multicast

mcsemcsemcse

link local address 224.0.0.0/24 used on a local segment ttl=1 routers not forward these packet
how is forwardingbetween routers if the TTL is always 1 224.0.0.5 244.0.0.6 ospf 244.0.0.10 eigrp

nicklauscombs

correct me if im wrong here...

think about ospf on an ethernet segment which would include DR BDR and DROTHERS. The DR is multicasting information to the DROTHERS and vice versa so everyone is on the same page. with a ttl of 1 the multicasts stay on the local segment, if it was set higher the multicast could keep going and get onto other segments which would be a mess.

mcsemcsemcse

you right but what about if there are routers ospf or eigrp changing routing table over frame relay we can say this routers are all same segments ?or other protocol you doin redistribution between them

networker050184

Frame relay is a layer 2 protocol. It doesn't affect TTL which increments by layer 3 hop.

mcsemcsemcse

if we change frame relay with MPLS its layer 3 protocol how will exchange routing table with TTL1

networker050184

MPLS is not a layer 3 protocol either.....

mcsemcsemcse

why routers forward this class d range's between routers 244.0.0.5 , 244.0.0.6 ,244.0.0.9 , 244.0.0.10 except 244.0.0.1

networker050184

224.0.0.1 is all hosts and you don't want your update going to all hosts, you want it to go to the router(s) on the segment. You should pick up a book and do some reading on multicast and how routing protocols use it.

mcsemcsemcse

but still is TTL 1 for 244.0.0.5 , 6 , 10

networker050184

224.0.0.0 - 224.0.0.255 are reserved for network protocols and always have a TTL of 1. Like I stated before a little reading up on multicast should clear all these issues up for you. I (and others here) are glad to answer a few questions, but it seems as though you aren't even trying to look for this information yourself.

tech-airman

mcsemcsemcse wrote:

link local address 224.0.0.0/24 used on a local segment ttl=1 routers not forward these packet
how is forwardingbetween routers if the TTL is always 1 224.0.0.5 244.0.0.6 ospf 244.0.0.10 eigrp

mcsemcsemcse,

From what I've gathered, Multicast seems to be a kind of a "frame relay" technology. Let's take basic Frame Relay for starters.

Let's say you've got a basic Frame Relay channel of Source Frame Relay Router, Frame Relay Switch, and a Destination Frame Relay Router. Let's assume all three routers have been pre-configured for Frame Relay.

Now, remember that even though the term is "Frame Relay Switch" the networking device is a Router.

So let's assume you're pinging from the Source Frame Relay Router to the Destination Frame Relay Router, via the Frame Relay Switch. That means the Layer 3 ICMP Echo Request packet becomes encapsulated into a Layer 2 frame. Then that Layer 2 frame is sent out the serial interface.

[ICMP Echo Request Packet]
|
| Encapsulation process
v
[Frame Trailer][ICMP Echo Request Packet][Frame Header]
|
| Encapsulation process
v
[Frame Trailer][ICMP Echo Request Packet][Frame Header] ---> Forwarded out of Serial interface
[Source Frame Relay Router]----z----[Frame Relay Switch]

Now, remember that we're dealing with a Layer 3 packet that is encapsulated in a Layer 2 frame. Now, when the Frame Relay Switch receives the Layer 2 frame, since it is configured as a Frame Relay Switch, that frame is NOT decapsulated from frame to packet. Since the frame was never decapsulated from frame to packet, the TTL field of the packet remains at 1. Then, that Frame is relayed by the Frame Relay Switch to the Destination Frame Relay Router.

[Frame Relay Switch]---z---{Frame Trailer][ICMP Echo Request Packet][Frame Header]--->[Destination Frame Relay Router]

So it is when the frame is received by the Destination Frame Relay Router the frame is actually decapsulated from a frame to packet. When the frame becomes decapsulated to a packet is when the TTL field is inspected which still carries the value of 1, decremented by 1 so now TTL is 0. Fortunately, the Destination Frame Relay Router is the destination for the ICMP Echo Request packet, so it's travelled far enough based on TTL.

Well, just like Frame Relay, Multicast packets are encapsulated into Multicast Frames. It is then these Multicast frames that are relayed around by the various Multicast routers and Multicast switches to reach their destinations. So since Multicasts travel as frames, the TTL in the Multicast Packet inside the Multicast frame, like the basic Frame Relay example above, is unmodified by the Multicast routers that forwarded the Multicast frames. So that's why in Multicast, the TTL of 1 is usually enough. Within the context of EIGRP, a TTL of 1 would be enough for the local EIGRP router to reach it's EIGRP neighbor. Within the context of OSPF, a TTL of 1 would be enough for a DROTHER to reach the DR and for the DR to reach a DROTHER.

Does this explanation help?

networker050184

A multicast packet has a TTL just like a unicast packet. It is decreased on a per hop basis just like a unicast. So your explanation is not accurate. If your explanation was true, a multicast packet could theoretically loop in a network indefinitely.

tech-airman

networker050184 wrote:

A multicast packet has a TTL just like a unicast packet. It is decreased on a per hop basis just like a unicast. So your explanation is not accurate. If your explanation was true, a multicast packet could theoretically loop in a network indefinitely.

networker050784,

The TTL decrementing process itself is a loop prevention technology. That's what RIP uses to determine the "unreachable" state of 16 hops. However, each of those hops are from RIP router to RIP router, thus TTL is decremented. Multicast "routing" protocols such as PIM maintain loop free paths for multicast frames. Actually, the TTL set to 1 is a safety feature because if SOMEHOW the multicast frame is forwarded from a multicast router to a non-multicast router, the receiving non-multicast router would probably drop the multicast frame which would also be a loop avoidance feature. So explain how a multicast packet would practically and actually be caught in a loop when there's so many loop avoidance technologies involved?

networker050184

I think you are missing the point that not all multicast packets have a TTL set to one.