IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
This ensures we follow the dynconfig convention for the socket
directory, and makes it easier to ensure that the client and server
both refer to the same socket.
Andrew Bartlett
Looking for smb.conf files with find -name \*.conf would find and not
find all sorts of things that were just not valid inputs. When
restricted to under source3 it then finds no inputs in the top level
build, as it only worked on the generated smb.conf files from 'make
test' environments.
Instead, just run it on the one file we know is valid, and if we want
to run it on more files, then we can make this test run against more
environments, or specific target files.
Andrew Bartlett
this fixes the top level build on svart and probably other hosts
Autobuild-User: Andrew Tridgell <tridge@samba.org>
Autobuild-Date: Wed Apr 27 08:27:49 CEST 2011 on sn-devel-104
Not all kerberos distributions have this function.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Wed Apr 27 07:39:08 CEST 2011 on sn-devel-104
This is constant data according to the man pages I find for this
fucntion, and causes a segfault to free() when linked to Heimdal. I
am advised that while it is constant for gss_mech_krb5, it may not be
for other mechanisms, so an assert will ensure this is dealt with by
the programmer who extends this code in future.
Andrew Bartlett
This only works for Heimdal and MIT Krb5 1.8, other versions will get
an ACCESS_DEINED error.
We no longer manually verify any details of the PAC in Samba for
GSSAPI logins, as we never had the information to do it properly, and
it is better to have the GSSAPI library handle it.
Andrew Bartlett
Signed-off-by: Jeremy Allison <jra@samba.org>
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Wed Apr 27 00:25:35 CEST 2011 on sn-devel-104
With newer GPFS releases, libgpfs contains a function
gpfs_lib_init(). It marks the pseudo-device the lib uses
as interface to the GPFS kernel module to not be to
destroyed after each call to GPFS.
This saves us some microseconds creating and removing the device
each time we do a call to libgpfs.
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Tue Apr 26 20:52:36 CEST 2011 on sn-devel-104
By reworking the 'fake DNS' file to use struct dns_rr_srv it should be
possible to emulate that resolver layer as well as the Samba4
sockaddr_storage* based layer. This will then give us a common DNS
emulation for 'make test'.
Andrew Bartlett
We always want the s3 binaries, except when building an smbtorture4
for the s3-waf and autoconf builds.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Tue Apr 26 04:27:28 CEST 2011 on sn-devel-104
This is done in name_query_send these days
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Mon Apr 25 19:24:58 CEST 2011 on sn-devel-104
This avoids getting IPv4 addresses as mapped IPv6 addresses
(e.g. ::ffff:192.168.0.1).
Before the bahavior was inconsistent between operating system
and distributions. Some system have IPV6_ONLY as default.
Now we consistently get AF_INET for IPv4 addresses and AF_INET6
for IPv6 addresses.
It also makes it possible to listen only on IPv6 now
as "::" doesn't imply "0.0.0.0" anymore. Which also
avoids confusing log messages that we were not able to
bind to "0.0.0.0".
metze
Most fault codes have a NTSTATUS representation, so use that.
This brings the fault handling in common with the source4/librpc/rpc code,
which make it possible to share more highlevel code, between source3 and
source4 as the error checking can be the same now.
metze
Autobuild-User: Stefan Metzmacher <metze@samba.org>
Autobuild-Date: Sun Apr 24 10:44:53 CEST 2011 on sn-devel-104
The interface version is at 28 now, and this defines cause problem
on some systems, e.g. Tru64 defines vfs_init in the system headers.
metze
Autobuild-User: Stefan Metzmacher <metze@samba.org>
Autobuild-Date: Sat Apr 23 12:07:35 CEST 2011 on sn-devel-104
vfsops struct on Tru64 has a vfs_init function pointer.
Guenther
Autobuild-User: Günther Deschner <gd@samba.org>
Autobuild-Date: Fri Apr 22 01:49:59 CEST 2011 on sn-devel-104
This should finally fix the AIX build and allow to remove AIX specific ifdefs.
Guenther
Signed-off-by: Jeremy Allison <jra@samba.org>
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Thu Apr 21 02:01:20 CEST 2011 on sn-devel-104
This is now API compatible with the existing code in lib/util/charset
lazy_initialize_conv() is no longer called as init_iconv() is called
when the smb.conf is processed, and get_conv_handle() will auto-init
with defaults if required.
load_case_tables_library() is no longer requried as all binaries and
libraries already load these in their entry points, as otherwise all
the other string functions would fail.
Andrew Bartlett
This uses the source3 PAC code (originally from Samba4) with some
small changes to restore functionality needed by the torture tests,
and to have a common API.
Andrew Bartlett
This Heimdal function does not set the global state, and allows the
GSSAPI server to progress further when compiled against Heimdal (such
as in the top level build).
The ability to specify a keytab has been removed from the API as it is
unused, and and the Heimdal function (avoiding setting global
variables) works with an open keytab.
Andrew Bartlett
Correct fix - ensure we have enough length, and correctly null out
passed in structs if not.
Jeremy.
