[Dnsmasq-discuss] host-record and cname quirks/questions
sam at sltosis.org
Tue Apr 10 08:12:50 BST 2012
On Mon, Apr 09, 2012 at 05:31:31PM -0700, Rob Zwissler wrote:
> Hey Simon -
> I'm running 2.61test5 and working with the new host-record directive you
> added, as well as the cname directive, and I noticed some behavior that was
> somewhat counter-intuitive:
> 1) host-record=hostname,alias,alias,IP the alias's are returned as A
> records instead of CNAMEs, I'm guessing that was a conscious decision on
> your part, but it seems that for a more typical hostname with multiple
> CNAMEs, this is maybe not ideal, what do you think?
> 2) cname=test_cname,target_name if I do a "dig CNAME test_cname", it will
> forward the request to the upstream server, but if I do a "dig A
> test_name", then it will resolve it as a CNAME and answer with the CNAME &
> A as expected. Is that a bug or a feature?
> 3) on a perhaps related note, if I define a cname= without a corresponding
> /etc/hosts or host-record entry, it will never answer for the CNAME. So we
> cannot use cname to define aliases on hosts that we are not hosting an A
> record for? Is that a bug or a feature?
3) comes back often(may be time for a FAQ entry?), As far as I
understood it, it is not a bug nor a feature but more of a consequence
of how dnsmasq works.
You can find more recent answer to that question, but here is the first
coming up in my quick search:
> Dnsmasq-discuss mailing list
> Dnsmasq-discuss at lists.thekelleys.org.uk
More information about the Dnsmasq-discuss