[c-nsp] Multilink Routing Problem

Fri Jun 13 14:27:05 EDT 2008

Hi folks...

Sorry for the long message but wanted to share what was causing this
problem...

We have two T1's - each connected to a separate 8 port T1 card on the 6500.
We were trying to do multilink across the two cards and it obviously doesn't
work (or at least with our config).  Moved both T1's to the same card and
everything is now fine.  Just an FYI in case someone else runs across
this..;)

Paul

-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net
[mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Paul Stewart
Sent: Thursday, June 12, 2008 8:59 PM
To: 'Eric Cables'
Cc: cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] Multilink Routing Problem

Thanks.

Here's the OSPF config from the 6500:

router ospf 1

 router-id xx.xx.98.184

 log-adjacency-changes

 redistribute connected subnets tag 116661016

 redistribute static subnets tag 116661016

 network xxx.xxx.227.0 0.0.0.7 area 0

 network xx.xx.246.16 0.0.0.7 area 0

 network xx.xx.64.48 0.0.0.7 area 0

 network xx.xx.64.56 0.0.0.7 area 0

.......more.....

Everything else connected to this device works. actually, I just checked
from our core devices and the point to point /30 and the statically routed
/29 are both in the routing tables:

dis1-rtr-pt#sh ip route xx.xx.120.202

Routing entry for xx.xx.xx.202/32

  Known via "connected", distance 0, metric 0 (connected, via interface)

  Redistributing via ospf 1

  Advertised by ospf 1 subnets tag 116661016

  Routing Descriptor Blocks:

  * directly connected, via Multilink1

      Route metric is 0, traffic share count is 1

dis1-rtr-pt#sh ip route xx.xx.120.201

Routing entry for xx.xx.120.200/30

  Known via "connected", distance 0, metric 0 (connected, via interface)

  Redistributing via ospf 1

  Advertised by ospf 1 subnets tag 116661016

  Routing Descriptor Blocks:

  * directly connected, via Multilink1

      Route metric is 0, traffic share count is 1

Now, my question is why is the far end point to point IP showing as a /32?

We have a static route to send the /29 to the "far end" and it looks fine:

dis1-rtr-pt#sh ip route xx.xx.120.232

Routing entry for xx.xx.120.232/29

  Known via "static", distance 1, metric 0

  Redistributing via ospf 1

  Advertised by ospf 1 subnets tag 116661016

  Routing Descriptor Blocks:

  * xx.xx.120.202

      Route metric is 0, traffic share count is 1

The /32 is throwing me off though.??

Thanks,

Paul

From: Eric Cables [mailto:ecables at gmail.com] 
Sent: Thursday, June 12, 2008 7:51 PM
To: Paul Stewart
Cc: cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] Multilink Routing Problem

I'd like to see more of the OSPF config.  Is this all part of area 0, or is
it configured as a stubby network?  If it's stubby, you may need to
configure it as NSSA to get the redistributed routes back into area 0.

On Thu, Jun 12, 2008 at 2:42 PM, Paul Stewart <paul at paulstewart.org> wrote:

Hi folks.. missing something obvious here.;)

Cisco 6500 on one side, 2621 on other (customer prem)

Relevent 6500 Config:

controller T1 5/0/7

 framing esf

 clock source internal

 linecode b8zs

 channel-group 1 timeslots 1-24

controller T1 6/0/7

 framing esf

 clock source internal

 linecode b8zs

 channel-group 1 timeslots 1-24

interface Multilink1

 ip address xx.xx.120.201 255.255.255.252

 ppp multilink

 multilink-group 1

interface Serial5/0/7:1

no ip address

 encapsulation ppp

 ppp multilink

 multilink-group 1

interface Serial6/0/7:1

no ip address

 encapsulation ppp

 ppp multilink

 multilink-group 1

router ospf 1

log-adjacency-changes

 redistribute connected subnets

xxxxxxxxxxxxx

xxxxxxxxxxxxx

2600 Config:

interface Multilink1

ip address xx.xx.120.202 255.255.255.252

 ppp multilink

 ppp multilink group 1

interface Serial0/0

 no ip address

 encapsulation ppp

 service-module t1 timeslots 1-24

 ppp multilink

 ppp multilink group 1

interface Serial0/1

no ip address

 encapsulation ppp

 service-module t1 timeslots 1-24

 ppp multilink

 ppp multilink group 1

ip route 0.0.0.0 0.0.0.0 xx.xx.120.201

>From the 6500 or the 2621 I can ping between the interfaces no problem but
can't get traffic past the 6500.  It's redistributing routes via OSPF to the
rest of our network but for some reason not the multilink IP address..
Anything make sense about this? I can ping the local 6500 interface from
anywhere in our network but not the 2621 end (except when on the 6500).

#sh ppp multilink

Multilink1, bundle name is dis1-rtr-pt

 Endpoint discriminator is dis1-rtr-pt

 Bundle up for 06:32:41, total bandwidth 3072, load 1/255

 Receive buffer limit 24000 bytes, frag timeout 1000 ms

   0/0 fragments/bytes in reassembly list

   0 lost fragments, 22 reordered

   0/0 discarded fragments/bytes, 0 lost received

   0xF19 received sequence, 0xC3E sent sequence

 Member links: 2 active, 0 inactive (max not set, min not set)

   Se0/0, since 06:32:41

   Se0/1, since 06:32:37

No inactive multilink interfaces

Originally we had OSPF on both sides of the link and the OSPF session would
come up no problem as "point to point" .. Confused, we moved to static
routing to try and prove this out.  Is there something special you have to
do to get a multilink bundle redistributed in OSPF??

Thanks in advance,

Paul Stewart

_______________________________________________
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/

-- 
Eric Cables 

_______________________________________________
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/

-- 
No virus found in this incoming message.
Checked by AVG. 
Version: 7.5.524 / Virus Database: 270.3.0/1500 - Release Date: 6/12/2008
4:58 PM