Andrew Bartlett [Thu, 10 Jul 2008 07:54:43 +0000 (17:54 +1000)]
Avoid the use of extensibleObject in ldap mapping backend.
Instead of extensibleObject, we use the new (more correct) ad2oLschema
tool, and a new objectClass called 'samba4Top', which we add and
remove in the same way we did extensibleObject.
Andrew Bartlett
Andrew Bartlett [Thu, 10 Jul 2008 05:52:44 +0000 (15:52 +1000)]
Make ad2oLschema even simpler, by moving the heavy work into dsdb.
This will allow the kludge_acl and schema code to leverage the same
work. (We might even get schema validation soon! :-)
Andrew Bartlett
Michael Adam [Fri, 27 Jun 2008 07:22:53 +0000 (09:22 +0200)]
ldb_map: eliminate "discards qualifyer" warning (const).
Michael
Michael Adam [Fri, 27 Jun 2008 07:22:11 +0000 (09:22 +0200)]
libnet_domain: eliminate "discards qualifier" warning (const).
Michael
Michael Adam [Thu, 26 Jun 2008 21:30:42 +0000 (23:30 +0200)]
dsdb: eliminate disards qualifyer compile warning.
Michael
Michael Adam [Wed, 9 Jul 2008 10:48:05 +0000 (12:48 +0200)]
Revert "tdbtool: fix off-by-one error in argument length. (bug #2344)"
This reverts commit
fafb8ad2b81b9a46cf8259bedc1dca5023b06115.
This fix is not valid:
1. convert_string() is not only used for key strings but also for data.
2. Some databases use string_tdb_data() i.e. non-null-terminated strings
as keynames and others (like the one I was using), use
string_term_tdb_data(), i.e. zero-terminated key strings.
After discussion with Metze, the easiest (and proper way) to
handle this is to specify key names as "keyname\0" for databases
which use string_term_tdb_data().
Sorry for the noise...
Michael
Michael Adam [Tue, 8 Jul 2008 13:33:36 +0000 (15:33 +0200)]
tdbtool: fix off-by-one error in argument length. (bug #2344)
This prevented all commands operating on keys (all non-traverse commands)
in tdbtool to fail with a "fetch failed" or "delete failed" message.
It seems that it fixes bug #2344 ...
Apparently this bug was introduced with
94e53472666ed in 2005.
Either nobody is using tdbtool or else tdb_find() has become
more strict about the key legth in the meantime. :-)
Michael
Stefan Metzmacher [Sat, 17 May 2008 00:40:37 +0000 (02:40 +0200)]
rpc_server: use the transport session_info if available
metze
Stefan Metzmacher [Mon, 7 Jul 2008 16:55:59 +0000 (18:55 +0200)]
packet: make it possible to free the packet_context from the send_callback
metze
(cherry picked from commit
20795c4a0d5f75561561470231de1a2fad2906ff)
Stefan Metzmacher [Mon, 7 Jul 2008 17:40:35 +0000 (19:40 +0200)]
smb_server/smb: handle incoming multi fragmented nttrans requests
metze
Stefan Metzmacher [Mon, 7 Jul 2008 17:37:14 +0000 (19:37 +0200)]
smb_server/smb: prepare multi fragmented nttrans requests
metze
Stefan Metzmacher [Mon, 7 Jul 2008 14:34:36 +0000 (16:34 +0200)]
libcli/raw: remove unused smb_raw_max_trans_data() function
metze
Stefan Metzmacher [Mon, 7 Jul 2008 16:07:47 +0000 (18:07 +0200)]
torture: .in.max_data should not depend on the smb max size
We now support async multi fragment SMBtrans calls.
metze
Stefan Metzmacher [Mon, 7 Jul 2008 16:04:44 +0000 (18:04 +0200)]
rap: trans->in.max_data should not depend on the smb max size
We now support async multi fragment SMBtrans calls.
metze
Stefan Metzmacher [Mon, 7 Jul 2008 16:01:28 +0000 (18:01 +0200)]
dcerpc_smb: trans->in.max_data should not depend on the smb max size
We now support async multi fragment SMBtrans calls.
