]>
git.ipfire.org Git - thirdparty/kea.git/log
JINMEI Tatuya [Fri, 7 Oct 2011 04:08:13 +0000 (21:08 -0700)]
[1212] updated bind10-guide with information about the IXFR support.
JINMEI Tatuya [Thu, 6 Oct 2011 22:57:20 +0000 (15:57 -0700)]
[master] Merge branch 'master' of ssh://git.bind10.isc.org/var/bind10/git/bind10
JINMEI Tatuya [Thu, 6 Oct 2011 22:56:56 +0000 (15:56 -0700)]
[master] a quickhack fix to discheck failure in dhcp6/tests.
okayed on jabber
Jeremy C. Reed [Thu, 6 Oct 2011 22:25:35 +0000 (17:25 -0500)]
[master] fix build on FreeBSD by removing net/in.h include
On FreeBSD these needs sys/socket.h.
But we realized that the net/in.h is not needed now.
(TODO: remove #include <netinet/in.h> and #include <arpa/inet.h> also?)
Jeremy C. Reed [Thu, 6 Oct 2011 21:00:48 +0000 (16:00 -0500)]
[master] for trac #1247: distcheck should enable gtest when possible
If --with-gtest defined on ./configure line, then do same for distcheck
target.
JINMEI Tatuya [Thu, 6 Oct 2011 18:22:25 +0000 (11:22 -0700)]
[master] Merge branch trac1261 with fixing conflicts, etc.
- I need to add another mock method in xfrin/test to be compatible with the
latest version of Diff class.
- I also replaced 'remove' with 'delete' more completely in the Diff class
(and adjusted the caller side accordingly) for consistency.
JINMEI Tatuya [Thu, 6 Oct 2011 17:47:47 +0000 (10:47 -0700)]
[1261] added more explanation about commit timing
Tomek Mrugalski [Thu, 6 Oct 2011 16:14:52 +0000 (18:14 +0200)]
Disabled tests dhcp6.sockets and dhcp6.sendReceive
Both fail on BSD systems due to different interface name (lo0 instead of lo)
Will reenable them once 1186 is merged that contains fix for that.
Tomek Mrugalski [Thu, 6 Oct 2011 14:11:16 +0000 (16:11 +0200)]
Merge branch 'master' into trac878
Conflicts:
ChangeLog
JINMEI Tatuya [Thu, 6 Oct 2011 03:34:40 +0000 (20:34 -0700)]
[master] disabled factory_unittest if we use static link. this should solve
buildbot failure:
http://git.bind10.isc.org/~tester/builder/BIND10/
20111005214500 -MacOSX10.6-x86_64-Clang-Static/logs/unittests.out
okayed on jabber.
JINMEI Tatuya [Wed, 5 Oct 2011 23:53:58 +0000 (16:53 -0700)]
[1261] made clear XfrinState().handle_rr() shouldn't be called directly
via an exception. added a test for that.
JINMEI Tatuya [Wed, 5 Oct 2011 23:43:46 +0000 (16:43 -0700)]
[1261] cleanup: removed an unused variable
JINMEI Tatuya [Wed, 5 Oct 2011 23:10:04 +0000 (16:10 -0700)]
[1261] added more detailed documentation about SOA mismatch case
for non incremental update
JINMEI Tatuya [Wed, 5 Oct 2011 22:28:03 +0000 (15:28 -0700)]
[1261] a small wording fix as suggested in review
JINMEI Tatuya [Wed, 5 Oct 2011 21:39:27 +0000 (14:39 -0700)]
[master] use freeaddrinfo to free an addrinfo structure, not free.
reported by Francis on bind10-dev. looks okay, directly pushing.
JINMEI Tatuya [Wed, 5 Oct 2011 18:25:25 +0000 (18:25 +0000)]
[master] fixed a build regression on g++(3.x)/Solaris. The original code
doesn't seem to work in a templated test for this compiler.
okayed on jabber.
