policy routing and netflow (was Re: karl and paul, expostulating )

Brett D. Watson bwatson at genuity.net
Thu Feb 20 22:37:05 UTC 1997


> On Feb 20, 1997, Brett D. Watson wrote:
> > 
> >   doesn't matter.  in production ios, policy routing (source based 
> > routing) is process switched.  there is code in the works to make it 
> > fast switched.  but  there is a bug wherein if you do policy routing, 
> > and you enable flow or optimum switching on the interface you're 
> > doing pr on , it disables the policy routing.
> > 
> >   that bug may be fixed now but in any case enabling flow switching 
> > will *not* speed up policy routing.  and if you're exporting the flow 
> > stats, you lose anywhere from 50kpps to 100Kpps of speed.
> 
> I have news for you; this isn't policy routing!  

  you better tell cisoc that then :)  they call source based routing 
"policy routing".  bad choice of names i guess but that's what they 
call it.  it's an inbound route map applied to an interface and you 
look at the *source* address of the packet, then use various "set" 
clauses based on that address.

> We aren't re-writting
> any source or destination addresses (which is what policy routing
> does).  

  i have news for you, policy routing has nothing to do with 
re-writing source or destination addresses.

>  We're just filtering based on source and destination
> parameters (such as address, protocol, port, etc).

  if you're filtering based on source-anything, and you're using a 
cisco, i'd like to know how you're doing it without policy 
route-maps.  please see:

www.cisco.com/warp/public/732/Tech/plicy_wp.htm

  i didn't imply that you were doing anything.  i may have 
misundestood justin but i thought he was implying that netflow 
switching would increase the switching speed of policy based routing, 
as cisco calls it, and it does not.  that's all i was getting at.

> Flow switching works very effectively (at least as of IOS 11.1.9).

  yes it does.  i never said it didn't work well.  who's mail were 
you reading anyway?

-brett






More information about the NANOG mailing list