metze
Stefan Metzmacher [Mon, 7 Jul 2008 13:04:59 +0000 (15:04 +0200)]
libcli/raw: make multi fragmented nttrans requests possible
metze
Stefan Metzmacher [Mon, 7 Jul 2008 12:00:53 +0000 (14:00 +0200)]
smb_server/smb: trans(2) setup count is uint8_t
metze
Stefan Metzmacher [Mon, 7 Jul 2008 12:00:09 +0000 (14:00 +0200)]
libcli/raw: trans(2) setup count is uint8_t
metze
Stefan Metzmacher [Fri, 4 Jul 2008 18:47:24 +0000 (20:47 +0200)]
libcli/raw: remove unused smbcli_request_receive_more() function
metze
Stefan Metzmacher [Fri, 4 Jul 2008 18:46:10 +0000 (20:46 +0200)]
libcli/raw: use the new recv_helper infrastructure for trans/trans2 replies
metze
Stefan Metzmacher [Fri, 4 Jul 2008 18:07:52 +0000 (20:07 +0200)]
libcli/raw: use the new recv_helper infrastructure for nttrans replies
metze
Stefan Metzmacher [Fri, 4 Jul 2008 17:52:23 +0000 (19:52 +0200)]
libcli/raw: add a recv_helper hook infrastructure
The recv helper will be called when a response comes
and the recv helper can decide to let the request
on the SMBCLI_REQUEST_RECV when more reponse packets
are expected. It's up to the helper function
to keep a reference to the in buffers, each incoming
response overwrites req->in.
metze
Stefan Metzmacher [Fri, 4 Jul 2008 18:16:36 +0000 (20:16 +0200)]
libcli/raw: the nttrans setup count is only 8-bit
metze
Stefan Metzmacher [Mon, 7 Jul 2008 11:00:24 +0000 (13:00 +0200)]
smb_server/smb: transs and transs2 calls have different word counts
Also add a note about NT_STATUS_DOS(ERRSRV, ERRerror).
metze
Stefan Metzmacher [Mon, 7 Jul 2008 10:56:26 +0000 (12:56 +0200)]
smb_server/smb: fix crash bug with fragmented trans calls
We need to use smbsrv_setup_secondary_request(req) to send the
trans ack, because smbsrv_send_reply(req) destroys 'req'
and the partial trans list had dead elements in the list.
Also make sure the partial list element is removed by a talloc
destructor.
metze
Stefan Metzmacher [Thu, 3 Jul 2008 11:40:31 +0000 (13:40 +0200)]
dcerpc.idl: the signing overhead can be more than 32 bytes
Windows seems to use 64 here, so we do now.
Before we got nca_proto_error fault because we send fragments
larger than the negotiated max frag size.
If the max frag size is 5840, we're sending 5837 bytes
when the auth_len is 45 and that matches w2k3 traffic.
metze
Stefan Metzmacher [Thu, 3 Jul 2008 11:39:55 +0000 (13:39 +0200)]
rpc_server: use the same chunk_size logic as we we use in the client
metze
Jeremy Allison [Wed, 2 Jul 2008 19:01:15 +0000 (12:01 -0700)]
Fix bug #5578, reported by sendel2000@hotbox.ru. Bad (non-Samba) use of strlcat gives error.
Jeremy.
Stefan Metzmacher [Wed, 2 Jul 2008 13:43:57 +0000 (15:43 +0200)]
Merge commit 'origin/v4-0-stable' into v4-0-test
Stefan Metzmacher [Wed, 2 Jul 2008 13:33:20 +0000 (15:33 +0200)]
Merge commit 'release-4-0-0alpha5' into v4-0-stable
Andrew Bartlett [Wed, 2 Jul 2008 11:47:21 +0000 (21:47 +1000)]
Merge branch 'v4-0-test' of ssh://git.samba.org/data/git/samba into 4-0-local
Andrew Bartlett [Wed, 2 Jul 2008 11:30:08 +0000 (21:30 +1000)]
Move ad2oLschema and oLschema2ldif into Samba4, out of LDB
LDB does not know about nor process the AD schema, so it makes no
sense to have this tool there. I've been changing it anyway, to use a
common schema manipulation library, and will enhance these links in
the future.
Andrew Bartlett
Stefan Metzmacher [Wed, 2 Jul 2008 08:28:42 +0000 (10:28 +0200)]
ldap_server: allow modifies to the root dse record and pass them to the ldb layer
metze
Andrew Bartlett [Wed, 2 Jul 2008 05:19:49 +0000 (15:19 +1000)]
Fill in the auxiliary classes into the dsdb_schema.
