Multi-Interface Patch

Brian Aker brian at
Wed Feb 13 14:48:52 UTC 2008


On Feb 13, 2008, at 9:54 AM, Dustin Sallings wrote:

> 	Should we assume that every interface that binds for the text  
> protocol should also bind for the binary protocol?  That seems  
> pretty straightforward, and will probably be consistent with the UDP  
> strategy.

I am up for doing that! That is how I would prefer it be done. We  
should add a command (maybe stat?) that can tell us what the binary  
port is.

> 	I've not been paying much attention to this thread, honestly.  Is  
> there a short summary of the goal of explicit multiple interface  
> binding?  It doesn't seem like it does much over IN_ADDR_ANY in the  
> general case but become a multiplier for file descriptors (once  
> there's a binary UDP implementation, it'd be n interfaces * m layer  
> 3 protocols * 4).

With ipv6 the interface is more natural for binding to multiple  

For a binary UDP, we should default it on and to the same port as the  
binary TCP (in my opinion).

> 	I'm not so much questioning its value as I am curious.  Right now,  
> it just looks like a lot of work for me.  :)

Well, the one thing I have been pinged about is that this will allow  
someone to push more data if they have been maxing out a single card's  
interface. Mainly though, it just makes the interface complete.


Brian "Krow" Aker, brian at
Seattle, Washington                     <-- Me                <-- Software    <-- Fun
You can't grep a dead tree.

More information about the memcached mailing list