[nsp] EIGRP looping problem but not SIA ?
Brian Turnbow
b.turnbow at twt.it
Mon Dec 22 10:35:06 EST 2003
In your case you have a static route to an interface so the route is viewed
as a connected not static route.
you then have a match on the interface vie the network statement for eigrp
so the route gets redistributed, as you can see by the show ip route.
are you routing internally by the /22 networks or by more specific nets ?
If you route via more specific place the null routes on the bgp edges with a
metric less than eigrp, route by more specific
and whatever doesn't match a more specific route gets dropped on ingress.
You can also configure the null route(s) on a single router and redistribute
to the others.
BGP will advertise it's configured networks as long as it has a match in the
routing table.
hope it helps
Brian
-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net
[mailto:cisco-nsp-bounces at puck.nether.net]On Behalf Of Eric Kagan
Sent: lunedì 22 dicembre 2003 13.06
To: Oliver Boehmer (oboehmer)
Cc: cisco-nsp at puck.nether.net
Subject: Re: [nsp] EIGRP looping problem but not SIA ?
> I'd start with "debug ip routing <acl>" with an <acl> which lists all
> the networks you're concerned with on different routers within your
> network. This should give you an idea on why these EIGRP routes are
> flapping. This looks like a config issue to me, like improper
> redistribution between EIGRP and a different protocol or between BGP and
> EIGRP.
I tried the above debug but there was no output. From telnet in I
configured:
term mon
access-list 30 permit 10.2.68.0 255.255.252.0
debug ip routing 30
>
> One comment though:
>
> > Router A: 10.2.68.1 /28 (problem subnet is 10.2.68.0 /22. It is
> > connected with Null route, yet show ip route show known from eigrp
> > nei 172.5.228.134)
>
> You configured a floating static NULL route (ip route 10.2.68.0
> 255.255.252.0 Null0 254), so if the router knows this route via a
> protocol with a lower admin-distance than 254, it will take this one,
> this is what you're seeing, and this is expected behaviour..
> You want to find out why the eigrp routes are flapping (you can see this
> by looking at the "00:00:11 ago"-timestamp).
>
> oli
But if I don't have a redistribute static on this router, why is it being
picked up by other routers ? Is there a better way to tie down the BGP
subnets so this will not occur ?
I should have mentioned I have the same exact config on Router B and this
scenario does not occur. If I issue a show ip route 10.2.64.0 255.255.252.0
I get the following.
Routing entry for 10.2.64.0/22
Known via "static", distance 250, metric 0 (connected)
Redistributing via eigrp 65555
Advertised by eigrp 65555
Routing Descriptor Blocks:
* directly connected, via Null0
Route metric is 0, traffic share count is 1
I am find with reconfiguring the setup, I'm just unsure about what changes
to make. Someone mentioned using the loopback interfaces on the routers for
BGP neighboe instead of the actual interface IP's. Would this have an
affect on the EIGRP routing as well ?
Thanks
Eric
>
> ----Original Message----
> From: cisco-nsp-bounces at puck.nether.net
> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Eric Kagan
> Sent: Freitag, 19. Dezember 2003 14:15 To: cisco-nsp at puck.nether.net
> Subject: [nsp] EIGRP looping problem but not SIA ?
>
> > I have been battling a problem for 2 weeks which I originally thought
> > was a BGP problem (subnets flapping and getting dampened). After
> > stripping BGP to bare config it didn't fix it and I stumbled across
> > an EIGRP problem on those same subnets. Basically I have my BGP
> > network statements being tied down with Static Null routes (I will
> > list config clips below). It seems these subnets are being
> > advertised and re-distributed through EIGRP to all neighbors and it
> > is a loop. I have read 100 docs, archive postings, etc and I can't
> > figure it out. Everything refers to SIA, but I have logging on and
> > don't have any error messages. Also, the actual neighbor sessions are
> > not getting reset (unless I reboot or clear them). The strange part
> > is it only happens on certain subnets and not all of them. There was
> > some mention of EIGRP routing loop bugs so I have upgraded IOS on all
> > 3 of the the routers as well. Here is a clip of the routers. I am
> > trying to only list relevant info. Any help or ideas would be
> > greatly appreciated. I am also willing to work with someone offline
> > on a consulting basis if interested.
> >
> > Router A: 10.2.68.1 /28 (problem subnet is 10.2.68.0 /22. It is
> > connected with Null route, yet show ip route show known from eigrp
> > nei 172.5.228.134)
> >
> > router eigrp 65555
> > passive-interface FastEthernet0/0
> > network 10.2.64.0 0.0.0.255
> > network 10.2.65.0 0.0.0.255
> > network 10.2.68.0 0.0.3.255
> > no auto-summary
> > eigrp log-neighbor-changes
> > !
