BIND 10 #1455: Add prerequisite checks to the DDNS module
BIND 10 Development
do-not-reply at isc.org
Wed May 23 09:56:15 UTC 2012
#1455: Add prerequisite checks to the DDNS module
-------------------------------------+-------------------------------------
Reporter: jelte | Owner: UnAssigned
Type: task | Status: reviewing
Priority: | Milestone:
medium | Sprint-20120529
Component: DDNS | Resolution:
Keywords: | Sensitive: 0
Defect Severity: N/A | Sub-Project: DNS
Feature Depending on Ticket: DDNS | Estimated Difficulty: 6
Add Hours to Ticket: 0 | Total Hours: 0
Internal?: 0 |
-------------------------------------+-------------------------------------
Changes (by jelte):
* owner: jelte => UnAssigned
* status: assigned => reviewing
Comment:
OK, that took a bit, but it's ready for review.
I implemented the algorithm as specified in RFC2136, except for the
'individual rr collection', since that is already handled by the message
parser. (I did include a test for it).
The branch is branched from trac1512, I rebased it after the first round
of review and before pushing, so I think that there will not be many
conflicts (if at all).
There are a few related changes in other parts of the code;
- libdns++ needed support for empty RRsets. To make sure it doesn't
interfere with rrsets that have not rdata, I copied the RFC2136 convention
of only allowing empty RRsets for classes ANY and NONE.
- Added a few namecomparison relation constants to pydnspp
- Fixed the python wrapper for find_zone() to always return a 3-tuple (it
sometimes dropped the third element and returned a 2-tuple)
As for the implementation itself
- there are a few methods that might be unnecessary; some prerequisites
are essentially the reverse of another. I still have separate methods for
those, since it looks nicer in the algorithm itself
- logging of failed prereqs is mostly like in bind9, but with slightly
different messages
- the zone, class, and datasrc_client are currently passed to the
check_prerequisites() method. I think they could just as well be set as
members of the Session class, but for now they are passed. (no strong
opinion here). But if the reviewer thinks so, we can probably remove them.
--
Ticket URL: <http://bind10.isc.org/ticket/1455#comment:10>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list