[3054] in Release_7.7_team
Re: named non-debugability problems, again
daemon@ATHENA.MIT.EDU (Greg Hudson)
Tue Nov 27 15:32:34 2001
From: Greg Hudson <ghudson@MIT.EDU>
To: John Hawkinson <jhawk@mit.edu>
Cc: bugs@mit.edu, release-team@mit.edu
In-Reply-To: <200111271910.OAA13081@multics.mit.edu>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Date: 27 Nov 2001 15:32:24 -0500
Message-Id: <1006893144.29943.2.camel@error-messages.mit.edu>
Mime-Version: 1.0
On Tue, 2001-11-27 at 14:10, John Hawkinson wrote:
> Hi. I'm not still not sure if this is the right place for this,
> and suggest still reelly feels wrong to me.
release-team is fine, but bugs is pointless for a deliberate behavior.
> I encountered this issue again today, where a user called the SIPB
> office with a linux machine that would not attach lockers. As best I
> could tell, it came down to a DNS or service switch problem problem on
> his machine, but I was unable to usefully debug it through him, and
> could think of nothing to do short of visitting his machine or
> requesting an account on it, neither of which seemed reasonable. So I
> ended up directing him to send mail to linux-help.
When we made the decision, we knew the benefit and we knew the cost.
Simply pointing out isolated examples of the cost is unlikely to change
our minds. If, say, Oliver were to say "this is seriously impacting our
ability to help users," that would be new information, but we expected
that people like you would occasionally miss the ability to remotely
debug named.
What we could do is add an athinfo query to check for something
specific. That wouldn't really present security problems, other than
perhaps aiding a DNS spoofing attack.
