BIND 10 #2464: Xfrin with TSIG broken: mistyped call

[BIND 10 Development]

#2464: Xfrin with TSIG broken: mistyped call
-------------------------------+-----------------------------------------
            Reporter:  jelte   |                        Owner:
                Type:  defect  |                       Status:  new
            Priority:  high    |                    Milestone:  New Tasks
           Component:  xfrin   |                     Keywords:
           Sensitive:  0       |              Defect Severity:  N/A
         Sub-Project:  DNS     |  Feature Depending on Ticket:
Estimated Difficulty:  0       |          Add Hours to Ticket:  0
         Total Hours:  0       |                    Internal?:  0
-------------------------------+-----------------------------------------
 I was updating my public server, and one zone wasn't coming in:

 {{{
 2012-11-05 22:43:40.305 ERROR [b10-xfrin.xfrin] XFRIN_XFR_OTHER_FAILURE
 AXFR transfer of zone tjeb.nl/IN failed: 'pydnspp.TSIGContext' object has
 no attribute 'last_has_signature'
 2012-11-05 22:43:40.316 INFO  [b10-xfrin.datasrc]
 DATASRC_DATABASE_UPDATER_ROLLBACK zone updates roll-backed for
 'tjeb.nl./IN' on sqlite3_zone.sqlite3
 2012-11-05 22:43:40.316 DEBUG [b10-xfrin.datasrc]
 DATASRC_DATABASE_UPDATER_DESTROYED zone updater destroyed for
 'tjeb.nl./IN' on sqlite3_zone.sqlite3
 }}}

 From looking at the TSIGContext code, that call should not be
 'last_has_signature' but 'last_had_signature'. Not sure if this is a
 regression due to tsig changes.

 This makes all transfers fail on my system, and probably on all systems
 that use TSIG.

 The fix is trivial (a one-letter change), but the unit tests should also
 be updated (a simple search-and-replace).

 If possible, a unit test that would have detected this should also be
 added.

 And we certainly need a lettuce test for this (in fact we should have
 already had this).

-- 
Ticket URL: <https://bind10.isc.org/ticket/2464>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development