]>
git.ipfire.org Git - thirdparty/kea.git/log
Jeremy C. Reed [Tue, 5 Feb 2013 00:03:27 +0000 (18:03 -0600)]
[master] remove __pycache__ directory with distclean
(reported by emhn: left over after a distclean)
Some of the makefiles clear that but not that one. We do
automate the distclean (via distcheck) as part of our
buildfarm. But I think this __pycache__ was introduced in python3.2 and
those build systems don't have it.
This is trivial so didn't get review.
Jeremy C. Reed [Tue, 5 Feb 2013 00:01:30 +0000 (18:01 -0600)]
[master] don't include run.sh in distfile
The run.sh.in is in the distfile.
(issue reported by emhn: was removed with distclean but was in tarball)
Trivial so didn't get review.
JINMEI Tatuya [Mon, 4 Feb 2013 22:23:47 +0000 (14:23 -0800)]
[1866] added note about not-implemented RR types dict in rdata/template.h.
Jelte Jansen [Mon, 4 Feb 2013 22:18:42 +0000 (23:18 +0100)]
[master] Fix cmdctl pem and csv file installation
They were installed twice (to /etc and to /share, removed the latter)
Also, the generated .pem files were included in the tarball, while they should be generated during the build process
JINMEI Tatuya [Mon, 4 Feb 2013 22:11:34 +0000 (14:11 -0800)]
[1866] revise generate_typeclasscode without using temporary text variables.
wouldn't matter for the usage of this script, but this should be slightly
more efficient.
JINMEI Tatuya [Mon, 4 Feb 2013 21:23:49 +0000 (13:23 -0800)]
[1866] removed more now-removed files from Makefile.am.
JINMEI Tatuya [Mon, 4 Feb 2013 19:50:05 +0000 (11:50 -0800)]
[1866] remove dummy text files from makefile
JINMEI Tatuya [Mon, 4 Feb 2013 19:49:36 +0000 (11:49 -0800)]
[1866] updated the gen script so it generates types/classes correctly.
JINMEI Tatuya [Mon, 4 Feb 2013 19:15:05 +0000 (11:15 -0800)]
[1866] list all meta/not-yet-supported RR types and classes in the gen script
that's probably simpler than using dummy files and directories.
JINMEI Tatuya [Mon, 4 Feb 2013 19:02:14 +0000 (11:02 -0800)]
[1866] removed unnecessary file
JINMEI Tatuya [Mon, 4 Feb 2013 18:49:09 +0000 (10:49 -0800)]
[1866] corrected some regexp's: make sure it matches '.' before file suffixes.
JINMEI Tatuya [Mon, 4 Feb 2013 17:25:14 +0000 (09:25 -0800)]
[2617] added comments about restrictions on run_kqueue/poller.
Jelte Jansen [Mon, 4 Feb 2013 17:13:54 +0000 (18:13 +0100)]
[1901] Rename source file for b10-init to init.py
Michal 'vorner' Vaner [Mon, 4 Feb 2013 16:14:23 +0000 (17:14 +0100)]
[1924] Use the constants in default parameters too
Michal 'vorner' Vaner [Mon, 4 Feb 2013 16:06:22 +0000 (17:06 +0100)]
[1924] Use the constants in the process_command_send
Michal 'vorner' Vaner [Mon, 4 Feb 2013 16:52:06 +0000 (17:52 +0100)]
[1924] Change the order of dirs
So the generated file is really generated before it is used.
Michal 'vorner' Vaner [Mon, 4 Feb 2013 16:48:55 +0000 (17:48 +0100)]
[1924] Use the constants in python group_sendmsg
Michal 'vorner' Vaner [Mon, 4 Feb 2013 16:43:57 +0000 (17:43 +0100)]
[1924] (minor) Fix copyright year
Michal 'vorner' Vaner [Mon, 4 Feb 2013 16:42:58 +0000 (17:42 +0100)]
[1924] Generate the common_defs.h header too
Michal 'vorner' Vaner [Mon, 4 Feb 2013 16:21:48 +0000 (17:21 +0100)]
[1924] Use constants when building CC message in C++
Use the newly introduced constants in the group_sendmsg method. Other
methods still need to be ported, but leaving that to other ticket.