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Tue Apr 19 00:17:08 CEST 2011 on sn-devel-104
The autoconf check for netgrent headers makes sure that the compiler flag used
(-Werror-implicit-function-declaration) is recognized by the compiler before
using it for the test. Mimic this behaviour in the waf checks.
Thanks to gd for noticing this.
Signed-off-by: Günther Deschner <gd@samba.org>
Autobuild-User: Günther Deschner <gd@samba.org>
Autobuild-Date: Mon Apr 18 16:24:15 CEST 2011 on sn-devel-104
(cherry picked from commit 69650a2aaa8649570261df7efccb35ad76d7cfc1)
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Sat Apr 16 16:20:08 CEST 2011 on sn-devel-104
This will help us have a top level 'make test', working around the
fact that the --eanble-s3build has different names for some of these
binaries, without fixing the names forever (we can just change the
mapping later).
Andrew Bartlett
If "inherit owner = yes", pass in the directory owner and group
owner as the target for CREATOR_OWNER and CREATOR_GROUP substitutions,
and also as the owner and primary group of the new security descriptor
being applied to the object.
Jeremy.
We need to read from the echo socket, pass the socket to read
from down to the read routines
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Thu Apr 14 15:34:36 CEST 2011 on sn-devel-104
This requires a small rework of the build system to ensure that the
correct #define statements are made in both the s3 and top level
builds. We now define the various HAVE_ macros in config.h at all
times, using heimdal_build/wscript_configure when that is in use.
Andrew Bartlett
Make sure we use a timeout of 60 seconds, not 60 milliseconds...
This prevented us from successfully using the ncacn_ip_tcp client in a lot of
places, I guess.
Guenther
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Wed Apr 13 18:59:19 CEST 2011 on sn-devel-104
When this was depended on directly as a subsystem, it ended up in
multiple libraries.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Wed Apr 13 08:39:35 CEST 2011 on sn-devel-104
The rap.sam test reads 0xFFFFFFFF as a string in the level 2
r->HomeDir attribute, which once we start validating ASCII strings
fails. This restores a unchecked dos charset for this test only,
until it is determined if the client or server RAP code is at fault.
Andrew Bartlett
Signed-off-by: Andrew Tridgell <tridge@samba.org>
The s4 implementation didn't do multibyte strings, so was only good
for '/' which is known to be safe in all multibyte charsets.
Andrew Bartlett
Signed-off-by: Andrew Tridgell <tridge@samba.org>
This means that there is no need for the 'valid.dat' table to be
loaded by anything other than smbd, so the unloader is also removed.
The concept of a 'valid dos character' has been replaced by the hash2
mangle method.
Andrew Bartlett
Signed-off-by: Andrew Tridgell <tridge@samba.org>
In general we don't manipulate UTF16 strings internally, particularly
as they are also multibyte, so are no easier to work with than UTF8.
Andrew Bartlett
Signed-off-by: Andrew Tridgell <tridge@samba.org>
We create a fake connection here and don't have an vuid. So work with
the session_info directly here.
Signed-off-by: Jeremy Allison <jra@samba.org>
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Mon Apr 11 22:56:12 CEST 2011 on sn-devel-104
This uses the provided session_info instead of searching the user via
the vuid. This is useful to work with fake connnection you need to
create if someone connects directly to a rpc service.
Signed-off-by: Jeremy Allison <jra@samba.org>
This makes the startup of smbd in make test much quicker and thus more reliable
(cherry picked from commit f1aa38b414e97d8687d0bebf65baa384f75301b4)
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Mon Apr 11 22:09:58 CEST 2011 on sn-devel-104
In the initial phase nmbd does not yet have timed events. This led to nmbd
not correctly registering its names in make test and certainly everywhere
else.
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Mon Apr 11 14:35:45 CEST 2011 on sn-devel-104
Print child descriptor instead of parent.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User: Stefan Metzmacher <metze@samba.org>
Autobuild-Date: Mon Apr 11 11:48:42 CEST 2011 on sn-devel-104
Caused by premature optimisation storing the parent ACL on the
module handle instead of (correctly) on the file fsp. Previous
code wasn't reentrant safe. This is less optimal but doesn't
crash in the specific case :-).
Jeremy.
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Sat Apr 9 02:05:15 CEST 2011 on sn-devel-104
Only print the value number for hwm, version and seqnum keys.
Pair-Programmed-With: Gregor Beck <gbeck@sernet.de>
Autobuild-User: Michael Adam <obnox@samba.org>
Autobuild-Date: Fri Apr 8 17:29:31 CEST 2011 on sn-devel-104
This reverts commit 77820d59fe.
I beleive this is now reliable, after the changes to wait for an
actual connection at startup and to use the fake DNS file.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Thu Apr 7 07:20:03 CEST 2011 on sn-devel-104
the ktest tests are currently flakey. Once andrew has found the
problem they can be re-enabled
Autobuild-User: Andrew Tridgell <tridge@samba.org>
Autobuild-Date: Thu Apr 7 03:36:36 CEST 2011 on sn-devel-104
this prevents a symbol duplication with the openssl library, which may
be linked in via a secondary library dependency
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
It is important that a machine account password change does not
invalidate existing tickets.
