[c-nsp] BGP memory on a 6500

[Buhrmaster, Gary]

> Folks who depended on such aggregation to keep their network 
> running, and who run it too close to the edge, might be in 
> for a rude shock when they change a policy and their router 
> blows up.

I think no one disputes this.  And that is why it should be 
an option, with the appropriate caveats just like enabling
portfast ("You have enable a feature which could have bad
side effects.  We trust you know what you are doing.  
Otherwise, you are screwed.  Have a nice day.").

That all said, I think the algorithms/heuristics are 
understood well enough that this is "just" a software 
engineering implementation issue when/if a decision
was to be made to do it (which, I have to admit, from
a BU point of view, what does it get me?  The only
possible benefit is good will of the customers, and
there are lots of costs).

There are days I wish IOS was "open" so you could
add in your own enhancements (such as this) in the
cases where Cisco did not agree that this was a wise 
thing to invest their efforts into.

Gary
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3376 bytes
Desc: not available
Url : https://puck.nether.net/pipermail/cisco-nsp/attachments/20061004/319b2e79/attachment.bin