Michal 'vorner' Vaner [Mon, 4 Feb 2013 15:59:36 +0000 (16:59 +0100)]
[1924] Generate the python version of constant defs
Generate it from the .cc file, so the are the same.
Michal 'vorner' Vaner [Mon, 4 Feb 2013 15:10:28 +0000 (16:10 +0100)]
[1924] Create a file for common protocol constants
This should be included and used from whatever libraries and programs
that talk to other bind10 programs. It allows defining constants shared
across the protocol.
Jelte Jansen [Mon, 4 Feb 2013 14:41:42 +0000 (15:41 +0100)]
[1901] Update version of lettuce test configs
Jelte Jansen [Mon, 4 Feb 2013 14:12:10 +0000 (15:12 +0100)]
[1901] Update manpages, log on certain update warnings
Second part of review as found in http://bind10.isc.org/ticket/1901#comment:11
Jelte Jansen [Mon, 4 Feb 2013 11:53:25 +0000 (12:53 +0100)]
[1901] Fix run script, add manpage, revert ChangeLog
According to the first part of the review in http://bind10.isc.org/ticket/1901#comment:11
Yoshitaka Aharen [Mon, 4 Feb 2013 03:29:32 +0000 (12:29 +0900)]
[2157] fix for
1c470ce was incomplete
Yoshitaka Aharen [Mon, 4 Feb 2013 02:08:57 +0000 (11:08 +0900)]
[2157] add a note for TSIG and opcode to b10-auth man page
Yoshitaka Aharen [Mon, 4 Feb 2013 02:04:13 +0000 (11:04 +0900)]
[2157] address occasional lettuce failure: add reply check for getstats command
JINMEI Tatuya [Sat, 2 Feb 2013 04:27:58 +0000 (20:27 -0800)]
[1901] fixed indentation
JINMEI Tatuya [Sat, 2 Feb 2013 03:47:12 +0000 (19:47 -0800)]
[2157] editorial: removed space in a blank line.
JINMEI Tatuya [Sat, 2 Feb 2013 03:43:55 +0000 (19:43 -0800)]
[2157] editorial: unify two short lines
JINMEI Tatuya [Fri, 1 Feb 2013 22:43:54 +0000 (14:43 -0800)]
[1725] disable sockcreator udp6_reate test on solaris 2.10 as OS is broken.
Yoshitaka Aharen [Sat, 2 Feb 2013 02:52:23 +0000 (11:52 +0900)]
[2157] fix opcode handling and added some documentation
JINMEI Tatuya [Sat, 2 Feb 2013 02:41:19 +0000 (18:41 -0800)]
[1924] constify
Yoshitaka Aharen [Sat, 2 Feb 2013 02:03:18 +0000 (11:03 +0900)]
[2157] make opcode a reference
Yoshitaka Aharen [Sat, 2 Feb 2013 01:55:56 +0000 (10:55 +0900)]
[2157] correct method naming of MessageAttributes
JINMEI Tatuya [Fri, 1 Feb 2013 23:37:00 +0000 (15:37 -0800)]
[1924] added blnak lines between method declarations for readability.
JINMEI Tatuya [Fri, 1 Feb 2013 23:20:27 +0000 (15:20 -0800)]
[master] use temp vars for SunStudio incapable of type matching with template.
committing at my discretion.
