From: "Steven Levine" <ecs-isp@2rosenthals.com>
Message-ID: <646b8033.28.mr2ice.fgrirsq@earthlink.net>
Date: Mon, 22 May 2023 07:46:11 -0700
To: "eCS ISP Mailing List" <ecs-isp@2rosenthals.com>
In-Reply-To: <list-7173914@2rosenthals.com>
Subject: Re: [eCS-ISP] Odd dig behavior

In <list-7173914@2rosenthals.com>, on 05/22/23
   at 12:04 PM, "Massimo S." <ecs-isp@2rosenthals.com> said:

Hi Massimo,

>bind 9.11.28 write this:
>dig  +nssearch  pmoylan.org
>couldn't get address for 'ns1.wiz.net.au': failure
>couldn't get address for 'ns2.wiz.net.au': failure

Too bad.  That pretty much confirms this is a minor defect in our current
bind builds.

FWIW, there's another minor defect that only affects those running a
secondary dns.  A secondary DNS stores the zone files in binary and the
binary files are written in text mode, rather than binary.  This results
in failures like:

d:\clients\dnacih\servers\ns1\files\c\bind9\etc
>named-compilezone -d -fraw -Ftext -o- scoug.com named/scoug.com.zone
loading "scoug.com" from "named/scoug.com.zone" class "IN"
dns_master_load: bad label type
zone scoug.com/IN: loading from master file named/scoug.com.zone failed:
bad label type zone scoug.com/IN: not loaded due to errors.

named-checkconf fails for the same reason.

This also means that when a secondary DNS starts up, it will not be able
to read the zone files and will need to contact the primary DNS to refresh
the in memory copies of the zone files.

Steven

-- 
----------------------------------------------------------------------
"Steven Levine" <steve53@earthlink.net>  Warp/DIY/BlueLion etc.
www.scoug.com www.arcanoae.com www.warpcave.com
----------------------------------------------------------------------