Jelte Jansen [Wed, 5 Oct 2011 13:32:28 +0000 (15:32 +0200)]
[1261] a few typos
Dima Volodin [Wed, 5 Oct 2011 11:31:31 +0000 (07:31 -0400)]
Merge branch 'master' of ssh://git.bind10.isc.org/var/bind10/git/bind10
Dima Volodin [Wed, 5 Oct 2011 11:30:39 +0000 (07:30 -0400)]
[1144] ChangeLog entry
JINMEI Tatuya [Wed, 5 Oct 2011 08:35:05 +0000 (01:35 -0700)]
[1261] a bit more documentation
JINMEI Tatuya [Wed, 5 Oct 2011 07:37:29 +0000 (00:37 -0700)]
[1261] overall comment update
JINMEI Tatuya [Wed, 5 Oct 2011 06:52:22 +0000 (23:52 -0700)]
[1261] editorial cleanup: no need for escaping '+' at EOL
JINMEI Tatuya [Wed, 5 Oct 2011 06:50:39 +0000 (23:50 -0700)]
[1261] log message update
JINMEI Tatuya [Wed, 5 Oct 2011 06:31:25 +0000 (23:31 -0700)]
[1261] corrected the type of zone_name arg for publish_xfrin_news (it's
now a Name object, not a string)
JINMEI Tatuya [Wed, 5 Oct 2011 06:17:41 +0000 (23:17 -0700)]
[1261] install NULL instead of "" for an empty column for backward
compatibility. We need this at least until we completely deprecate
the old API.
JINMEI Tatuya [Wed, 5 Oct 2011 05:26:35 +0000 (22:26 -0700)]
[1261] use IXFR or AXFR depending on the configuration
JINMEI Tatuya [Wed, 5 Oct 2011 04:17:10 +0000 (21:17 -0700)]
[1261] refactoring: pass request type (IXFR or AXFR) from the command handler
to do_xfrin().
JINMEI Tatuya [Wed, 5 Oct 2011 04:16:37 +0000 (21:16 -0700)]
[1261] more adjustment for distcheck
JINMEI Tatuya [Wed, 5 Oct 2011 04:15:53 +0000 (21:15 -0700)]
[1261] some more adjustment for distcheck
JINMEI Tatuya [Wed, 5 Oct 2011 03:54:02 +0000 (20:54 -0700)]
[1261] corrected env var for the build dir.
JINMEI Tatuya [Wed, 5 Oct 2011 03:51:50 +0000 (20:51 -0700)]
[1261] missing EXTRA_DIST, and corrected data dir for the writable DB.
JINMEI Tatuya [Wed, 5 Oct 2011 03:39:10 +0000 (20:39 -0700)]
[1261] one more missing update for the Diff class about remove->delete
JINMEI Tatuya [Wed, 5 Oct 2011 03:34:30 +0000 (20:34 -0700)]
[1261] added another sqlite3-based test. also make sure any remaining diff
object is cleared as soon as possible after the xfrin session.
JINMEI Tatuya [Wed, 5 Oct 2011 03:18:47 +0000 (20:18 -0700)]
[1261] added a test using a workable sqlite3 DB file and its data source
client to see it really works for an operational environment (and indeed
I found a bug in the Diff class). To make it possible, also updated
how to set the data source client for XfrinConnection.
JINMEI Tatuya [Wed, 5 Oct 2011 03:16:54 +0000 (20:16 -0700)]
[1261] fixed the interface mismatch: actual DataSource API expectes
'delete_rrset', not 'remove_rrset'. To make things consistent,
also removed the corresponding Diff methods from 'remove_data' to
'delete_data'.
JINMEI Tatuya [Tue, 4 Oct 2011 21:49:46 +0000 (14:49 -0700)]
[1261] comment cleanup: this TODO is mostly a non issue (our data source API
internally takes care of it)
JINMEI Tatuya [Tue, 4 Oct 2011 21:47:34 +0000 (14:47 -0700)]
[1261] added some error case tests. updated do_xfrin() so that it will
handle exceptions more correctly. in particular it should have caught
various other exceptions from other modules (e.g. one from pydnspp/isc.dns)
without incorrectly assuming it uses Boost.Python (we don't use it any more).
This change fixes another problem in an existing test, so I also fixed
that test case, too.
Michal 'vorner' Vaner [Tue, 4 Oct 2011 20:54:05 +0000 (22:54 +0200)]
Merge #1259
Michal 'vorner' Vaner [Tue, 4 Oct 2011 17:57:02 +0000 (19:57 +0200)]
[1259] Add a note about the library name
Michal 'vorner' Vaner [Tue, 4 Oct 2011 08:34:09 +0000 (10:34 +0200)]
[1259] More TTL tweaks
These are just tests and logging messages.
