A couple of things on first pass..
You should only run sparse-dense on internal interfaces, AND ONLY IF you
are using auto-RP to announce RP addresses to all your internal routers.
Even so, I must recommend against it. Whenever possible, use static RP
assignments.
Next, your PIM config needs to set a rib-group to inet.2 since your MBGP
peering is populating inet.2, PIM then needs to RPF against inet.2. Also,
your IGP, interface routes, and any statics also need to be installed into
inet.2 Something like this:
groups {
statics {
routing-options {
rib <inet.*> {
static {
route 198.58.3.0/24 discard;
route 0.0.0.0/0 discard;
}
}
}
}
}
routing-options {
interface-routes {
rib-group inet if-rib;
}
rib inet.0 {
apply-groups statics;
}
rib inet.2 {
apply-groups statics;
}
rib-groups {
mcast-rib {
export-rib inet.2;
import-rib inet.2;
}
ospf-rib {
export-rib inet.0;
import-rib [ inet.0 inet.2 ];
}
if-rib {
import-rib [ inet.0 inet.2 ];
}
}
}
protocols {
pim {
rib-group inet mcast-rib;
rp {
local {
address 198.58.3.254;
}
}
interface all {
mode sparse;
version 2;
}
}
}
..and as you see directly above, set the PIM version to 2 on your
interfaces also.
One more note, when assiging a static RP on your Juniper routers, be sure
to also set the RP version to 2.
I hope this helps. Let me know if you'd like me to work directly with
anyone on this.
Greg
On Fri, 15 Jun 2001, Aviva Garrett wrote:
> This question came in on juniper-nsp. Does anyone want to respond (if
> it's appropriate -- juniper-nsp@puck.nether.net)?
>
> Thanks,
> ..Aviva
>
> In message <Pine.BSI.4.30.0106150014440.29239-100000@malasada.lava.net>you writ
> e:
> > Greetings,
> >
> > I was hoping someone more experienced with Juniper routers could assist me
> > in configuring an M10 with which I'm quite the novice with. I'm trying to
> > move 2 multicast tunnels and 1 direct peering connection from a cisco to a
> > Juniper M10 router but for some reason can't get it to work.
> >
> > The interface portion of the config has:
> >
> > fe-0/0/0 {
> > description "Fast Ethernet on eth-srv VLAN";
> > unit 0 {
> > family inet {
> > no-redirects;
> > primary;
> > address 64.65.64.30/25;
> > }
> > }
> > }
> > t3-0/1/0 {
> > description "Sprint, 82.HFGS.710001.GTEW";
> > encapsulation cisco-hdlc;
> > t3-options {
> > compatibility-mode larscom subrate 2;
> > no-payload-scrambler;
> > cbit-parity;
> > }
> > unit 0 {
> > family inet {
> > address 160.81.200.30/30;
> > }
> > }
> > }
> > gr-0/3/0 {
> > unit 0 {
> > description "MBone tunnel to WorldCom/UUNET";
> > tunnel {
> > source 64.65.64.149;
> > destination 208.205.11.91;
> > }
> > family inet {
> > address 157.130.204.122/30;
> > }
> > }
> > unit 1 {
> > description "Multicast tunnel to Yahoo Broadcast (broadcast.com)";
> > tunnel {
> > source 64.65.64.149;
> > destination 206.190.40.61;
> > }
> > family inet {
> > address 206.190.40.186/30;
> > }
> > }
> > }
> > lo0 {
> > unit 0 {
> > family inet {
> > address 64.65.64.148/32;
> > address 127.0.0.1/32;
> > address 64.65.64.149/32;
> > }
> > }
> > }
> >
> > The pertinent part of the protocol config looks like:
> >
> > bgp {
> > group Sprint {
> > family inet {
> > any;
> > }
> > peer-as 1239;
> > neighbor 160.81.200.29;
> > }
> > group tunnels {
> > family inet {
> > multicast;
> > }
> > neighbor 157.130.204.121 {
> > peer-as 704;
> > }
> > neighbor 206.190.40.185 {
> > peer-as 5779;
> > }
> > }
> > }
> > msdp {
> > peer 144.228.240.253 {
> > local-address 64.65.64.149;
> > }
> > peer 206.190.40.61 {
> > local-address 64.65.64.149;
> > }
> > peer 157.130.204.121 {
> > local-address 157.130.204.122;
> > }
> > }
> > pim {
> > dense-groups {
> > 224.0.1.39/32;
> > 224.0.1.40/32;
> > }
> > rp {
> > local {
> > address 64.65.64.148;
> > priority 250;
> > }
> > auto-rp mapping;
> > }
> > interface all {
> > mode sparse-dense;
> > }
> > }
> >
> > The MBGP peering comes up:
> >
> > Groups: 4 Peers: 5 Down peers: 0
> > Table Tot Paths Act Paths Suppressed History Damp State Pendi
> > ng
> > inet.0 278399 102804 50 150 778
> > 0
> > inet.2 6373 3751 1083 10 1774
> > 0
> > Peer AS InPkt OutPkt OutQ Flaps Last Up/Dwn State
> > |#Active/Received/Damped...
