Dumb Question is an A or AAAA record required?

Matthew Richardson matthew-l at itconsult.co.uk
Thu Jul 9 14:06:27 UTC 2020

On a related issues there were (perhaps long ago) issues if the A record
for a domain had an SMTP server on it, where email could sometimes be
delivered to that A record rather than the MX.  I had (again long ago:
10-15 years) actually seen this occur.

Do people think that this problem could still occur these days?  What sort
of transient (presumably DNS) failure might cause an SMTP server to deliver
to A rather than MX?

Best wishes,

>From: Anand Buddhdev <anandb at ripe.net>
>To: "@lbutlr" <kremels at kreme.com>, bind-users <bind-users at lists.isc.org>
>Date: Thu, 9 Jul 2020 14:43:04 +0200
>Subject: Re: Dumb Question is an A or AAAA record required?

>On 09/07/2020 14:21, @lbutlr wrote:
>> Given a domain that is hosted and used for email and web, is an A
>> record for that domain actually required?
>It's not *required*. But see below.
>> That is, if bob.tld is hosted by example.com can you simply have
>> 	NS ns1.example.com
>> 	NS ns2.example.com
>> 	MX mx.example.com
>> www	CNAME www.example.com
>> Without specifying
>> 	A
>These days, many folk try to reach websites by typing just the bare 
>domain name without the "www" prefix.
>If a user types "bob.tld" into a browser, the browser will issue an 
>address lookup for "bob.tld", causing the resolver to ask for A and AAAA 
>records for "bob.tld". If you don't have an A record at the zone apex, 
>the browser will not get back any address and display an error message 
>for the user. An alert user might try "www.bob.tld" but most users are 
>likely to just give up.
>So while it's not *required* to have an address record at the apex, it's 
>good practice to have one.
>Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe from this list
>ISC funds the development of this software with paid support subscriptions. Contact us at https://www.isc.org/contact/ for more information.
>bind-users mailing list
>bind-users at lists.isc.org

More information about the bind-users mailing list