Re: Was multicast routing behavior changed in new 15.0T code?

Answering myself, FTR:

http://www.ciscosystems.com/en/US/docs/ios/ipmulti/configuration/guide/imc_mfib_overview_ps10591_TSD_Products_Configuration_Guide_Chapter.html

describes the new "IPv4 enabled CEF mcast forwarding", and the MFIB.
It introduces the PIM tunnel interfaces used for SM registration.

BTW, I used to think that CEF was linked to "push" control as opposed to
"pull" (aka on demand) control. Well, mcast CEF is on demand, so I would
have called it fast switching.

-Carlos

Carlos G Mendioroz @ 18/8/2010 12:15 -0300 dixit:
> Jorge,
> have you received any info on this ?
> CEF forwarding for mcast seems a new feature of 15, and it would be
> nice to have a piece of documentation describing it.
> I see that CEF can be disabled on *output* to an interface... strange.
> Please post your findings !
>
> -Carlos
>
> Jorge Cortes @ 16/8/2010 23:57 -0300 dixit:
>> Hi Armin,
>>
>> Thanks for taking the time for looking and responding to this query. The
>> racks I use for studying were upgraded to 15.0 code two weeks ago. Those
>> have the full license set, I'm sure of it.
>> My question however is more focused on the actual behavior, which
>> looks like
>> it has been changed. It looks like you have to disable Multicast CEF
>> for the
>> routers to learn the RP mappings, and also a tunnel interface is now
>> created. At least this is what I observed while going thru this exercise,
>> but I am just curious to know if anybody has tried this and what are the
>> outcomes.
>>
>> Thanks,
>> Jorge
>>
>> On Mon, Aug 16, 2010 at 6:08 PM, Armin Mirsepassi
>> <amirsepassi_at_ccgrp.com>wrote:
>>
>>> Jorge,
>>>
>>> 15.x Quick read faq:
>>>
>>> http://www.cisco.com/en/US/prod/collateral/iosswrel/ps8802/ps10587/ps10591/p
>>>
>>> s10621/qa_c67_561940.html<http://www.cisco.com/en/US/prod/collateral/iosswrel/ps8802/ps10587/ps10591/p%0As10621/qa_c67_561940.html>
>>>
>>>
>>> I havnt bought or used 15.x with ISR Gen 1, I will take a risk and
>>> assume
>>> it's the same as with Gen 2, unless im missing something. Look at
>>> pictures
>>> in below ordering doc for a little better explanation of the license
>>> features, pretty much same as above.
>>>
>>> http://www.cisco.com/en/US/prod/collateral/modules/ps10598/ordering_guide_c0
>>>
>>> 7_557736_ps10537_Products_Data_Sheet.html<http://www.cisco.com/en/US/prod/collateral/modules/ps10598/ordering_guide_c0%0A7_557736_ps10537_Products_Data_Sheet.html>
>>>
>>>
>>> Otherwise use the feature navigator tool (http://www.cisco.com/go/fn) to
>>> make sure everything you need is available with the licenses you have
>>> purchased.
>>>
>>> If you're using for studies you can request a temp 60 day license
>>> (http://www.cisco.com/go/license click temp license link at top of page)
>>>
>>> Armin
>>>
>>> -----Original Message-----
>>> From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of
>>> Jorge Cortes
>>> Sent: Monday, August 16, 2010 1:06 PM
>>> To: Cisco certification
>>> Subject: Was multicast routing behavior changed in new 15.0T code?
>>>
>>> Hi GS Gurus.
>>>
>>> Just ran into an strange issue with multicast routing where the router
>>> wasn't getting the RP mapping information. The topology is very
>>> simple as
>>> it
>>> is a daisy chain and no RPF failures:
>>>
>>> SW---SW(RP)---SW---R
>>>
>>> Router interface configuration
>>>
>>> interface FastEthernet0/1
>>> ip address 61.61.128.193 255.255.255.224
>>> ip pim sparse-mode
>>> ip ospf mtu-ignore
>>> duplex auto
>>> speed auto
>>> ipv6 address 2001:61:100::4/64
>>> ipv6 rip RIPNG enable
>>> !
>>>
>>> R4(config)#do sh ip pim rp map
>>> PIM Group-to-RP Mappings
>>>
>>> R4(config)#
>>>
>>> So I turned on "debug ip mpacket" on the router to see what was going
>>> on,
>>> and also tried to disable "mroute cache" as follows:
>>>
>>> R4(config)#int fa0/1
>>> R4(config-if)#no ip mroute-ca
>>> The above command is going to be deprecated. Use MFIB commands instead.
>>>
>>> Well, looks like this command is not supported anymore, so lets use
>>> the new
>>> MFIB commands.
>>>
>>> R4(config-if)#no ip mfib ?
>>> cef Multicast CEF to/from interface
>>> forwarding Multicast forwarding to/from interface
>>>
>>> Not sure the difference between both, so will disable both. The final
>>> configuration of the interface looks as follows:
>>>
>>> interface FastEthernet0/1
>>> ip address 61.61.128.193 255.255.255.224
>>> ip pim sparse-mode
>>> no ip mfib forwarding input
>>> no ip mfib forwarding output
>>> no ip mfib cef input
>>> no ip mfib cef output
>>> ip ospf mtu-ignore
>>> duplex auto
>>> speed auto
>>> ipv6 address 2001:61:100::4/64
>>> ipv6 rip RIPNG enable
>>> !
>>>
>>> And right after I'm done with that configuration -which was suppossed
>>> to be
>>> used only for debugging purposes- I get the following message:
>>>
>>> R4(config-if)#
>>> *Aug 15 22:54:16.490: %LINEPROTO-5-UPDOWN: Line protocol on Interface
>>> Tunnel0, changed state to up
>>>
>>> And now the RP mappings appear to be working:
>>>
>>> R4(config-if)#do sh ip pim rp map
>>> PIM Group-to-RP Mappings
>>>
>>> Group(s) 239.10.5.0/24
>>> RP 150.1.9.9 (?), v2v1
>>> Info source: 150.1.10.10 (?), elected via Auto-RP
>>> Uptime: 00:01:52, expires: 00:02:05
>>>
>>> Not sure what exactly is going on with the new code, but I had never
>>> seen
>>> this before. First looks like the behavior is now like in IPv6
>>> multicast,
>>> where a tunnel interface is created, but I don't understand why it was
>>> necessary to turn off MFIB for it to work.
>>>
>>> Is anybody aware of this behavior? I couldn't find it documented.
>>>
>>> R4(config-if)#do sh ver
>>> Cisco IOS Software, 2800 Software (C2800NM-ADVIPSERVICESK9-M), Version
>>> 15.0(1)M2, RELEASE SOFTWARE (fc2)
>>>
>>> Thanks,
>>>
>>> Jorge
>>>
>>>
>>> Blogs and organic groups at http://www.ccie.net
>>>
>>> _______________________________________________________________________
>>> Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>>
>>
>>
>>
>>
>>
>>
>

-- 
Carlos G Mendioroz  <tron_at_huapi.ba.ar>  LW7 EQI  Argentina
Blogs and organic groups at http://www.ccie.net