Wilco Baan Hofman [Mon, 30 Aug 2010 13:27:05 +0000 (15:27 +0200)]
5 uint32_t's should be set, not 5 bytes.
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Wilco Baan Hofman [Mon, 30 Aug 2010 13:05:25 +0000 (15:05 +0200)]
The root key should be created as a root key, not subkey.
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Wilco Baan Hofman [Mon, 30 Aug 2010 10:21:29 +0000 (12:21 +0200)]
Do not include the closing doublequote in .reg files.
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Wilco Baan Hofman [Wed, 28 Jul 2010 12:46:38 +0000 (14:46 +0200)]
Fix hex():00,00,00 bugs and strtol dword bugs, these values are explicitly hex.
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Wilco Baan Hofman [Tue, 27 Jul 2010 15:21:48 +0000 (17:21 +0200)]
Make sure we write the value stored in value, not line (doublequoted).
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Wilco Baan Hofman [Tue, 27 Jul 2010 09:13:24 +0000 (11:13 +0200)]
Make sure that regf only writes to disk every five seconds or when flushed or destroyed
Brings speed up on a large .reg import from 17.5s to 0.95s, over 18 times faster
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Wilco Baan Hofman [Mon, 26 Jul 2010 20:10:43 +0000 (22:10 +0200)]
Fix .reg file format parsing.
* multiline data
* doublequoted value name
* handle windows format CRLF
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Wilco Baan Hofman [Mon, 26 Jul 2010 18:16:44 +0000 (20:16 +0200)]
Fix registry parsing of strings to also support the windows method.
Specifically: dword:
00000000, hex:00,00,00,00, hex(0..B):00,00,00,00
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Volker Lendecke [Sun, 3 Oct 2010 11:21:21 +0000 (13:21 +0200)]
s3: I thought I had compiled this....
Matthias Dieter Wallnöfer [Sat, 2 Oct 2010 11:45:22 +0000 (13:45 +0200)]
s4:dsdb_dn_val_rmd_flags - memmem - scan the whole string for occourences
Do this as in "dsdb_dn_is_upgraded_link_val". There is really no reason to
truncate before search.
Autobuild-User: Matthias Dieter Wallnöfer <mdw@samba.org>
Autobuild-Date: Sun Oct 3 10:45:39 UTC 2010 on sn-devel-104
Matthias Dieter Wallnöfer [Wed, 29 Sep 2010 17:50:20 +0000 (19:50 +0200)]
s4:ldap.py - delete the right object after test completition
Matthias Dieter Wallnöfer [Wed, 29 Sep 2010 17:39:17 +0000 (19:39 +0200)]
s4:ldap.py - fix "system only" test
A part was missing
Matthias Dieter Wallnöfer [Mon, 27 Sep 2010 05:46:11 +0000 (07:46 +0200)]
s4:acl_read LDB module - fix counter type
Matthias Dieter Wallnöfer [Thu, 30 Sep 2010 06:47:24 +0000 (08:47 +0200)]
s4:getncchanges.c - fix some counter types
Matthias Dieter Wallnöfer [Mon, 27 Sep 2010 05:42:29 +0000 (07:42 +0200)]
s4:schema_samba4.ldif - update allocated controls list
This needs always to be done after a control allocation otherwise we end up in
double-allocations and unexpected behaviour.
Matthias Dieter Wallnöfer [Tue, 28 Sep 2010 10:33:40 +0000 (12:33 +0200)]
s4:libcli/resolve/dns_ex.c - fix "const" warning
Volker Lendecke [Sun, 3 Oct 2010 09:56:35 +0000 (11:56 +0200)]
s3: Move #define VALGRIND to config.h
This fixes the valgrind overrun in the tdb jenkins hash
Stefan Metzmacher [Sat, 2 Oct 2010 00:13:36 +0000 (02:13 +0200)]
s4:rpc_server/echo: fix compiler warning
metze
Autobuild-User: Stefan Metzmacher <metze@samba.org>
Autobuild-Date: Sun Oct 3 09:41:51 UTC 2010 on sn-devel-104
Volker Lendecke [Sun, 3 Oct 2010 08:54:06 +0000 (10:54 +0200)]
s3: Attempt to fix the IRIX build
Volker Lendecke [Sun, 3 Oct 2010 08:46:22 +0000 (10:46 +0200)]
s3: Attempt to fix a ton of warnings on the build farm
Volker Lendecke [Sat, 2 Oct 2010 21:49:06 +0000 (23:49 +0200)]
s3: Remove talloc_autofree_context() from dmapi.c
This is a place where an explicit dmapi_destroy_session would be needed. But we
don't use a destructor for this.