Andrew Bartlett
Andrew Bartlett [Wed, 2 Jul 2008 05:15:54 +0000 (15:15 +1000)]
Collapse auxillary classes in LDAP schema conversion.
MS-ADTS 3.1.1.3.1.1.5 describes the behaviour of auxiliary classes.
In effect, these are additional MUST or MAY attributes that are
appeneded to the parent class (the auxiliary does not become listed in
the objectClass attribute), and so we do just that, and merge them
here, for export to OpenLDAP as it's schema.
Andrew Bartlett
Stefan Metzmacher [Mon, 30 Jun 2008 15:17:24 +0000 (17:17 +0200)]
schema_fsmo: prepare auto allocation of schema oid prefixes
This implements the logic in the schema_fsmo_add() function,
but it only calls a dummy dsdb_create_prefix_mapping() yet.
metze
Stefan Metzmacher [Tue, 1 Jul 2008 14:35:13 +0000 (16:35 +0200)]
schema_fsmo: move fsmo info into struct dsdb_schema
metze
Andrew Bartlett [Tue, 1 Jul 2008 04:54:23 +0000 (14:54 +1000)]
Merge branch 'v4-0-test' of ssh://git.samba.org/data/git/samba into 4-0-local
Andrew Bartlett [Tue, 1 Jul 2008 04:53:31 +0000 (14:53 +1000)]
Create PREFIX/var/lib and PREFIX/var/run in 'make install'.
This is a problem because the location of the winbind privilaged pipe
moved.
Andrew Bartlett
Stefan Metzmacher [Mon, 30 Jun 2008 15:15:49 +0000 (17:15 +0200)]
schema_fsmo: only set module->private_data in case we keep the struct for a long time
metze
Stefan Metzmacher [Mon, 30 Jun 2008 11:03:17 +0000 (13:03 +0200)]
drsuapi.idl: remove some unknows from DsGetNCChanges() (update samba4 callers)
metze
Stefan Metzmacher [Mon, 30 Jun 2008 11:01:02 +0000 (13:01 +0200)]
drsuapi.idl: remove some unknows from DsGetNCChanges()
metze
Andrew Bartlett [Mon, 30 Jun 2008 11:54:43 +0000 (21:54 +1000)]
A couple more package updates
Andrew Bartlett [Mon, 30 Jun 2008 11:54:28 +0000 (21:54 +1000)]
On the way to alpha6!
Andrew Bartlett [Mon, 30 Jun 2008 11:54:01 +0000 (21:54 +1000)]
Merge branch '4-0-stable' into 4-0-local
Andrew Bartlett [Mon, 30 Jun 2008 09:34:11 +0000 (19:34 +1000)]
Merge branch '4-0-local' into v4-0-stable
Andrew Bartlett [Mon, 30 Jun 2008 09:34:02 +0000 (19:34 +1000)]
More updates for spec file, for alpha5
Andrew Bartlett [Mon, 30 Jun 2008 09:31:20 +0000 (19:31 +1000)]
Ensure we install smbd as well...
Andrew Bartlett [Mon, 30 Jun 2008 08:12:20 +0000 (18:12 +1000)]
Mark as 'not a git snapshot'.
Andrew Bartlett [Mon, 30 Jun 2008 08:07:26 +0000 (18:07 +1000)]
Prepare for alpha5 tarball.
Now we have packaging/Fedora in the tree, make that directory or copy
files into there.