JINMEI Tatuya [Fri, 1 Feb 2013 16:12:32 +0000 (08:12 -0800)]
[2617] more editorial fix; remove unnecessary quote, use double-quote for """
Jelte Jansen [Fri, 1 Feb 2013 13:40:28 +0000 (14:40 +0100)]
[1901] rely on set path for calling b10-init
Jelte Jansen [Fri, 1 Feb 2013 13:33:50 +0000 (14:33 +0100)]
[1901] Remove b10-init from sbin (was already in libexec)
Mukund Sivaraman [Fri, 1 Feb 2013 13:32:32 +0000 (19:02 +0530)]
[master] Avoid static init fiasco in libdns++ unittests due to merge of #2390, #2656
Jelte Jansen [Fri, 1 Feb 2013 13:31:00 +0000 (14:31 +0100)]
[1901] small style fixes
Jelte Jansen [Fri, 1 Feb 2013 11:35:42 +0000 (12:35 +0100)]
[1901] Auto-update to new config (Init instead of Boss)
Mukund Sivaraman [Fri, 1 Feb 2013 10:39:31 +0000 (16:09 +0530)]
[master] Update ChangeLog for #2390 and #2656
Mukund Sivaraman [Fri, 1 Feb 2013 10:38:01 +0000 (16:08 +0530)]
Merge branch 'master' into trac2656_2
Mukund Sivaraman [Fri, 1 Feb 2013 10:29:55 +0000 (15:59 +0530)]
Merge branch 'master' into trac2390_2
Jelte Jansen [Fri, 1 Feb 2013 10:10:49 +0000 (11:10 +0100)]
[1901] Some more Bob replacements
Jelte Jansen [Fri, 1 Feb 2013 09:47:04 +0000 (10:47 +0100)]
[1901] Rename bob and BoB as well
JINMEI Tatuya [Fri, 1 Feb 2013 04:19:54 +0000 (20:19 -0800)]
[2659] updated the behavior for empty name derived from opt-out nsec3.
it now returns the closest provable encloser proof as described in
RFC 5155 errata 3441, instead of returning SERVFAIL. test case was
adjusted and test was renamed to explain the intent more correctly.
JINMEI Tatuya [Fri, 1 Feb 2013 03:48:48 +0000 (19:48 -0800)]
[2659] unrelated editorial fix: update the code comment to match impl.
the mock zone finder is now actually smart enough to support empty non
terminals.
JINMEI Tatuya [Fri, 1 Feb 2013 03:47:48 +0000 (19:47 -0800)]
[2659] update nxrrsetMissingNSEC3 test using actual test RRs, not dynamic ones.
this made the test work for other actual (non-mock) data sources, so it's
enabled for them too. no behavior change yet.
JINMEI Tatuya [Fri, 1 Feb 2013 03:46:56 +0000 (19:46 -0800)]
[2659] unrelated editorial fix: updated the comment to match the actual data.
JINMEI Tatuya [Fri, 1 Feb 2013 03:31:43 +0000 (03:31 +0000)]
[master] qualify a namespace within InMemoryZoneFinder more strictly.
without this SunStudio will be confused about the namespace for the
declaration of internal Context below and refuse to compile.
confirmed it works, okayed on jabber.
JINMEI Tatuya [Thu, 31 Jan 2013 22:26:06 +0000 (14:26 -0800)]
[master] changelog for #2667
JINMEI Tatuya [Thu, 31 Jan 2013 22:18:29 +0000 (14:18 -0800)]
[2667] minor doc update: mentioning --disable-rpath and not mentioning -rpath
(-rpath isn't actually checked in this macro)
JINMEI Tatuya [Thu, 31 Jan 2013 22:04:05 +0000 (14:04 -0800)]
[2667] editorial fix: folded some long lines
Adam Tkac [Thu, 10 Jan 2013 17:04:04 +0000 (18:04 +0100)]
[2667] Add --disable-rpath configure switch.
Some dynamic linkers are wise enough (at least the Linux one) to correctly pick
needed shared libraries without hardcoded rpath. The --disable-rpath configure
switch allows distributors to avoid hardcoded rpath paths in libraries and
executables.
Signed-off-by: Adam Tkac <atkac@redhat.com>
JINMEI Tatuya [Thu, 31 Jan 2013 20:15:47 +0000 (12:15 -0800)]
[2157] removed white spaces at EOL
JINMEI Tatuya [Thu, 31 Jan 2013 19:26:12 +0000 (11:26 -0800)]
[2617] another editorial fix, in a comment.
