[76855] in North American Network Operators' Group
Re: Anycast 101
daemon@ATHENA.MIT.EDU (Paul Vixie)
Sat Jan 1 20:25:04 2005
To: nanog@merit.edu
From: Paul Vixie <vixie@vix.com>
Date: 02 Jan 2005 00:51:15 +0000
In-Reply-To: <Pine.GSO.4.61.0412312014320.3963@pants.snark.net>
Errors-To: owner-nanog-outgoing@merit.edu
> i've also been thinking that AXFR's known incoherency could be reduced
> by using some kind of in-band embargo that would bring a new zone
> version online synchronously on servers supporting this feature and
> configured to enable it for a particular zone.
>
> Or a different storage abstraction for your zone data. Flat text zone
> files are so 90's. How about an rdbms backend on each nameserver, with
> updates delivered 'reliably' by a message queue service. It sounds a lot
> easier than a bunch of protocol additions.
yes, it's easier to do something vendor-specific. with things like dlz-bind
or the other sql backends for bind9, it can be done today. however, my goal
(and isc's goal) is to advance the state of the art, not just by providing a
reasonable implementation of the current protocol and then adding a bunch of
vendor-specific extensions, but by advancing the state of the protocol, such
that multi-vendor interoperability will be possible even when using advanced
features. this requires "a bunch of protocol additions", which are indeed a
lot less easy than just doing vendor-specific extensions.
--
Paul Vixie