> > 64.65.64.2 6435 30416 6010 0 0 22:28 83542
> > /83565/0 21/21/0
> > 64.65.64.66 6435 36106 3454 0 0 22:23 18340
> > /94331/0 0/0/0
> > 157.130.204.121 704 1014 48 0 3 22:08 0/0/0
> > 179/2788/1083
> > 160.81.200.29 1239 38438 56 0 0 18:07 922/1
> > 00503/50 3551/3561/0
> > 206.190.40.185 5779 85 84 0 0 40:25 0/0/0
> > 0/3/0
> >
> > The MSDP sessions come up:
> >
> > Peer address Local address State Last up/down Peer-Group
> > 144.228.240.253 64.65.64.149 Established 00:19:47
> > 157.130.204.121 157.130.204.122 Established 00:09:10
> > 206.190.40.61 64.65.64.149 Established 00:40:48
> >
> > The ciscos on the same ethernet see the Juniper as an RP:
> >
> > PIM Group-to-RP Mappings
> >
> > Group(s) 224.0.0.0/4
> > RP 64.65.64.148 (iiwi.lava.net), v2
> > Info source: 64.65.64.148 (iiwi.lava.net), via Auto-RP
> > Uptime: 00:09:40, expires: 00:02:14
> >
> > >From a cisco I can run an mtrace between the Juniper to one of our
> > upstreams:
> >
> > mamo>mtrace iiwi 144.228.240.253
> > Type escape sequence to abort.
> > Mtrace from 64.65.64.148 to 144.228.240.253 via RPF
> > From source (iiwi.lava.net) to destination (rp-stk.sprintlink.net)
> > Querying full reverse path...
> > 0 rp-stk.sprintlink.net (144.228.240.253)
> > -1 sl-bb20-stk-14-0.sprintlink.net (144.232.4.234) PIM/MBGP [64.65.64.0/22]
> > -2 sl-bb21-stk-14-0.sprintlink.net (144.232.4.233) PIM/MBGP [64.65.64.0/22]
> > -3 sl-bb22-stk-15-0.sprintlink.net (144.232.4.242) PIM/MBGP [64.65.64.0/22]
> > -4 sl-bb21-prl-11-3.sprintlink.net (144.232.8.218) PIM/MBGP [64.65.64.0/22]
> > -5 sl-gw1-prl-12-0-0.sprintlink.net (144.232.30.3) PIM/MBGP [64.65.64.0/22]
> > -6 sl-lavanet-1-0.sprintlink.net (160.81.200.30) PIM [64.65.64.148/32]
> > -7 iiwi.lava.net (64.65.64.148)
> > mamo>mtrace 144.228.240.253 iiwi
> > Type escape sequence to abort.
> > Mtrace from 144.228.240.253 to 64.65.64.148 via RPF
> > From source (rp-stk.sprintlink.net) to destination (iiwi.lava.net)
> > Querying full reverse path...
> > 0 iiwi.lava.net (64.65.64.148)
> > -1 iiwi.lava.net (64.65.64.148) PIM [144.228.0.0/16]
> > -2 sl-gw1-prl-1-1-1.sprintlink.net (160.81.200.29) PIM [144.228.240.253/32]
> > -3 sl-bb21-prl-2-0.sprintlink.net (144.232.30.2) PIM/MBGP [144.228.240.253/
> > 32]
> > -4 sl-bb22-stk-2-1.sprintlink.net (144.232.8.217) PIM/MBGP [144.228.240.253
> > /32]
> > -5 sl-bb21-stk-15-0.sprintlink.net (144.232.4.241) PIM/MBGP [144.228.240.25
> > 3/32]
> > -6 sl-bb20-stk-14-0.sprintlink.net (144.232.4.234) PIM [144.228.240.253/32]
> > -7 rp-stk.sprintlink.net (144.228.240.253)
> >
> > But no multicast traffic flows into our network. If anyone has any ideas
> > I'd appreciate hearing from you. BTW, the tunnels and Sprint connection
> > work when they're terminated on a cisco so the upstreams are configured
> > ok. I suspect I'm overlooking something very obvious...
> >
> >
>
This archive was generated by hypermail 2b29 : Mon Aug 05 2002 - 10:42:37 EDT