Dima Volodin [Tue, 4 Oct 2011 20:52:12 +0000 (16:52 -0400)]
[1144] Merge branch 'trac1144'
JINMEI Tatuya [Tue, 4 Oct 2011 20:51:23 +0000 (13:51 -0700)]
[1261] minor editorial cleanups
JINMEI Tatuya [Tue, 4 Oct 2011 17:30:31 +0000 (10:30 -0700)]
[1144] added one more test case: comparing incompatible types of RDATAs
Dima Volodin [Tue, 4 Oct 2011 11:33:55 +0000 (07:33 -0400)]
[1144] more comparison tests in tests/rdata_ds_like_unittest.cc
Dima Volodin [Tue, 4 Oct 2011 11:33:05 +0000 (07:33 -0400)]
[1144] read instead of peek in rdata/generic/detail/ds_like.h
Jelte Jansen [Tue, 4 Oct 2011 08:47:26 +0000 (08:47 +0000)]
[master] use c-style cast for functionptr cast
JINMEI Tatuya [Tue, 4 Oct 2011 07:46:55 +0000 (00:46 -0700)]
[1261] some documentation updates
JINMEI Tatuya [Tue, 4 Oct 2011 07:37:15 +0000 (00:37 -0700)]
[1261] updated do_xfrin() for the IXFR case
JINMEI Tatuya [Tue, 4 Oct 2011 06:22:24 +0000 (23:22 -0700)]
[1261] added finish_message() method for the xfrin state classes and
supported multiple-message IXFR session.
JINMEI Tatuya [Tue, 4 Oct 2011 06:00:52 +0000 (23:00 -0700)]
[1261] added multi-sequence case. this should already succeed, and it does indeed.
JINMEI Tatuya [Tue, 4 Oct 2011 05:34:21 +0000 (22:34 -0700)]
[1261] added some more tests for the simple IXFR case. also revised
the code so that the Diff obj is created in the RR processing on demand.
JINMEI Tatuya [Tue, 4 Oct 2011 02:23:38 +0000 (19:23 -0700)]
[1261] an initial step of IXFR RRs: copy _handle_axfrin_response() and
made minimal adjustment to support the IXFR state transitions. Refactored
and updated tests for a simple IXFR test and sharing some of the existing
frameworks.
JINMEI Tatuya [Tue, 4 Oct 2011 00:29:54 +0000 (17:29 -0700)]
[1261] comment update
JINMEI Tatuya [Tue, 4 Oct 2011 00:25:58 +0000 (17:25 -0700)]
[1261] extended _create_query() so that it can send an IXFR query.
JINMEI Tatuya [Mon, 3 Oct 2011 22:12:39 +0000 (15:12 -0700)]
[1261] a small cleanup: separated simpler cases of tests for _send_query()
to _create_query(). we don't have to examine the resulting wire-format
binary at this level of tests.
JINMEI Tatuya [Mon, 3 Oct 2011 19:19:36 +0000 (12:19 -0700)]
[1259] a minor wording fix: s/Bind9/BIND 9/
JINMEI Tatuya [Mon, 3 Oct 2011 18:06:10 +0000 (18:06 +0000)]
[master] (re)fixed the recent solaris build issue without relying on reinterpret_cast. okayed on jabber.
Jelte Jansen [Mon, 3 Oct 2011 16:36:19 +0000 (18:36 +0200)]
[1206] somehow i lost some indentation. put it back
Jelte Jansen [Mon, 3 Oct 2011 16:17:22 +0000 (18:17 +0200)]
[1206] new stuff makes some of the old (soon to be removed) fail cppcheck
Jelte Jansen [Mon, 3 Oct 2011 14:44:28 +0000 (14:44 +0000)]
[1206] fix segfault on sunstudio
Michal 'vorner' Vaner [Mon, 3 Oct 2011 13:35:33 +0000 (15:35 +0200)]
[1259] Missing EXTRA_DIST
Michal 'vorner' Vaner [Mon, 3 Oct 2011 13:13:51 +0000 (15:13 +0200)]
[1259] Better names
Michal 'vorner' Vaner [Mon, 3 Oct 2011 13:11:01 +0000 (15:11 +0200)]
[1259] Log when TTLs don't match
Michal 'vorner' Vaner [Mon, 3 Oct 2011 09:40:46 +0000 (11:40 +0200)]
[1259] Check the thing is not used after exception
As it would not be safe anyway.
