On Thu, 24 May 2001, Joe Abley wrote:
On Thu, May 24, 2001 at 03:53:30PM +1200, Peter Mott wrote:
Thanks to those who chipped in to help sort this problem. We are now back on the air. I have come to the conclusion that reliable domestic peering is not something we should expect anytime soon.
It doesn't sound like domestic peering was the issue; more like "transit through netgate is broken". A sane network which didn't learn a route through some domestic method would send the traffic overseas to reach its destination rather than simply dropping it.
Mmm, not quite. The problem is at layer 2 - Attica learn routes from Clear, who learn them from Netgate, and so Attica have a bunch of routes that direct them to directly traverse APE to Netgate's router. Unfortunately, at this time we *do not* have layer 2 connectivity between Attica and Netgate - they can both ping most everything else on APE, but they can't ping each other. Reasons for this are unclear - it seems to involve spanning tree, VLAN (mis)configuration, increasing topology complexity, and the phase of the mooon - according to Peter M, it seems to come and go in 20 minute cycles. It certainly changes depending on who does or doesn't advertise spanning tree packets. *This* is why I'm harping on about spanning tree and total filtering of connected switches - because right here, right now, I believe it's causing APE to go like crap. Cheers Si --------- To unsubscribe from nznog, send email to majordomo(a)list.waikato.ac.nz where the body of your message reads: unsubscribe nznog