Volker Lendecke [Sat, 2 Oct 2010 21:44:58 +0000 (23:44 +0200)]
s3: Remove talloc_autofree_context() from unexpected.c
Same argument as in
997a64f6fb11e19d78c: If tdb_close() was needed at exit
time, we'd have more severe problems by now.
Volker Lendecke [Sat, 2 Oct 2010 21:40:48 +0000 (23:40 +0200)]
s3: Remove talloc_autofree_context() from srv_samr_nt.c
The dispinfo structs need to survive within a process across pipe opens,
but they don't have a special destructor
Volker Lendecke [Sat, 2 Oct 2010 21:38:33 +0000 (23:38 +0200)]
s3: Remove talloc_autofree_context from pdb_interface
None of the pdb backends have special destructors that need to be run at
program exit.
Andrew Tridgell [Sat, 2 Oct 2010 21:56:28 +0000 (14:56 -0700)]
s4-kcc: silence "Testing kcctpl_create_intersite_connections" message
Autobuild-User: Andrew Tridgell <tridge@samba.org>
Autobuild-Date: Sun Oct 3 04:51:44 UTC 2010 on sn-devel-104
Andrew Tridgell [Sat, 2 Oct 2010 21:23:43 +0000 (14:23 -0700)]
s4-selftest: support 'make testenv SCREEN=1'
this can be used to start a test envioronment in screen windows.
Use:
make testenv SCREEN=1 SELFTEST_TESTENV=dc
to launch just one environment (in this case, "dc")
Andrew Tridgell [Sat, 2 Oct 2010 21:21:52 +0000 (14:21 -0700)]
s4-selftest: silence warnings about bind chown
Andrew Tridgell [Sat, 2 Oct 2010 21:21:11 +0000 (14:21 -0700)]
s4-test: silence a tap2subunit error
Andrew Tridgell [Sat, 2 Oct 2010 20:08:37 +0000 (13:08 -0700)]
s4-test: silence the Failed to chown message in make test
Andrew Tridgell [Sat, 2 Oct 2010 20:03:40 +0000 (13:03 -0700)]
s4-ldif: get rid of the ndr_pull_error message on startup
detect if the prefixmap blob is text or binary using the first 4 bytes
Andrew Tridgell [Sat, 2 Oct 2010 19:44:28 +0000 (12:44 -0700)]
s4-server: exit with status 127 on SIGTERM
this ensures that make test detects a failure if samba is killed by
SIGTERM
Andrew Tridgell [Sat, 2 Oct 2010 19:43:50 +0000 (12:43 -0700)]
s4-selftest: added --screen option for test
you can now do:
make test TESTS="some test" SCREEN=1
while in GNU screen, and all the samba servers will launch in their
own new screen, named after the server name.
You can also do:
make test TESTS="some test" SCREEN=1 VALGRIND_SERVER=1
to run valgrind on each samba server, or
make test TESTS="some test" SCREEN=1 GDBTEST=1
to run gdb on each server
Andrew Tridgell [Sat, 2 Oct 2010 19:42:00 +0000 (12:42 -0700)]
s4-selftest: fixed up exit codes on signals for make test
also set $ENVNAME in launched servers
Jelmer Vernooij [Sun, 3 Oct 2010 02:24:46 +0000 (04:24 +0200)]
land: Force always emailing when there is no other mechanism of progress reporting.
Jelmer Vernooij [Sun, 3 Oct 2010 01:56:09 +0000 (03:56 +0200)]
land: Attach tarball of logs rather than individual logs to keep the
mail size reasonable.
Jelmer Vernooij [Sun, 3 Oct 2010 00:16:11 +0000 (02:16 +0200)]
land: Some cosmetic fixes.
Jelmer Vernooij [Sat, 2 Oct 2010 23:31:11 +0000 (01:31 +0200)]
land: Attach test output files to result emails.
Jelmer Vernooij [Sat, 2 Oct 2010 22:50:53 +0000 (00:50 +0200)]
land: Add --revision argument.
Jelmer Vernooij [Sat, 2 Oct 2010 22:10:47 +0000 (00:10 +0200)]
land-remote: Run remote land command unbuffered.