Jelte Jansen [Mon, 3 Oct 2011 09:33:16 +0000 (11:33 +0200)]
Merge branch 'master' of ssh://git.bind10.isc.org/var/bind10/git/bind10
Jelte Jansen [Mon, 3 Oct 2011 09:29:36 +0000 (09:29 +0000)]
[master] fix for solaris build failure
simply a forgotten cast, reviewed on jabber
Michal 'vorner' Vaner [Mon, 3 Oct 2011 09:19:11 +0000 (11:19 +0200)]
[1259] Check the class
If it corresponds to the updater. This would be a bug, but we better
detect it sooner.
Michal 'vorner' Vaner [Mon, 3 Oct 2011 09:04:07 +0000 (11:04 +0200)]
[1259] Rename parameter to ds_client
Because datasource could represent something bigger behind the scenes.
Michal 'vorner' Vaner [Mon, 3 Oct 2011 09:02:27 +0000 (11:02 +0200)]
[1259] Use a constant for 100 updates
Jelte Jansen [Mon, 3 Oct 2011 08:14:13 +0000 (10:14 +0200)]
Merge branch 'master' of ssh://git.bind10.isc.org/var/bind10/git/bind10
JINMEI Tatuya [Mon, 3 Oct 2011 07:57:44 +0000 (00:57 -0700)]
[1261] a small cleanup: use a Name object for the zone name throughout the
XfrinConnection class and process_xfrin(), and only convert it to a string
when necessary.
JINMEI Tatuya [Mon, 3 Oct 2011 07:17:43 +0000 (00:17 -0700)]
[1261] a small refactoring: separate the current AXFR(+sqlite3) specific code
clearly. no behavior change in effect at this moment.
JINMEI Tatuya [Mon, 3 Oct 2011 06:52:55 +0000 (23:52 -0700)]
[1261] implemented IXFRAdd, IXFREnd and AXFR (currently just raise an
exception) classes.
JINMEI Tatuya [Mon, 3 Oct 2011 06:18:15 +0000 (23:18 -0700)]
[1261] implemented IXFRDelete class (state)
JINMEI Tatuya [Mon, 3 Oct 2011 06:01:17 +0000 (23:01 -0700)]
[1261] implemented IXFRDeleteSOA class (state)
Naoki Kambe [Mon, 3 Oct 2011 04:23:19 +0000 (13:23 +0900)]
[master] changelog entry for #928, #929, #930 and #1175
JINMEI Tatuya [Mon, 3 Oct 2011 04:00:56 +0000 (21:00 -0700)]
[1261] added XfrinFirstData class (state)
Naoki Kambe [Wed, 28 Sep 2011 04:53:21 +0000 (13:53 +0900)]
[1175] remove the retrying part because there is no clear necessity
Naoki Kambe [Wed, 28 Sep 2011 04:45:49 +0000 (13:45 +0900)]
[1175] comment the reason why socket.has_ipv6 is not used
Naoki Kambe [Mon, 26 Sep 2011 13:13:09 +0000 (22:13 +0900)]
[1175] correct the comment according to the purpose of the retry
Naoki Kambe [Mon, 26 Sep 2011 13:12:48 +0000 (22:12 +0900)]
[1175] correct the names and comments of the functions according to their
purposes
Naoki Kambe [Mon, 26 Sep 2011 09:35:22 +0000 (18:35 +0900)]
[1175] remove the two conditions because they do the same tests for IPv4
whether IPv6 is enabled or not
Naoki Kambe [Mon, 26 Sep 2011 09:04:13 +0000 (18:04 +0900)]
[1175] modify b10-stats_test.py, b10-stats-httpd_test.py and test_utils.py
- exclude the setup of handler in SIGALRM and add it as an external
function in the SignalHandler class in test_utils.py
- define the function in that class to reset the handler in
test_utils.py, and add it in tearDown() in each testcase
Naoki Kambe [Mon, 26 Sep 2011 06:46:30 +0000 (15:46 +0900)]
[1175] fix double hash signs
Naoki Kambe [Mon, 26 Sep 2011 13:24:35 +0000 (22:24 +0900)]
[1175] correct the comment of the function according to its purpose
Naoki Kambe [Tue, 13 Sep 2011 09:22:23 +0000 (18:22 +0900)]
[1175] correct the logging id
Naoki Kambe [Tue, 13 Sep 2011 09:21:55 +0000 (18:21 +0900)]
[1175] add the package name for SessionError
Naoki Kambe [Tue, 13 Sep 2011 08:53:22 +0000 (17:53 +0900)]
[1175] modify test_utils.py
- explicitly shut down the socket of the msgq before shutting down
the msgq
- do nothing in the shutdown method of MockMsgq for avoiding shutting
down the msgq twice
- replace the stop method in the shutdown method of the MyStatsHttpd
with the shutdown command of stats_httpd
Naoki Kambe [Thu, 8 Sep 2011 01:52:19 +0000 (10:52 +0900)]
[1175] modify test_utils.py
- move up an assignment of the BIND10_MSGQ_SOCKET_FILE environment
variable (BaseModules uses a constant file name during each
testcase)
- BaseModules checks whether msgq is ready after it started the msgq
object. A SessionTimeout is raised here if not.
