- Handling of fallback to AXFR in knot_ns_process_ixfrin().
- Handling of this condition in xfrin_process_ixfr_packet().
- Added new KNOT error code for this condition.
- Handling of this condition in xfr_process_event().

refs #1316 @2h

- Distinguishing completed transfer by the return value from
  xfrin_process_ixfr_packet().
- Handling case of only one SOA record in response (RFC1995bis 4
  a and b).

refs #1210 @1h

- Packet with TC bit set is regarded malformed (RFC1995bis 4 b).
- Proper cleanup in some cases.
- Returning XFRIN_RES_COMPLETE after complete transfer.

refs #1210 @1h

Conflicts:
	libknot/nameserver/name-server.c
	src/knot/other/debug.h

refs #1314

- The canonical domain name was not taken from the CNAME RRSet in
  zone, but from the one synthetized from the wildcard. This caused
  the node's owner to be different than the CNAME name.
- Added the case causing the cycling in to the example.com zone.

fixes #1313 @1h30m

Conflicts:
	src/knot/server/tcp-handler.c

When it fails to write PID file, it logs warning and continues to run.

fixes #1310

Executed debug code now updated:
* Renamed from DEBUG_KNOT_*(cmds) to debug_knot_*_exec(cmds) (available also in _verb and _detail)
* Default level is also 'brief', but verb and detail are available.

refs #1297

* Default level is 'brief' for all current debug messages.

refs #1297

refs #1297

* Default level is 'brief' for all current debug messages.

refs #1297

When not enabled, all debugging messages are disabled.

refs #1297

* 60-90s retry interval on AXFR bootstrap failure.
* Improved logging.

fixes #1306

fixes #1308

Requests for same zone are serialized.

To do:
* More intelligent scheduling of requests among threads.

fixes #982

refs #1289

refs #1289

refs #1289

refs #1289

- Only zones from old db having the exact same pointer as the zone
  in new db will be removed from the old db prior to destruction of
  the old db. Others will remain there and be destroyed also.

fixes #1257 @1h30m

- Created separate function for creating error response from
  partially assembled response packet. This function will take only
  the portion of the wireformat appropriate to the size of
  the query (just the Header and Question section).
- Added packet_question_size() to get the size of Header + Question

fixes #1300 @1h

Reverted back to blocking sockets, using NB sockets caused recv() returning too early in tcp_recv().

refs #1289

refs #1289

- Should fix problem with not reading whole packet.

Todo: implement recv() timeout.

refs #1289