JINMEI Tatuya [Thu, 31 Jan 2013 19:24:02 +0000 (11:24 -0800)]
[2617] some more editorial fixes in pydoc
JINMEI Tatuya [Thu, 31 Jan 2013 19:00:09 +0000 (11:00 -0800)]
[2617] editorial cleanups for MsgQCloseOnReceive pydoc
JINMEI Tatuya [Thu, 31 Jan 2013 18:49:57 +0000 (10:49 -0800)]
[2617] treat ECONNRESET on recv as error unconditionally with detailed notes
Curtis Blackburn [Thu, 31 Jan 2013 18:26:57 +0000 (12:26 -0600)]
[trac2665] fixed two unchecked return codes in src/lib/config/tests/ccsession_unittests.cc
Jelte Jansen [Thu, 31 Jan 2013 16:28:29 +0000 (17:28 +0100)]
[1901] Some more changes
Should be about working with the new names now
Stephen Morris [Thu, 31 Jan 2013 16:03:37 +0000 (16:03 +0000)]
[2673] Further changes to fix build problems under g++
The previous commit disabled unused parameter warnings in the DHCP
code when compiling with clang. g++ picked up some additional
warnings/errors, and this fix corrects those.
Jelte Jansen [Thu, 31 Jan 2013 15:42:31 +0000 (16:42 +0100)]
[1901] Replace Boss with Init
Jelte Jansen [Thu, 31 Jan 2013 15:36:25 +0000 (16:36 +0100)]
[1901] Replace 'boss' with 'b10-init' or 'b10_init'
Jelte Jansen [Thu, 31 Jan 2013 14:47:39 +0000 (15:47 +0100)]
[1901] Rename 'bind10' script files
- renamed 'bind10' to 'b10-init'
- renamed related files as well
- left alone the 'run' script, as well as the directory name
- added new 'bind10' script that execs b10-init
note this commit only handles file names, and internal references have not been updated yet
Michal 'vorner' Vaner [Thu, 31 Jan 2013 14:09:25 +0000 (15:09 +0100)]
[1924] Update mocks with the new parameter
Michal 'vorner' Vaner [Thu, 31 Jan 2013 14:05:23 +0000 (15:05 +0100)]
[1924] Send the want_answer header from C++
Michal 'vorner' Vaner [Thu, 31 Jan 2013 14:01:57 +0000 (15:01 +0100)]
[1924] Test for the C++ want_answer parameter
Yoshitaka Aharen [Thu, 31 Jan 2013 13:31:49 +0000 (22:31 +0900)]
[2157] increment opcode counter if message header parse succeeded
Michal 'vorner' Vaner [Thu, 31 Jan 2013 13:15:20 +0000 (14:15 +0100)]
[1924] Unittest for the original C++ group_sendmsg
Stephen Morris [Thu, 31 Jan 2013 12:16:22 +0000 (12:16 +0000)]
[2673] Modifications to avoid clang warnings and errors in DHCP code
Michal 'vorner' Vaner [Thu, 31 Jan 2013 11:49:30 +0000 (12:49 +0100)]
[1924] Method to read messages sent over wire in tests
Not used yet, the test needs to be yet written.
JINMEI Tatuya [Thu, 31 Jan 2013 06:53:54 +0000 (22:53 -0800)]
[2617] added "exiting" log message to those that don't have it until now.
JINMEI Tatuya [Thu, 31 Jan 2013 06:19:40 +0000 (22:19 -0800)]
[2617] logged shutdown event of cfgmgr
JINMEI Tatuya [Thu, 31 Jan 2013 06:16:58 +0000 (22:16 -0800)]
[2617] reorder log message; some more comments and doc
JINMEI Tatuya [Thu, 31 Jan 2013 05:24:37 +0000 (21:24 -0800)]
[2617] separated real recv error and close-on-recv; treat the latter as debug.
JINMEI Tatuya [Thu, 31 Jan 2013 04:27:14 +0000 (20:27 -0800)]
[2617] make sure FD handling is performed only when it's still valid.
This eliminates the possibility of hitting MSGQ_READ_UNKNOWN_FD, and, in fact,
we can even deprecate this message. And then, process_socket() becomes
a 2-line function called only by _process_fd() and can be integrated into
the latter. so it's done, too.
JINMEI Tatuya [Thu, 31 Jan 2013 03:56:27 +0000 (19:56 -0800)]
[2617] internal refactoring: extract FD handling into a separate method.
