Re: No SOLUTION till DATE for this SCNERIO ???? Challenging

OK, well it's approaching my bedtime here after a very long day but
please let me ask you to clarify a couple of things and then maybe
others in more favorable time zones can pick up the ball and be of
further assistance before I roll back out of the sack tomorrow:

1. When you say "outgoing interface" below, do you actually mean RPF
interface?
2. Are there any restrictions preventing you from making use of
static mroutes? Other restrictions?

I agree that it's a "basic thing which is not coming up," which is why
I encourage a thorough review of some of the mcast troubleshooting
tutorials that are freely available. Multicast is not actually all
that difficult (granted, sometimes less than intuitive when approached
from a unicast mindset!) if you consistently apply fundamental mcast
logic in your approach to troubleshooting. You need to fully
understand how mcast traffic is flowing and how the RPF checks are
taking place/succeeding prior to any modifications and then you simply
need to track how the scenario changes following said modifications
(and be 100% certain that you are making the proper cause/effect
association!). Then you need to resolve the problem within the
constraints of your given lab, with which I am completely unfamiliar
(and please bear in mind that I never did complete (nor attempt) the
SP track - not even close! ;-)).

Just remember that tunnels and static routes - and especially static
routes pointing down tunnels - can be a network engineer's best
friend, so long as we aren't somehow arbitrarily restricted from using
these magic potions!

____________________________________________
There are only 10 types of people in the world:
Those who understand binary and those who do not...

On Feb 2, 2011, at 7:31 , Asfer Shah wrote:

> Thanks
> Scoottt, I surely will try to check the tutorials but the problem is
> that Its very basic thing which is not coming up ?
> I am trying to run the Multicast in INTRA-AS between my two sites(R9
> and R7/R8) connected via MPLS - VPN's . (R4 is RP and client & R5 is
> SOURCE)
> Everything runs smoothly unless, I introduce the default route from
> the (R6) which is the internet gateway for my PE's of ABC Vpn's.
>
> Whenever the default route is injected my PE outgoing interface
> donot point towards the tunnel and starts to follow the 0.0.0.0
> source !! which breaks my multicast communication and messes
> everything !
> I can share the details with you even further, I discussed this
> scenerio will couple of other CCIE's but somehow no one has managed
> to answer my solution ?
> thanks
>
> CC: markom_at_ipexpert.com; ccielab_at_groupstudy.com
> From: scott_ccie_list_at_it-ag.com
> To: s_asfar_at_hotmail.com
> Subject: Re: No SOLUTION till DATE for this SCNERIO ???? Challenging
> MULTICAST TASK ?? Please help
> Date: Wed, 2 Feb 2011 11:17:11 -0700
>
> So it looks as if you've got some vendor's full-scale SP lab up and
> running there. Lots of layers to an onion like that which are not
> going to be readily apparent with just a glance over the configs.
> It's also not evident from your text exactly what was working before
> the introduction of the default route came along and broke things.
> I see you've got a couple of 'ip igmp join-group' commands on your
> CEs but it's not clear to me what you were pinging - and from where
> - that worked and then later ceased working. What mcast
> troubleshooting have you done thus far?
>
> I try to keep my posts as vendor-neutral as possible but if you
> google "multicast troubleshooting" you will find a couple of very,
> very good tutorials out there. If you follow a logical process of
> validating all of your various control and data plane RPF checks,
> etc, you are bound to find the problem for yourself quickly...
>
> ____________________________________________
> There are only 10 types of people in the world:
> Those who understand binary and those who do not...
>
> On Feb 2, 2011, at 2:50 , Asfer Shah wrote:
>
>
> thanks for the quick Reply, Actually I am appearing in les than a
> week time and I am failing to understand this buG !!
> nothing much I could figure out !!!
> I am hell stuck,since either multcast or Internet will wok at a
> single time !!
>
> CE1 is dual homed to PE5 and PE3, we have three GRE (MDT) tunnels
> between PE2 , PE 5 and PE3
> thanks
>
>
> |--------------------------
> PE5----------
>
> | |
> > > INT---------PE1--------PE2--------------------------PE3--------CE1
> > > | |
> > > | |
> > > PE4 CE2
>
>
>
> > From: markom_at_ipexpert.com
> > Date: Wed, 2 Feb 2011 10:16:40 +0100
> > Subject: Re: No SOLUTION till DATE for this SCNERIO ????
> Challenging MULTICAST TASK ?? Please help
> > To: s_asfar_at_hotmail.com
> > CC: bmcgahan_at_ine.com; scott_ccie_list_at_it-ag.com; ccielab_at_groupstudy.com
> ; narbikk_at_gmail.com
> >
> > Can you please post all the relevant configurations to the list so
> we
> > can actually troubleshoot it? :-)
> >
> > --
> > Marko Milivojevic - CCIE #18427
> > Senior Technical Instructor - IPexpert
> >
> > FREE CCIE training: http://bit.ly/vLecture
> >
> > Mailto: markom_at_ipexpert.com
> > Telephone: +1.810.326.1444
> > Web: http://www.ipexpert.com/
> >
> > On Wed, Feb 2, 2011 at 09:03, Asfer Shah <s_asfar_at_hotmail.com>
> wrote:
> > > Hello ,
> > > I have a Query to which no one have answered so far ??
> > >
> > > PE2-PE3 are in AS20 and are MP-iBGP neighbor to each other.
> > > CE1 and CE2 are two sites connected to PE2 and PE3 respectively
> talking
> > > through the MPLS VPN cloud.
> > > The multicast distribution trees are also configured and the
> Intra-AS
> > > multicast end-to-end is working smoothly fine between CE1-CE2.
> > >
> > > I have the below topology in which I have a trouble.
> > >
> > >
> > >
> > > INT---------PE1--------PE2-------PE3--------CE1
> > > | |
> > > | |
> > > PE4 CE2
> > >
> > >
> > > Now,I have PE1 in AS 10 which is gateway to the internet and is
> peering with
> > > PE2 for ipv4 (add-fam) only.
> > > I have PE4 in (AS30) which is peering with PE1 for both ipv4 and
> vpnv4
> > > (add-families)
> > > PE4 is Multihop Mp-eBGP neighbor to both the PE2 and PE3.
> > >
> > > PE1 is advertising the internet (0.0.0.0 in vpnv4 add family) to
> PE4 and then
> > > PE4 is throwing it to PE2 and PE3
> > > After manipulating the Next-Hop , PE2 and PE3 do watch the
> internet default
> > > vpnv4 route 0.0.0.0 coming from PE1 and they do access the
> internet easily.
> > >
> > > At this point when I check back the multicast between CE1 and
> CE2 , it will
> > > stop working and the traffic is dropped
> > > The momnet , I remove the 0.0.0.0 route adver in the vpnv4 from
> PE1 , my
> > > multicast will run perfectly fine.
> > >
> > > There is PIM running between PE1 PE2 and PE4 but no solution to
> the problem
> > > till date !!
> > > Can you please help me in this special problem ???
> > >
> > >
> > > Blogs and organic groups at http://www.ccie.net
> > >
> > >
> _______________________________________________________________________
> > > Subscription information may be found at:
> > > http://www.groupstudy.com/list/CCIELab.html
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> <CE1.cfg><CE2.cfg><PE1.cfg><PE2.cfg><PE3.cfg><PE4.cfg><PE5.cfg>

Blogs and organic groups at http://www.ccie.net
Received on Wed Feb 02 2011 - 20:54:53 ART