This is only for the default kerberos method with a password in
secrets.tdb. The keytab based methods are still not tested.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Wed Apr 6 05:24:48 CEST 2011 on sn-devel-104
Andreas Schneider <asn@samba.org> correctly points out that this input
parameter should now be const, and that found a bug where I used then
used it incorrectly as a talloc context.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Wed Apr 6 00:33:31 CEST 2011 on sn-devel-104
This changes auth_session_info_transport to just be a wrapper, rather
than a copy that has to be kept in sync.
As auth_session_info was already wrapped in python, this required
changes to the existing pyauth wrapper and it's users.
Andrew Bartlett
Remove the final incorrect uses of conn->session_info->utok.uid.
When we're in the "admin users" list, then this value is not set
to zero.
Inspired by the comment on this bug by Dmitry Butskoy <dmitry@butskoy.name>.
I'll create a different fix for 3.5.x.
Jeremy.
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Tue Apr 5 21:53:59 CEST 2011 on sn-devel-104
This test looks for any .conf file with [global] in it, and assumes
it's a Samba3 format smb.conf file. If 'make test' has been run in
the top level build, it was finding the smb.conf files it generated.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Mon Apr 4 23:50:06 CEST 2011 on sn-devel-104
Before a auth_serversupplied_info struct can be used for
authorization, the local groups and privileges must be calculated.
create_local_token() now copies the server_info, and then sets the
calulated token and unix groups.
Soon, it will also transform the result into an expanded struct
auth_session_info. Until then, the variable name (server_info vs
session_info provides a clue to the developer about what information
has been entered in the structure).
By moving the calls to create_local_token within the codebase, we
remove duplication, and ensure that the session key (where modified)
is consistently copied into the new structure.
Andrew Bartlett
Ensure we do this for all cases where the stat fails.
Jeremy
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Mon Apr 4 20:08:45 CEST 2011 on sn-devel-104
This is a tool to check the consistency of an idmap tdb database.
The default mode is to scan the database and list invalid entries,
e.g. records with an invalid format, or records which are valid
but for which the reverse mapping entry is missing.
With the "--repair" switch, one can enter an interactive
repair mode which will prompt for each invalid entry found
with the option to delete, skip or edit the record.
There is also a non-interactive repair mode triggered by "--auto"
which will remove all records with invalid content and fill up
mappings which are missing the reverse entry.
The "--test" parameter lets "net idmap check" only list the
changes that would be written and not actually commit them to
the database.
The "--lock" option allows to lock the database already in the
first reading traverse, in order to remove the race when the
database has to be closed and reopened again before writing
the changes.
Signed-off-by: Michael Adam <obnox@samba.org>
Autobuild-User: Michael Adam <obnox@samba.org>
Autobuild-Date: Mon Apr 4 18:21:09 CEST 2011 on sn-devel-104
With a case insensitive file system the stat cache lookup leaked the parent
directorys stat information from unix_convert into the smb_filename. This led
open_file_ntcreate to believe it just created a directory.
In the case where we do the search we already invalidate the stat struct.
Thanks to TAKAHASHI Motonobu for insisting! :-)
Volker
Autobuild-User: Volker Lendecke <vlendec@samba.org>
Autobuild-Date: Sun Apr 3 14:54:27 CEST 2011 on sn-devel-104
The checks are roughtly taken from the autoconf ctdb checks.
I was not able to implement checks with CHECK_DECL, CHECK_TYPE,
CHECK_HEADER and friends, because the ctdb headers seem to need too
special a setup of includes and defines in order to compile.
So I used CHECK_CODE() in all checks.
In the long run, this should be changed.
I supported a --with-ctdb-dir options to allow for building
against a ctdb that is not installed into /usr (e.g. against
a local git checkout). In order to implement this, I had to
hand includes in to the CHECK_CODE function.
Here I found a problem with CHECK_CODE (or even the core waf
conf.check() function: The CHECK_CODE function does not
expand the includes it gets (i.e. '#' is not expanded to the
base dir, and relative paths are left relative). But the core
check() function seems to ignore all include paths that are
not absolute paths. Hence in particular the usual default '# .'
for the includes is useless. So I preprocessed the list of includes
for the cluster checks. But I assume that it would be useful
to move this expansion into CHECK_CODE or even into the core
waf check function.
Autobuild-User: Michael Adam <obnox@samba.org>
Autobuild-Date: Sat Apr 2 03:26:55 CEST 2011 on sn-devel-104
There is no reason for smbd with Windows ACLs to use chmod
or fchmod unless it's a file opened with UNIX extensions or
with posix pathnames.
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Sat Apr 2 02:40:43 CEST 2011 on sn-devel-104
This way, the record will be scheduled for fast vacuuming.
This is sent with the NOREPLY flag, so ctd should not sent
a reply packet and samba does not expect one. Hence, it
is not important for the success of the db_ctdb_delete command
whether or not the ctdbd we are running against supports the
SCHEDULE_FOR_DELETION control.
