From owner-freebsd-bugs  Sat May 13 15:37:30 1995
Return-Path: bugs-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.10/8.6.6) id PAA14243
          for bugs-outgoing; Sat, 13 May 1995 15:37:30 -0700
Received: from Root.COM (implode.Root.COM [198.145.90.1])
          by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id PAA14237
          for <freebsd-bugs@freefall.cdrom.com>; Sat, 13 May 1995 15:37:28 -0700
Received: from corbin.Root.COM (corbin.Root.COM [198.145.90.18]) by Root.COM (8.6.8/8.6.5) with ESMTP id PAA00486; Sat, 13 May 1995 15:40:33 -0700
Received: from localhost (localhost [127.0.0.1]) by corbin.Root.COM (8.6.11/8.6.5) with SMTP id PAA00131; Sat, 13 May 1995 15:37:29 -0700
Message-Id: <199505132237.PAA00131@corbin.Root.COM>
To: Charles Henrich <henrich@crh.cl.msu.edu>
cc: freebsd-bugs@freefall.cdrom.com
Subject: Re: bin/402: w -n doesnt work as advertised. 
In-reply-to: Your message of "Sat, 13 May 95 18:30:25 EDT."
             <199505132233.PAA00451@Root.COM> 
From: David Greenman <davidg@Root.COM>
Reply-To: davidg@Root.COM
Date: Sat, 13 May 1995 15:37:28 -0700
Sender: bugs-owner@FreeBSD.org
Precedence: bulk

>> >    I modified w.c to attempt to figure out the ip address of the hosts
>> >    that were non-IP.  Because some folks might need an option to make w
>> >    not do any resolver lookups, I also added -l which mirrors the
>> >    the existing -n flag, and ensures no nameserver calls get executed.
>>
>>    Actually, I strongly dislike the 4.4 "w" doing nameserver lookups by
>> default. On any reasonable system that has > 20 users on it, it can take
>> anywhere from several seconds to several minutes for the w output to finish.
>> In my opinion, the default for w should be reverted back to the old behavior
>> of trusting what is in wtmp.
>
>Must be a slow nameserver?  Most local nameervers will cache all retrieved data
>for an hour to a day, given that the responses should 80% of the time come from
>your local nameserver.  I never have any problems with the lookups, w never
>takes even a second to run.

   This works if the nameserver for all the domains is currently reachable. If
it isn't, it can take a minute for the query to timeout. About half the time I
do a 'w' on freefall, it takes > 1 minute to complete because of this. In
other cases, it's often that nameservers are slow to respond - 2 seconds isn't
unusual for non-cached entries. If this happens with just a few of the users,
the delay can approach 5-10 seconds. The problem grows as the number of users
increases. It's extremely annoying.

-DG