[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


To: <dnsop@cafax.se>
From: Daniel Senie <dts@senie.com>
Date: Tue, 19 Mar 2002 10:33:41 -0500
In-Reply-To: <Pine.BSF.4.30.0203190556360.22662-100000@spider.nic-se.se>
Sender: owner-dnsop@cafax.se
Subject: Re: draft-ietf-dnsop-v6-name-space-fragmentation-01.txt

At 12:03 AM 3/19/02, Mats Dufberg wrote:
>The draft suggests that the master server checks that all delegations each
>time it loads the zone. The check of the delegations includes making sure
>that all nameservers in the delegation is reachable.
>
>It is unrealistic to expect an ordinary TLD to do such frequent checks. A
>delegation should be varified each time it is updated, and maybe at other
>times. And such checks could preferably use a separate resolver if the
>master server does not resolv.

Beyond "unrealistic," it would seem this could be quite damaging, and not 
just in the TLD case. Any large organization may have very large number of 
delegations. A flood of packets to verify the function of delegated servers 
represents a serious problem. The time delay to initiate thewe checks at 
service load/reload (and periodic checks thereafter?) may render the 
service unavailable to answer queries.

Might it not be better to think in terms of a separate function (perhaps 
implemented as a separate thread or subprocess of the name service, perhaps 
as a separate entity) which takes a slow, continuous walk around the name 
space looking for and reporting errors? This would eliminate the need to 
make the checks at start-up point, permitting the service to begin and 
continue functioning without being impeded by checks of other systems.


-----------------------------------------------------------------
Daniel Senie                                        dts@senie.com
Amaranth Networks Inc.                    http://www.amaranth.com


Home | Date list | Subject list