> > router bgp 65555
> > no synchronization
> > bgp log-neighbor-changes
> > network 10.2.68.0 mask 255.255.252.0
> > network 10.2.88.0 mask 255.255.252.0
> > network 10.2.95.0 mask 255.255.255.0
> >
> > ip route 10.2.68.0 255.255.252.0 Null0 254
> > ip route 10.2.88.0 255.255.252.0 Null0 254
> > ip route 10.2.95.0 255.255.255.0 Null0 254
> >
> > IP-EIGRP neighbors for process 65555
> > 1 10.2.65.6 Fa4/0 13 00:37:19 1 200 0
> > 26051 0 172.5.228.134 AT2/0.2 13 00:37:47 9 200 0
> > 27618 2 10.2.68.5 Fa1/0 13 00:38:51 1 200
> > 0 2386721
> >
> > show ip rou 10.2.68.0 255.255.252.0
> > Routing entry for 10.2.68.0/22
> > Known via "eigrp 65555", distance 90, metric 581120, type internal
> > Redistributing via eigrp 65555
> > Last update from 172.5.228.134 on ATM2/0.2, 00:01:04 ago
> > Routing Descriptor Blocks:
> > * 172.5.228.134, from 172.5.228.134, 00:01:04 ago, via ATM2/0.2
> > Route metric is 581120, traffic share count is 1
> > Total delay is 480 microseconds, minimum bandwidth is 4500 Kbit
> > Reliability 255/255, minimum MTU 1500 bytes
> > Loading 8/255, Hops 3
> >
> >
> > show ip eigrp top act
> > IP-EIGRP Topology Table for AS(65555)/ID(10.2.64.193)
> >
> > Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,
> > r - reply Status, s - sia Status
> >
> > A 10.2.68.0/22, 1 Successors, FD is 581120
> > 1 replies, active 00:00:39, query-origin: Clear
> > via 10.2.65.6 (Infinity/Infinity), FastEthernet4/0, serno 3
> > * via Redistributed (Infinity/Infinity)
> > Remaining replies:
> > via 172.5.228.134, r, ATM2/0.2
> >
> >
> > Router B: (10.2.64.1 /25) - The loop is -
> >
> > IP-EIGRP neighbors for process 65555
> > 2 10.2.64.2 Fa1/0 11 00:45:59 1 200 0
> > 470103 1 10.2.65.2 AT2/0.1 12 00:46:00 1 200
> > 0 27845 0 10.2.65.5 Fa4/0 14 00:46:01 1
> > 200 0 30186
> >
> > show ip rou 10.2.68.0 255.255.252.0
> > Routing entry for 10.2.68.0/22
> > Known via "eigrp 65555", distance 90, metric 28160, type internal
> > Redistributing via eigrp 65555
> > Last update from 10.2.65.5 on FastEthernet4/0, 00:00:11 ago
> > Routing Descriptor Blocks:
> > * 10.2.65.5, from 10.2.65.5, 00:00:11 ago, via FastEthernet4/0
> > Route metric is 28160, traffic share count is 1
> > Total delay is 100 microseconds, minimum bandwidth is 100000
> > Kbit Reliability 255/255, minimum MTU 1500 bytes
> > Loading 9/255, Hops 1
> >
> >
> > Router C (10.2.72.1 /27)
> >
> > IP-EIGRP neighbors for process 65555
> >
> > 0 10.2.65.1 AT2/0.1 10 00:45:01 22 200 0
> > 26205 2 172.5.228.133 AT2/0.2 12 00:45:34 8 200
> > 0 30143 1 10.2.72.5 Fa1/0.1 14 00:45:37 1
> > 200 0 3402274
> >
> > show ip route 10.2.68.0 255.255.252.0
> > Routing entry for 10.2.68.0/22
> > Known via "eigrp 65555", distance 90, metric 64256, type internal
> > Redistributing via eigrp 65555
> > Last update from 10.2.65.1 on ATM2/0.1, 00:00:54 ago
> > Routing Descriptor Blocks:
> > * 10.2.65.1, from 10.2.65.1, 00:00:54 ago, via ATM2/0.1
> > Route metric is 64256, traffic share count is 1
> > Total delay is 290 microseconds, minimum bandwidth is 45045 Kbit
> > Reliability 255/255, minimum MTU 1500 bytes
> > Loading 15/255, Hops 2
> >
> > Hopefully that is enough config to get an idea of what is going on.
> > I tried to post as little as possible while still being thorough. I
> > am welcome to offline chat, ideas, etc and I am willing to hire
> > someone for a few hours to help troubleshoot and resolve.
> >
> > Thanks
> > Eric
> >
> > _______________________________________________
> > cisco-nsp mailing list cisco-nsp at puck.nether.net
> > https://puck.nether.net/mailman/listinfo/cisco-nsp
> > archive at http://puck.nether.net/pipermail/cisco-nsp/
>
_______________________________________________
cisco-nsp mailing list cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/
More information about the cisco-nsp
mailing list