On Fri, 3 Nov 2000, Joe Shaw wrote:
If they're filtering on the /19 or /20 boundry on legacy space, they're VERY misconfigured and breaking a whole bunch of connectivity.
I thought filtering at /19-/20 space was still considered best common practice by some of the older Tier 1's who were interested in keeping the routing tables small. Maybe with less legacy equipment deployed this isn't an issue anymore.
I don't think it's a matter of legacy equipment and many networks do filter on the /19-/20 boundry for address space that is assigned as shorter than /21. I don't know of anyone who is _purposely_ filtering out legacy /24's though. By legacy /24's, I am referring to address space that was allocated by the RIR as /24's. The chances of those /24's being aggregated into /20 or shorter prefixes are pretty slim.
Wouldn't it be better, at least from an engineering standpoint, to still announce their routes with AS padding to increase the AS-path so in the event their other connection(s) goes down they still have some type of inbound connectivity? It seems like your example would work in a best case scenario, but customer X would drop off of the planet in the event of a partial outage without some manual reconfiguration. I did something similar to what you are suggesting, but we still announced the routes, with padding, so that in the event of a failure the network could still function. The link did fail eventually (would you believe me if I mentioned there was a backhoe and a contractor involved?), and while the network was certainly slower than normal, it continued to function adequately so that there was no perceivable outage seen by our customers.
I agree that it would be better. I know of several instances where people are doing it as I described though, for whatever reasons. --- John Fraizer EnterZone, Inc