Andrew Bartlett [Sat, 2 Oct 2010 06:55:06 +0000 (16:55 +1000)]
s4-kdc Fix up after import of new lorikeet-heimdal
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Sun Oct 3 01:56:04 UTC 2010 on sn-devel-104
Andrew Bartlett [Sat, 2 Oct 2010 06:52:52 +0000 (16:52 +1000)]
Add new files for sha512 support
Andrew Bartlett [Sat, 2 Oct 2010 06:32:56 +0000 (16:32 +1000)]
Andrew Bartlett [Sat, 25 Sep 2010 04:34:46 +0000 (14:34 +1000)]
s4-heimdal We don't need HDBDIR any more
Matthieu Patou [Fri, 20 Aug 2010 08:01:46 +0000 (12:01 +0400)]
heimdal: change the version string
Andrew Bartlett [Sat, 25 Sep 2010 04:58:09 +0000 (14:58 +1000)]
s4-heimdal_build fix up build after heimdal import
Heimdal has it's own dynconfig like system, and so we need the
LIBDIR etc in the CFLAGS here.
We also need to define build rules for the new files imported in
the merge commit, and fix up some header files.
This includes the work by Matthieu Patou <mat@matws.net>
Andrew Bartlett
Matthieu Patou [Sun, 15 Aug 2010 14:31:28 +0000 (18:31 +0400)]
s4:heimdal: import lorikeet-heimdal-
201009250123 (commit
42cabfb5b683dbcb97d583c397b897507689e382)
I based this on Matthieu's import of lorikeet-heimdal, and then
updated it to this commit.
Andrew Bartlett
Kamen Mazdrashki [Fri, 1 Oct 2010 00:30:50 +0000 (03:30 +0300)]
s4-ldapcmp.py: Don't guess credentials for second Credentials object
This allow us to fallback to first credentials given.
Autobuild-User: Kamen Mazdrashki <kamenim@samba.org>
Autobuild-Date: Sat Oct 2 23:05:20 UTC 2010 on sn-devel-104
Kamen Mazdrashki [Fri, 1 Oct 2010 00:29:33 +0000 (03:29 +0300)]
s4-getopt.py: Make Anonymous creds when no credentials
are supplied on command line and caller doesn't want us
to guess credentials from environment
Kamen Mazdrashki [Thu, 30 Sep 2010 18:42:17 +0000 (21:42 +0300)]
s4-test-fsmo.py: Implement kind of busy-wait for role transfer
This allows us not to hard code timeout we are going to wait for
before failing the test - if DCs are not very busy, we won't wait
at all (speeds up 'make test' by a minute also)
Maximum timeout to wait for is set to 20 secs currently,
which should be enough even for slow machines and tested
DCs being under load
Kamen Mazdrashki [Wed, 29 Sep 2010 12:56:23 +0000 (15:56 +0300)]
s4-test-dsdb_schema_info.py: Get rid of global module variables
Kamen Mazdrashki [Wed, 29 Sep 2010 12:51:11 +0000 (15:51 +0300)]
s4-test-dsdb_schema_info.py: Simplify connection SamDB
by using samba.tests.connect_samdb() helper
Kamen Mazdrashki [Wed, 29 Sep 2010 13:00:43 +0000 (16:00 +0300)]
s4-test-fsmo.py: Use common implementation for env_get_var_value()
Kamen Mazdrashki [Wed, 29 Sep 2010 13:00:04 +0000 (16:00 +0300)]
s4-test-delete_object.py: Use common implementation for env_get_var_value()
Kamen Mazdrashki [Wed, 29 Sep 2010 11:53:12 +0000 (14:53 +0300)]
s4-python-test: Common implementation for getting environment variable value
Unit-test based python tests require certain input parameters
to be set in environment, otherwise they can't be run
Kamen Mazdrashki [Wed, 29 Sep 2010 01:31:02 +0000 (04:31 +0300)]
s4-python-samba: Remove trailing ';'s
Kamen Mazdrashki [Wed, 29 Sep 2010 00:02:11 +0000 (03:02 +0300)]
s4-test-fsmo.py: Use common implementation for connect_samdb()
Connections are restricted to LDAP only (remote),
as this test can't be run against local LDBs
Kamen Mazdrashki [Wed, 29 Sep 2010 00:00:18 +0000 (03:00 +0300)]
s4-test-delete_object.py: Use common implementation for connect_samdb()
Connections are restricted to LDAP only (remote),
as this test can't be run against local LDBs
Kamen Mazdrashki [Tue, 28 Sep 2010 23:58:23 +0000 (02:58 +0300)]
s4-python-test: Implement global connect_samdb() function
This helper makes proper ldb url to connect to
and is a shorthand for test to create SamDB connections
Kamen Mazdrashki [Tue, 28 Sep 2010 22:55:22 +0000 (01:55 +0300)]
s4-samba.samdb: Fix masking names from outer context
- 'filter' is built-in
- 'ldb' is a module name we imported
Kamen Mazdrashki [Tue, 28 Sep 2010 22:53:22 +0000 (01:53 +0300)]
s4-python-samba: 'file' is a built-in
Kamen Mazdrashki [Tue, 28 Sep 2010 22:35:51 +0000 (01:35 +0300)]
s4-samba.samdb: Fix leading indention and trailing ';'
Kamen Mazdrashki [Tue, 28 Sep 2010 22:09:09 +0000 (01:09 +0300)]
s4-python-samba: Fix few cosmetics
- we have sys module already imported
- _glue module is part of samba package so be more precise how to import
Volker Lendecke [Sat, 2 Oct 2010 22:15:39 +0000 (00:15 +0200)]
s3: Attempt to fix the non-ads build
Jelmer Vernooij [Sat, 2 Oct 2010 22:04:59 +0000 (00:04 +0200)]
land-remote: Pass extra arguments on to land.
