BGP part of the lab which requires an aggregate address. The
> >aggregate address puts
> >a static route to null 0 in my routing table and I can no longer default
> >route to the OSPF network:
> >
> >
> >
> >OSPF Network with 24, 28 and 29 bit masks of 170.100.x.x
> > |
> > |
> > |
> > Router 5 - OSPF and IGRP
> > |
> > | 170.100.1.x/24
> > |
> > Frame-relay point-to-point link
> > |
> > |
> > |
> > Router 4 - running IGRP and BGP
> > ip default-network 200.0.5.0
> > router bgp 5
> > no auto-summary
> > network 170.100.1.0 mask 255.255.255.0
> > aggregate-address 170.0.0.0 255.0.0.0 summary-only
> >
> >
> >Now I get the following routing table on R4 and cannot get to any of the
> >non 24 bit subnets in OSPF.
> >How do I fix WITHOUT using static routes on R5 that "summarize" the 28 and
> >29 bit subnets to
> >24 bits and redistribute them into IGRP. Can I stop the aggregate address
> >from being entered into
> >R4's routing table. I really only want to advertise that aggregate to my
> >EBGP neighbor anyways.
> >
> >Gateway of last resort is 170.100.1.5 to network 210.0.5.0
> >
> > 170.100.0.0/24 is subnetted, 9 subnets
> >C 170.100.200.0 is directly connected, Ethernet0
> >I 170.100.233.0 [100/10476] via 170.100.1.5, 00:01:03, Serial0
> >I 170.100.129.0 [100/10476] via 170.100.1.5, 00:01:03, Serial0
> >I 170.100.150.0 [100/10476] via 170.100.1.5, 00:01:03, Serial0
> >I 170.100.68.0 [100/10476] via 170.100.1.5, 00:01:03, Serial0
> >I 170.100.67.0 [100/10476] via 170.100.1.5, 00:01:03, Serial0
> >I 170.100.64.0 [100/10476] via 170.100.1.5, 00:01:04, Serial0
> >I 170.100.65.0 [100/10476] via 170.100.1.5, 00:01:04, Serial0
> >C 170.100.1.0 is directly connected, Serial0
> > 1.0.0.0/24 is subnetted, 1 subnets
> >C 1.1.1.0 is directly connected, Loopback0
> >I* 210.0.5.0/24 [100/10476] via 170.100.1.5, 00:01:04, Serial0
> >B 205.15.150.0/24 [20/0] via 170.100.200.7, 1d18h
> >B 170.0.0.0/8 [200/0] via 0.0.0.0, 1d18h, Null0 <---------------this
> >entry is stopping my default-network from being usefull.
> >tserve#ping 170.100.42.241
> >
> >Type escape sequence to abort.
> >Sending 5, 100-byte ICMP Echos to 170.100.42.241, timeout is 2 seconds:
> >.....
> >Success rate is 0 percent (0/5)
> >tserve#
> >
> >
> >Thanks,
> >
> >Julie Ann
> >
> >------------------------------------------------------------------------
> > Julie Ann Connary
> > | | Network Consulting Engineer
> > ||| ||| Federal Support Program
> > .|||||. .|||||. 13635 Dulles Technology Drive,
> >Herndon VA 20171
> > .:|||||||||:.:|||||||||:. Pager: 1-888-642-0551
> > c i s c o S y s t e m s Email: jconnary@cisco.com
> >
> >------------------------------------------------------------------------
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:27:30 GMT-3