JINMEI Tatuya [Thu, 31 Jan 2013 02:01:48 +0000 (18:01 -0800)]
[2617] log EPIPE at the warn level. added more detailed tests for send_data.
Mukund Sivaraman [Thu, 31 Jan 2013 00:38:59 +0000 (06:08 +0530)]
[2656] Check that origin argument is used for relative names
Mukund Sivaraman [Thu, 31 Jan 2013 00:33:04 +0000 (06:03 +0530)]
[2656] Include bad data in the exception message
Mukund Sivaraman [Thu, 31 Jan 2013 00:22:55 +0000 (05:52 +0530)]
[2390] Check that origin argument is used for relative names
Mukund Sivaraman [Thu, 31 Jan 2013 00:16:15 +0000 (05:46 +0530)]
[2390] Include bad data in the exception message
JINMEI Tatuya [Wed, 30 Jan 2013 21:46:17 +0000 (13:46 -0800)]
[2617] make sure error info isn't lost on socket.error in getbytes()
the previous code (probably unintentionally) converted it to an empty
socket.error, resulting in useless debug message.
JINMEI Tatuya [Wed, 30 Jan 2013 21:45:04 +0000 (13:45 -0800)]
[2617] added a debug message on session establishment with lname. debug aid.
JINMEI Tatuya [Wed, 30 Jan 2013 21:31:10 +0000 (13:31 -0800)]
[2617] log lname after establishing cc session; for debug aid
JINMEI Tatuya [Wed, 30 Jan 2013 21:29:20 +0000 (13:29 -0800)]
[2617] extend python cc module so it logs some events, and log local lname
this is for debugging the issue of this task, but will be useful in general.
Tomek Mrugalski [Wed, 30 Jan 2013 18:07:54 +0000 (19:07 +0100)]
[2673] clangs-specific flags removed, fix in Dhcpv{4,6}srv::writeServerID
Jeremy C. Reed [Wed, 30 Jan 2013 15:35:48 +0000 (09:35 -0600)]
[master] DHCP4_SERVERID_LOADED and DHCP6_SERVERID_LOADED need server-id too
On my home system I received error when running unit tests like:
Message DHCP4_SERVERID_LOADED server-id /home/reed/opt/bind10/var/b10-dhcp4-serverid has been loaded from file %2
assertion ""Excess logger placeholders still exist in message" == NULL" failed: file "../../../../src/lib/log/log_formatter.cc", line 69, function "void isc::log::checkExcessPlaceholders(std::string*, unsigned int)"
I am using --enable-logger-checks. The build farm also uses that.
I am not sure why this was not seen on the build farm.
I added the missing value for dhcp4 and dhcp6 and the make check completed.
It had slight review on jabber.
Jelte Jansen [Wed, 30 Jan 2013 15:10:47 +0000 (16:10 +0100)]
[2390] small style fix
Michal 'vorner' Vaner [Wed, 30 Jan 2013 13:46:43 +0000 (14:46 +0100)]
[1924] Add python group_sent want_reply parameter
So the reporting of errors can be turned on from python.
Michal 'vorner' Vaner [Wed, 30 Jan 2013 13:23:58 +0000 (14:23 +0100)]
[1924] (minor) Style cleanups
* Spaces
* Remove commented-out code
* Use full relative path to header
Michal 'vorner' Vaner [Wed, 30 Jan 2013 13:22:21 +0000 (14:22 +0100)]
[1924] Provide proper headers
It wouldn't be needed in practice, but provide headers as close to if it
was sent by another client as reasonable and possible.
Also test we can explicitly say we don't want the answer, instead of
having just the default.
Shane Kerr [Wed, 30 Jan 2013 10:47:57 +0000 (11:47 +0100)]
Typo fixes.
Shane Kerr [Wed, 30 Jan 2013 10:27:16 +0000 (11:27 +0100)]
Change lettuce test timeouts from 10 seconds to 1 minute.
Michal 'vorner' Vaner [Wed, 30 Jan 2013 10:17:27 +0000 (11:17 +0100)]
[1924] Send the errors about missing recipient
If the msgq finds no recipient to send to and the sender wants an
answer, provide an error as the answer. Also check the error messages
look sane in tests.