On 12/13/2010 11:21 PM, Robert Story wrote:
> On Mon, 13 Dec 2010 11:16:54 -0500 Bill wrote:
> BF>  I'm not sure what to suggest as a solution.  For those who use
> BF>  net-snmp as the agent on a router, where asymmetric routing like this
> BF>  is very common, this black hole for the responses is a real problem.
>
> IMHO, support for asymmetric routing is more important that supporting
> broadcast queries.

Well, I'd like to have working both, I have many customers who require 
responses to broadcast packets and other really weird stuff.

> BF>  I think a better solution for responding to broadcast requests is to
> BF>  not use the if_index, and just don't specify the source address when
> BF>  the request was to a broadcast address.  [...]  Unfortunately,
> BF>  this means having to figure out if a given address is a broadcast, you
> BF>  have to get a list of interfaces on the system.
>
> Do we really need to figure out if it is a broadcast address? Or is it
> sufficient to see if the packet destination address does not match any of the
> addresses currently bound to the interface the packet came in on? If there's a
> match, don't use the if_index.

This could work, needs some testing though. And if you have list of IP 
addresses of an interface, you probably can get broadcast addresses (or 
netmask) easily. The problem is not to get this information, the real 
challenge is to do it effectively, i.e. cache them somewhere, refresh 
the cache sometimes, look up in the cache effectively, ... Or do we want 
to run netlink query on every outgoing UDP message?

Jan

------------------------------------------------------------------------------
Lotusphere 2011
Register now for Lotusphere 2011 and learn how
to connect the dots, take your collaborative environment
to the next level, and enter the era of Social Business.
http://p.sf.net/sfu/lotusphere-d2d
_______________________________________________
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to