Andrew Bartlett
Andrew Bartlett [Mon, 30 Jun 2008 08:06:02 +0000 (18:06 +1000)]
Merge branch 'v4-0-test' of ssh://git.samba.org/data/git/samba into 4-0-local
Stefan Metzmacher [Sat, 28 Jun 2008 18:42:27 +0000 (20:42 +0200)]
nbt_server/wins: don't force wins_ldb as shared_module
metze
Stefan Metzmacher [Fri, 27 Jun 2008 13:34:31 +0000 (15:34 +0200)]
dsdb: don't force the build of ldb modules as shared_module
metze
Stefan Metzmacher [Sat, 28 Jun 2008 18:41:45 +0000 (20:41 +0200)]
ldb_ildap: don't force the build as shared_module
metze
Stefan Metzmacher [Sat, 28 Jun 2008 08:50:49 +0000 (10:50 +0200)]
auth: don't force the build of auth_server as shared_module
metze
Stefan Metzmacher [Sat, 28 Jun 2008 09:01:20 +0000 (11:01 +0200)]
lib/talloc: build as MERGED_OBJ instead of STATIC_LIBRARY
metze
Stefan Metzmacher [Sat, 28 Jun 2008 08:59:02 +0000 (10:59 +0200)]
lib/events: build as MERGED_OBJ instead of STATIC_LIBRARY
metze
Stefan Metzmacher [Fri, 27 Jun 2008 13:33:36 +0000 (15:33 +0200)]
smb_build: use MERGED_OBJ instead of STATIC_LIBRARY
metze
Stefan Metzmacher [Fri, 27 Jun 2008 13:31:35 +0000 (15:31 +0200)]
smb_build: enable implicit make rules
metze
Stefan Metzmacher [Sat, 28 Jun 2008 08:49:49 +0000 (10:49 +0200)]
ldb: allow ldb modules to specify LDB_MODULE(name) or LDB_BACKEND(name)
metze
Stefan Metzmacher [Sat, 28 Jun 2008 08:49:49 +0000 (10:49 +0200)]
ldb: fix the init function names for some modules
metze
Andrew Bartlett [Mon, 30 Jun 2008 04:08:03 +0000 (14:08 +1000)]
Mark as alpha5 release
Andrew Bartlett [Mon, 30 Jun 2008 04:04:43 +0000 (14:04 +1000)]
Fix small formatting details in WHATSNET for an alpha5 release.
Andrew Bartlett
Andrew Bartlett [Mon, 30 Jun 2008 03:42:50 +0000 (13:42 +1000)]
Update Fedora packaging per feedback on review ticket.
https://bugzilla.redhat.com/show_bug.cgi?id=453083
Andrew Bartlett [Mon, 30 Jun 2008 02:22:50 +0000 (12:22 +1000)]
Rework samdb handling for 'netlogon' packets.
Don't reopen the samdb for every netlogon packet, and use the
system_session(), as we must access data not available to anonymous.
Perhaps we should consider a 'authenticated but not system' token, if
we want more control on this.
Andrew Bartlett
Andrew Bartlett [Mon, 30 Jun 2008 01:27:55 +0000 (11:27 +1000)]
Relax the ldap.py tests so that they pass against OpenLDAP CVS.
This requires a patched OpenLDAP checked out from their CVS HEAD. I
hope to see the patches incorporated soon.
Andrew Bartlett
Andrew Bartlett [Sun, 29 Jun 2008 23:38:52 +0000 (09:38 +1000)]
Loosen ldap.py tests on the LDAP backend.
Allow LDAP backend to return parts of the DN in a different case, when
verifying memberOf.
Andrew Bartlett
Jelmer Vernooij [Sat, 28 Jun 2008 13:26:46 +0000 (15:26 +0200)]
Non-zero exit code on failure.
Andrew Bartlett [Sat, 28 Jun 2008 12:04:19 +0000 (22:04 +1000)]
Merge branch 'v4-0-test' of ssh://git.samba.org/data/git/samba into 4-0-local
Andrew Bartlett [Sat, 28 Jun 2008 12:03:52 +0000 (22:03 +1000)]
selftest changes for 'winbind privileged dir' change.
Andrew Bartlett [Sat, 28 Jun 2008 12:02:58 +0000 (22:02 +1000)]
dynconfig changes for 'winbind privileged dir' changes.
Andrew Bartlett [Sat, 28 Jun 2008 12:02:19 +0000 (22:02 +1000)]
Use a configure-specified directory for the winbind priv pipe
This makes it easier for RPMs to specify an group for access to that
directory.
Andrew Bartlett
Andrew Bartlett [Sat, 28 Jun 2008 11:38:19 +0000 (21:38 +1000)]
More work to use %{name} rather than 'samba'
Andrew Bartlett [Sat, 28 Jun 2008 11:37:00 +0000 (21:37 +1000)]
Update WHATSNEW towards an alpha5 release.
