BIND 10 #1782: update SQLite3Accessor iterator so it will return NSEC3
BIND 10 Development
do-not-reply at isc.org
Fri Apr 13 12:25:11 UTC 2012
#1782: update SQLite3Accessor iterator so it will return NSEC3
-------------------------------------+-------------------------------------
Reporter: | Owner: UnAssigned
jinmei | Status: reviewing
Type: task | Milestone:
Priority: low | Sprint-20120417
Component: data | Resolution:
source | Sensitive: 0
Keywords: | Sub-Project: DNS
Defect Severity: N/A | Estimated Difficulty: 4
Feature Depending on Ticket: NSEC3 | Total Hours: 0
Add Hours to Ticket: 0 |
Internal?: 0 |
-------------------------------------+-------------------------------------
Changes (by vorner):
* owner: vorner => UnAssigned
* status: accepted => reviewing
Comment:
It is ready for review. The changes are rather simple, I used SQL to
produce all the needed results, so the actual change is only to the
statement.
It, however, contains two small hacks:
* As the nsec3 table does not contain the sigtype column, I hardcode
"NSEC3"
as the value for the column. This should be OK, because we don't use it
and
there can be no other signature than for NSEC3 anyway. But it would be
nice
if we could actually get rid of the column, if it is not used.
* The results were ordered by rname, to group the RRs of the same RRset
together and many tests depended on this order. But the nsec3 table
does not
contain the rname column. So now I order it by rname of the records
table
and name of the nsec3 table, which keeps the order for the tests
relying on
it and does what is needed. But, didn't we talk about actually removing
the
ordering sometime?
Also, as this could strip the NSEC3 records on xfr-out, I suggests this is
marked as a bugfix in the changelog:
{{{
[bug] vorner
The NSEC3 records are now included when transferring a signed zone out.
}}}
--
Ticket URL: <http://bind10.isc.org/ticket/1782#comment:6>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list