Reverse look-up issue

[Michele Chubirka]

1. Thanks! The mysterious nscd cache. I've been reading up on this on
Sunsolve and it controls everything. (By the way, can I have the name of
that Solaris listserv?)
2. Looks like one of our SOA's is trying to make DNS entries for other
subnets, not associated with their authority, INCLUDING the problem subnet.
I think this may be our problem.

-----Original Message-----
From: Bill Moseley [mailto:moseley at hank.org]
Sent: Wednesday, September 13, 2000 4:41 PM
To: Michele Chubirka; bind-users at isc.org
Subject: Re: Reverse look-up issue


At 04:28 PM 09/13/00 -0400, Michele Chubirka wrote:
>
>We're a class "b" subnet running Bind 4.97 (in the process of upgrading,
but
>it's taking time), and are experiencing the following problem with one of
>our subnets: 1 minute(average)hangs when telneting/ftping to some of our
>Solaris and AIX3.2 boxes.

I don't know if this applies to you or not, but we had the exact same
problem on Solaris.  Telnet or ftp would often take a minute to connect.
One web server also had host name lookups enabled for logging, and boy was
that server slow.

I was instructed on the Solaris list that there were issues with nscd and
disabling host caching would fix the problem.  And it did.

What we did was uncomment this line in /etc/nscd.conf
        enable-cache            hosts           no

There may be other issues at work here, but there's something you can try.
The machine in question was not running named (DNS was provided by other
machines).

Good luck,



Bill Moseley
mailto:moseley at hank.org