Jelmer Vernooij [Sat, 2 Oct 2010 21:40:19 +0000 (23:40 +0200)]
pytdb: Include Python.h first to prevent warning.
Jelmer Vernooij [Sat, 2 Oct 2010 21:32:12 +0000 (23:32 +0200)]
subunithelper: Remove accidentally added line.
Kirill Smelkov [Sat, 2 Oct 2010 13:43:50 +0000 (17:43 +0400)]
pytdb: Check errors after PyObject_New() calls
The call could fail with e.g. MemoryError, and we'll dereference NULL
pointer without checking.
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Kirill Smelkov [Sat, 2 Oct 2010 13:43:46 +0000 (17:43 +0400)]
pytdb: Add support for tdb_repack()
Cc: 597386@bugs.debian.org
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Kirill Smelkov [Sat, 2 Oct 2010 13:43:40 +0000 (17:43 +0400)]
pytdb: Add TDB_INCOMPATIBLE_HASH open flag
In 2dcf76 Rusty added TDB_INCOMPATIBLE_HASH open flag which selects
Jenkins lookup3 hash for new databases.
Expose this flag to python users too.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Kirill Smelkov <kirr@mns.spb.ru>
Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Jelmer Vernooij [Sat, 2 Oct 2010 20:31:31 +0000 (22:31 +0200)]
subunithelper: Fix format time.
Jelmer Vernooij [Sat, 2 Oct 2010 18:32:16 +0000 (20:32 +0200)]
land: Implement --fail-immediately directly in Python, provide subunit file.
Jelmer Vernooij [Sat, 2 Oct 2010 16:41:14 +0000 (18:41 +0200)]
subunithelper: Make filter options optional.
Jelmer Vernooij [Sat, 2 Oct 2010 16:40:44 +0000 (18:40 +0200)]
format-subunit: Split out summary file writing.
Jelmer Vernooij [Sat, 2 Oct 2010 15:02:02 +0000 (17:02 +0200)]
selftest: Move plain text formatter to subunithelper.
Jelmer Vernooij [Sat, 2 Oct 2010 11:26:36 +0000 (13:26 +0200)]
land: Add separate treestagebuilder for subunit.
Jelmer Vernooij [Sat, 2 Oct 2010 11:17:25 +0000 (13:17 +0200)]
land: Cherry-pick tridges changes to autobuild.
Jelmer Vernooij [Sat, 2 Oct 2010 11:13:37 +0000 (13:13 +0200)]
land: Move stage building into a separate class so we can have subclasses (e.g. subunit-specific).
Jelmer Vernooij [Sat, 2 Oct 2010 00:59:21 +0000 (02:59 +0200)]
land: Only pass shell=True if necessary.
Jelmer Vernooij [Fri, 1 Oct 2010 20:10:13 +0000 (22:10 +0200)]
land: Add --fail-slowly option.
Jelmer Vernooij [Fri, 1 Oct 2010 19:55:10 +0000 (21:55 +0200)]
Several smaller cleanups, use python coding style.
Jelmer Vernooij [Fri, 1 Oct 2010 19:43:20 +0000 (21:43 +0200)]
land: Avoid running things in a shell where not necessary.
Jelmer Vernooij [Fri, 1 Oct 2010 17:24:20 +0000 (19:24 +0200)]
land-remote: Checkout repository first.
Jelmer Vernooij [Fri, 1 Oct 2010 19:35:35 +0000 (21:35 +0200)]
land-remote: Use --repository option.
