Root Servers not updated

Carl Oppedahl carl at oppedahl.com
Sat Jul 5 01:40:32 UTC 1997


At 02:22 PM 07/03/97 -0400, David H. Holtzman wrote:

>There's a lot of smoke going around this list and others about the root
>servers.  The internic generated a corrupt zone file on Tuesday which Bind
>did not load and prevented XFERs to the secondaries.  We became aware of it
>last night and it is currently fixed.  New zones will be ready for XFER
>within the next hour.  
>
>The issue with requiring a login for retrieving the zone files was designed
>to help reduce spamming.  We are currently experiencing a huge system load
>due to several individuals who are trying to register recently deleted
>domain names.  The requests are being repeated and fired off every few
>milliseconds.  One individual has over 50,000 templates in the system as we
>speak.  We believe that by requiring an identity to download the zone
>files we will better protect the community.
>
>Note that this ftp policy has not yet gone into effect.
>
>Lastly, it appears irresponsible to fan the flames of "the internic is
>trying to get over" via lists like these.  How about giving us the benefit
>of the doubt?  We generally stay silent on these issues, but feel that it
>is important to explain what we are doing and why.  No, we are not trying
>to take over the net.  Sorry for the confusion.

I didn't see any irresponsible fanning of flames.  I saw NSI (rather
characteristically, I might add) making what now turns out to be a secret
(you call it "silent") plan to change something about how information
passes from the A root server to the other root servers, and not sharing it
with the Internet community until someone (in this case, Paul Vixie as I
recall) noticed that NSI had screwed up and done it wrong.  And only after
someone outside of NSI noticed the screwup, did someone within NSI bother
to acknowledge it.  And only then did NSI fix it, as far as I can tell, by
the way.

If people had "given NSI the benefit of the doubt" as you request,
presumably by not posting anything about this screwup, it is anyone's guess
how long it would have taken for the screwup to get fixed.

I also didn't see any "smoke".  What I saw was factual discussions of how
the root level servers were out of synch with each other, indeed almost a
week out of synch before it got fixed.

If it is "important to explain what NSI is doing and why", then why did
this explanation not happen *before* NSI made this ill-fated change?  Why
did this explanation only happen after people outside of NSI discovered the
problem and commented upon it?





More information about the NANOG mailing list