BIND 10 #495: Hook up NSAS to iterator/cache

[BIND 10 Development]

#495: Hook up NSAS to iterator/cache
-------------------------------------+-------------------------------------
                 Reporter:  shane    |                Owner:  jelte
                     Type:           |               Status:  reviewing
  enhancement                        |            Milestone:  R-Team-
                 Priority:  major    |  Sprint-20110316
                Component:           |           Resolution:
  resolver                           |            Sensitive:  0
                 Keywords:           |  Add Hours to Ticket:  0
Estimated Number of Hours:  3.0      |          Total Hours:  0
                Billable?:  1        |
                Internal?:  0        |
-------------------------------------+-------------------------------------
Changes (by stephen):

 * owner:  stephen => jelte


Comment:

 >> operator(): when calculating the elapsed time, do you need a test
 before doing the calculation?
 >> :
 > I think so, as we don't want the difference to be negative or 0, which
 are results that are possible with gettimeofday. I think there are some
 clock-tick-dependent time functions that wouldn't have this property, but
 I don't think any of those are as portable.
 As both cur_time and current_ns_qsent_time are obtained via calls to
 gettimeofday() (which should provide a monotonically increasing time), we
 should not need to worry about a negative difference.  It could be a zero,
 in which case a check after the calculation (the statement "rtt =
 std::max(1, rtt)"?) is required.

 > but i was wondering this about our readUint16() method in buffer
 actually
 I've added a line to ResolverTestingBrainstorming about checking TCP
 communication between machines of different "endian-ness" to address this.

 All OK, please merge.

-- 
Ticket URL: <https://bind10.isc.org/ticket/495#comment:6>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development