Jelmer Vernooij [Fri, 1 Oct 2010 19:34:31 +0000 (21:34 +0200)]
land: Add --repository option.
Jelmer Vernooij [Fri, 1 Oct 2010 17:03:18 +0000 (19:03 +0200)]
land: Move more functionality onto builder.
Jelmer Vernooij [Fri, 1 Oct 2010 16:52:10 +0000 (18:52 +0200)]
land: Avoid more uses of chdir().
Jelmer Vernooij [Fri, 1 Oct 2010 19:28:32 +0000 (21:28 +0200)]
Split land.py back out of autobuild.py, so I can change it without
risking to break the autobuild system.
Volker Lendecke [Sat, 2 Oct 2010 09:50:26 +0000 (11:50 +0200)]
s3: Attempt to fix bug 7665
Quite a few of our internal routines put stuff on talloc_tos() these days.
In top-level netapi routines, properly allocate a stackframe and clean it
again. Also, don't leak memory in the rpccli_ callers onto the libnetapi
context.
Volker Lendecke [Sat, 2 Oct 2010 09:13:34 +0000 (11:13 +0200)]
s3: Fix a 64-bit BUG
Quick explanation why this is a problem:
It is not safe to pass a pointer to a uint32_t into a routine that expects
a size_t *. The routine that this pointer is passed to has no chance to find
out that in reality it was passed a pointer to a uint32_t. On platforms where
sizeof(size_t)==8, we get a nice overwrite of a stack area that should not
be overwritten.
As this happens over an over again: Does someone know a means to make gcc
stop with an error in this case?
Volker
Andrew Tridgell [Sat, 2 Oct 2010 05:07:04 +0000 (22:07 -0700)]
bisect: more bisection options
and fixes for the old ones
Autobuild-User: Andrew Tridgell <tridge@samba.org>
Autobuild-Date: Sat Oct 2 06:14:46 UTC 2010 on sn-devel-104
Andrew Tridgell [Sat, 2 Oct 2010 04:10:39 +0000 (21:10 -0700)]
script: added bisect-test.py git bisect script
this can be used to work out what commit broke a set of tests
Andrew Tridgell [Sat, 2 Oct 2010 02:53:34 +0000 (19:53 -0700)]
autobuild: show top commit in emails
Andrew Tridgell [Sat, 2 Oct 2010 02:53:13 +0000 (19:53 -0700)]
autobuild: include autobuild.log in logs.tar.gz
Andrew Tridgell [Sat, 2 Oct 2010 02:07:01 +0000 (19:07 -0700)]
s4-repl: use the GC principal name for DRS replication connection
this is required when talking to RODCs (for notify calls), and is good
practice for all DCs
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
Andrew Tridgell [Sat, 2 Oct 2010 02:05:30 +0000 (19:05 -0700)]
s4-rpc: added target_principal binding handle option
this allows you to specify a target SPN for a connection
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
Andrew Tridgell [Sat, 2 Oct 2010 02:04:44 +0000 (19:04 -0700)]
s4-dsdb: added dsdb_search_by_dn_guid()
this is more efficient than first searching for the DN, then doing a
search. We should look at using this in lots of existing code
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
Stefan Metzmacher [Fri, 1 Oct 2010 07:53:05 +0000 (09:53 +0200)]
s4:rpc_server/netlogon: don't use dcerpc_binding_handle_call_send/recv() directly
metze
Autobuild-User: Stefan Metzmacher <metze@samba.org>
Autobuild-Date: Sat Oct 2 03:11:38 UTC 2010 on sn-devel-104
Stefan Metzmacher [Wed, 29 Sep 2010 22:26:33 +0000 (00:26 +0200)]
netlogon.idl: add missing flags to DSGETDC_VALID_FLAGS
metze
Andrew Bartlett [Sat, 2 Oct 2010 00:58:46 +0000 (10:58 +1000)]
s4-gensec Always honour the set server principal
The spengo code won't set this unless it is allowed to by this
same option, but other callers may need it.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Sat Oct 2 02:27:39 UTC 2010 on sn-devel-104
Andrew Tridgell [Fri, 1 Oct 2010 23:19:28 +0000 (16:19 -0700)]
s4-drs: fixed comparison login in replicated renames
we need to ensure we only ever compare USNs from the same originating
invocation ID.
Autobuild-User: Andrew Tridgell <tridge@samba.org>
Autobuild-Date: Sat Oct 2 01:45:19 UTC 2010 on sn-devel-104