Re: [pim] Question about TTL field in dense mode state refresh message

[Beau Williamson <bwilliam at cisco.com>]

Sowmya,

Why are you worrying about PIM-DM???

It is safe to say that PIM-DM is a dead-end protocol as it is
defined now.  No production network should be using it as it's
non-deterministic behavior often results in route loops that have been
known to melt down networks as well as blackhole traffic.  (Other
than those draw backs, I guess it is just fine.)

PIM-DM is now just an academic exercise or a protocol to test single
routers on a lab bench to see how fast they can replicate multicast
packets.

As for questions on TTL, there is no longer any need to set the TTL of a
multicast packet to anything but 255 when the application transmits
it.  This is because we have deprecated the use of TTL Thresholds as
a method of scoping multicast in favor of administratively scoped zones.

Beau Williamson
At 6/27/2006 05:31 PM, Sowmya.Krishnaswamy at nokia.com wrote:

Content-class: urn:content-classes:message
Content-Type: multipart/alternative;
         boundary="----_=_NextPart_001_01C69A39.655555D8"

Hello,

The PIM-DM rfc 3973 describes the pseudo code for forwarding the state refresh message in section 4.5.1. In the pseudo code there is a for loop where we walk the list of interfaces with pim nbrs and inside the loop we check if the TTL in the state refresh message is 0, we skip refreshing the prune state of this interface and skip forwarding the state refresh message out this interface.

Is it safe to skip processing the state refresh message when the TTL in the state refresh message is 0?

Could do this check outside the for loop in the pseudocode?

Thanks,
Sowmya


_______________________________________________
pim mailing list
pim at ietf.org
https://www1.ietf.org/mailman/listinfo/pim

_______________________________________________
pim mailing list
pim at ietf.org
https://www1.ietf.org/mailman/listinfo/pim