From: Scott Morris (smorris@internetworkexpert.com)
Date: Wed Oct 01 2008 - 08:38:45 ART
PIM allows the multicast conversations to occur without issue. So yes, they
much be PIM neighbors. That will simply help the forwarding of traffic.
By using the "bsr-border" or using a "multicast-boundary" (for auto-rp) that
will prevent the leakage of RP information so you don't unnecessarily mess
up the tree structure. That way MSDP can do what it's supposed to.
Have fun!
Scott Morris, CCIE4 #4713, JNCIE-M #153, JNCIS-ER, CISSP, et al.
CCSI/JNCI-M/JNCI-ER
Senior CCIE Instructor
smorris@internetworkexpert.com
Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll Free: 877-224-8987
Outside US: 775-826-4344
Online Community: http://tinyurl.com/6dmnsu
CCIE Blog: http://tinyurl.com/2nxxaq
Knowledge is power.
Power corrupts.
Study hard and be Eeeeviiiil......
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Hobbs
Sent: Wednesday, October 01, 2008 12:50 AM
To: Cisco certification
Subject: Re: Strange MSDP scenario, every 99th ping is good
Well I got this one figured out. R7 and R8 have to be PIM neighbors. But I
use a bsr-border command on R8 and multicast boundary with an ACL on R7 to
block the auto-rp groups from getting across. This separates the two
domains, each with one RP. Pings are smooth
On Tue, Sep 30, 2008 at 9:42 PM, Hobbs <deadheadblues@gmail.com> wrote:
> By the way if R7 and R8 are PIM neighbors, the ping works great. I
> guess I should be asking this:
>
> Are R7 and R8 required to be PIM neighbors or is there another way
> around it?
>
> I thought the R1 sending the MSDP message to R8, would populate R8's
> OIL, but it appears not...
>
>
> On Tue, Sep 30, 2008 at 9:17 PM, Hobbs <deadheadblues@gmail.com> wrote:
>
>> Here is my scenarion, SW4 is attempting to ping group joined by BB1:
>>
>> multicast domain 1 multicast domain 2:
>> [BB1--->R1--->R5--->R7] --- [R8---SW3---SW4]
>>
>> R1 is the RP for domain 1.
>> R8 is the RP for domain 2.
>> R1 and R8 are MSDP peers and the peer connections are UP:
>>
>> BB1 is joined to group 225.0.0.25
>> SW4 is the sender to 225.0.0.25.
>>
>> R7 and R8 are PIM enabled on their connected interfaces but are NOT
>> PIM neighbors. Otherwise, it would be one big domain.
>>
>> I'm gonna spoil my last mail by showing every 99th ping is successful:
>>
>> Cat3560-4#ping 225.0.0.25 re 1000
>>
>> Type escape sequence to abort.
>> Sending 1000, 100-byte ICMP Echos to 225.0.0.25, timeout is 2 seconds:
>> ..................................................................
>> Reply to request 66 from 100.100.100.100, 17
>>
ms..........................................................................
................
>> Reply to request 157 from 100.100.100.100, 17
>>
ms..........................................................................
................
>> Reply to request 248 from 100.100.100.100, 17
>>
ms..........................................................................
................
>> Reply to request 339 from 100.100.100.100, 16
>>
ms..........................................................................
................
>> Reply to request 430 from 100.100.100.100, 16
>>
ms..........................................................................
................
>> Reply to request 521 from 100.100.100.100, 17
>> ms........................................
>>
>> I have noticed the mroute entry on R8 is timing out. 150.100.34.14 is
>> the address of SW4's sending interface.
>>
>> R8#show ip mroute 225.0.0.25 | Beg \* (*, 225.0.0.25),
>> 00:18:09/stopped, RP 200.0.0.8, flags: SP
>> Incoming interface: Null, RPF nbr 0.0.0.0
>> Outgoing interface list: Null
>>
>> (150.100.34.14, 225.0.0.25), 00:02:56/00:00:03, flags: PA
>> Incoming interface: FastEthernet0/1, RPF nbr 150.100.81.13
>> Outgoing interface list: Null
>>
>> When this entry times out, I switch to SW4 and the ping is good one
>> time only. Then the entry is populated again immediately and the OIL
>> becomes NULL. I turned off SPT switchover on all routers with "ip pim
>> spt-threshold infinity"
>>
>> Any ideas?
>>
>> thanks,
Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Sat Nov 01 2008 - 15:35:18 ARST