BIND 10 #192: Data source hotspot cache
BIND 10 Development
do-not-reply at isc.org
Fri Jun 25 23:05:35 UTC 2010
#192: Data source hotspot cache
-------------------------+--------------------------------------------------
Reporter: each | Owner: jinmei
Type: enhancement | Status: reviewing
Priority: major | Milestone: 05. 3rd Incremental Release: Serious Secondary
Component: b10-auth | Resolution:
Keywords: | Sensitive: 0
-------------------------+--------------------------------------------------
Comment(by jinmei):
Replying to [comment:28 jinmei]:
> > I compromised by having ZoneInfo do it, but not DataSrcMatch.
(However, I had to disable one of the unit tests because of this. I still
think it makes more sense to have this live in DataSrcMatch.)
>
> Okay, I think I can live with that compromize. It's very counter-
intuitive that that made a test fail, though...I'll look into it.
Please check the attached diff (not committed to the branch).
The failed test itself is a simple regression. Now that the search key
(name) is adjusted in the caller side, type dependent consideration isn't
necessary.
This also means we don't have to (or even shouldn't) pass RRtype to the
!DataSrcMatch constructor. In fact, the type isn't used any other part of
this class. It was the case in your initial implementation, too, so we
could have avoided having rrtype as a class member in the first place.
This is another reason why I didn't think it makes sense to impose this
responsibility to the Name/DataSrcMatch class. The attached diff cleans
this up, too.
Furthermore, we now don't need a separate test for !DataSrcMatchByType
because it's no different from other lookups. My proposed diff didn't do
this cleanup yet.
On the other hand, we need more tests of other types for this class,
including the case where class doesn't match.
--
Ticket URL: <http://bind10.isc.org/ticket/192#comment:29>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list