[bind10-dev] Data source configuration

[JINMEI Tatuya / 神明達哉]

At Tue, 22 May 2012 11:40:38 +0200,
Michal 'vorner' Vaner <michal.vaner at nic.cz> wrote:

> I took the #1974 ticket which is to define data source configuration. I got some
> ideas and wrote them down into http://bind10.isc.org/wiki/datasrc_config. I'd
> like your input on which you think is the best one. I have a slight preference
> of the third one, but I'm OK with any of them.
> 
> It would be nice if everybody could read it and we could discuss it at the call,
> but I don't think it is realistic, as at least Jinmei is sleeping right now and
> will be just having his breakfast during the call. But if at least some of us
> you can read it, it might be worth a try to share some ideas.

I've not yet fully thought over it, but have some random, preliminary
comments:

- The requirement seems reasonable.
- In either approach, I'm not sure whether we can reasonably hardcode
  the matching strategy as part of configuration, because it could be
  application specific: auth allows partial match, but others
  generally don't.  Also, in our current API, "partial match" can only
  happen for findZone(); others only look for an exact match.
- Basically I'd like to avoid including specific application like
  "auth" in the configuration.
- I was not sure what "slaves" mean in the meta-data-source example.
  If it means "backend", or at least not related to DNS secondary, I
  think it's better to use a different term than "slave".
- I was also not sure about "zonefiles".  Does that mean an
  RFC1035-style master file for a single zone, or some textual file
  for a set of zones?

---
JINMEI, Tatuya
Internet Systems Consortium, Inc.