Naoki Kambe [Tue, 6 Sep 2011 11:27:44 +0000 (20:27 +0900)]
[1175] fix a typo
Naoki Kambe [Tue, 6 Sep 2011 11:17:57 +0000 (20:17 +0900)]
[1175] modify b10-stats-httpd_test.py
- The function get_availaddr uses socket.getaddrinfo for getting the
address family.
- The function is_ipv6_enabled uses 3 random ports for checking
whether IPv6 is enabled.
Naoki Kambe [Tue, 6 Sep 2011 02:42:42 +0000 (11:42 +0900)]
[1175] send the command 'status' to the stats when it started, and
then send the command 'shutdown', and also check each value returned
by each invoked command
Naoki Kambe [Mon, 5 Sep 2011 08:14:11 +0000 (17:14 +0900)]
[1175] modify stats_httpd.py.in, b10-stats-httpd_test.py and
b10-stats_test.py
- The stats httpd doesn't need to return an argument when it's
shutting down.
- The testcase sends the 'status' command or the 'shutdown' command
to the stats or the stats httpd when they started, and then their
returned values are checked.
Naoki Kambe [Mon, 5 Sep 2011 05:01:18 +0000 (05:01 +0000)]
[1175] modify test_utils.py
- add 3-time retry to creating the server object when it fails in the
ThreadingServerManager class
- suppress outputs by Msgq, and add dummy sys module and the output
methods
- pass Exceptions raised while it's running with a thread
Naoki Kambe [Mon, 5 Sep 2011 04:53:24 +0000 (04:53 +0000)]
[1175] remove a logging name from unittest
Naoki Kambe [Mon, 5 Sep 2011 04:53:11 +0000 (04:53 +0000)]
[1175] modify b10-stats-httpd_test.py
- remove a logging name from unittest
- do stats_httpd.stop() in teadDown() instead of each test case
- send 'shutdown' command to kill stats_httpd when testing address
already in use
Naoki Kambe [Mon, 5 Sep 2011 04:46:08 +0000 (04:46 +0000)]
[1175] remove -v option from pycoverage
Naoki Kambe [Thu, 1 Sep 2011 11:34:28 +0000 (11:34 +0000)]
[1175] add -v option in pycoverage for debugging the failure in
buildbot. (unittest.main() with verbosity option is not supported
in Python3.1.)
Naoki Kambe [Thu, 1 Sep 2011 04:19:53 +0000 (04:19 +0000)]
[1175] modify b10-stats-httpd_test.py, b10-stats_test.py and
test_utils.py
- more strictly close the io object whether it's successfully opened
or not
- add verbosity=2 in unittest.main for debugging the failure in the
buildbot
- don't redict sys.stderr in MockMsgq
- rename the function name to create_specfile
- switch the verbose in Msgq into True
Naoki Kambe [Fri, 26 Aug 2011 03:01:47 +0000 (12:01 +0900)]
[1175] fix wrong list-type handling in the function
get_spec_defaults and add more tests into test_get_spec_defaults