Andrew Bartlett
Stefan Metzmacher [Sat, 28 Jun 2008 08:28:15 +0000 (10:28 +0200)]
pvfs: create a pvfs_acl subsystem
That means that the pvfs_acl implementations no longer register
as ntvfs modules (which was wrong)
metze
Stefan Metzmacher [Sat, 28 Jun 2008 08:27:16 +0000 (10:27 +0200)]
pvfs: pvfs_aio should be a subsystem instead of an extra ntvfs module
metze
Andrew Bartlett [Sat, 28 Jun 2008 08:25:47 +0000 (18:25 +1000)]
Place the Fedroa package into Samba4's GIT tree
I'll keep this here at least until Fedora picks things up.
Andrew Bartlett
Andrew Bartlett [Sat, 28 Jun 2008 08:12:10 +0000 (18:12 +1000)]
Tone down the language in BUGS.txt
Andrew Bartlett [Sat, 28 Jun 2008 08:04:35 +0000 (18:04 +1000)]
Fix the wbinfo test on the LDAP backend.
The problem was that we would do a blocking wait for the LDAP server,
which was also blocking on us returning (because we were in single
process mode).
The LDAP connection being made here is useless anyway, and will need
to be an async ldb_connect() before anybody reintroucues it (nobody in
their right mind would program a winbindd backend on pure LDAP, when
the ldb abstraction is available).
Andrew Bartlett
Michael Adam [Fri, 27 Jun 2008 11:21:42 +0000 (13:21 +0200)]
ntlm_auth: eliminate "cast pointer from integer of different size" warnings on 64bit.
Michael
Michael Adam [Fri, 27 Jun 2008 10:42:07 +0000 (12:42 +0200)]
ntlmssp: get rid of "cast to pointer from integer of different size" warning on 64 bit
Not sure if this check is actually necessary, though:
the second should be sufficient.
Michael
Michael Adam [Fri, 27 Jun 2008 06:23:26 +0000 (08:23 +0200)]
tdb_unpack: Eliminate "cast to pointer from integer of different size" warning on 64bit.
The 'p' type is just a flag to mark the presence of a pointer,
not a real pointer itself.
The code is now the same as in Samba3's tdb_unpack.
Michael
Stefan Metzmacher [Fri, 27 Jun 2008 09:34:05 +0000 (11:34 +0200)]
gsskrb5: just don't force, but allow the flags when GSS_CF_NO_CI_FLAGS is given
metze
Stefan Metzmacher [Fri, 27 Jun 2008 09:33:00 +0000 (11:33 +0200)]
gsskrb5: fix gss_krb5_cred_no_ci_flags_x_oid_desc variable name
metze
Stefan Metzmacher [Fri, 27 Jun 2008 09:31:49 +0000 (11:31 +0200)]
credentials: gss_set_cred_option() doesn't like GSS_C_NO_BUFFER
metze
Julien Kerihuel [Fri, 30 May 2008 22:47:14 +0000 (00:47 +0200)]
Fix if_version for interfaces with a if_version_minor field different from 0
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Volker Lendecke [Fri, 27 Jun 2008 10:14:41 +0000 (12:14 +0200)]
Fix a non-empty line with just spaces
Volker Lendecke [Fri, 27 Jun 2008 10:14:19 +0000 (12:14 +0200)]
Fix the definition of abspath for gnu make < 3.81
Julien Kerihuel [Fri, 27 Jun 2008 07:55:30 +0000 (09:55 +0200)]
Remove events from require line:
The events library is not installed and prevent OpenChange from linking
with ldb. The following patch removes events reference from ldb pc file.
We are currently fixing this bug in openchange upstream using 'sed' in
our installsamba4.sh script, but this hack won't work with a packaged
version of samba4.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Michael Adam [Fri, 27 Jun 2008 07:12:57 +0000 (09:12 +0200)]
pynbt: eliminate "initialization from incompatible pointer type" warning
by fixing the signature of py_nbt_node_init().
Jelmer - please check!
Michael
Stefan Metzmacher [Fri, 27 Jun 2008 07:40:00 +0000 (09:40 +0200)]
nbt_server/wins: add forward declaration for struct event_context
metze
Stefan Metzmacher [Fri, 27 Jun 2008 07:39:30 +0000 (09:39 +0200)]
lib/ldb: add forward declaration for struct event_context
metze
Michael Adam [Thu, 26 Jun 2008 21:32:37 +0000 (23:32 +0200)]
credentials: gss_set_cred_option() needs gss_cred_id_t * argument, not gss_cred_id_t.
This call was added in
f573c1ff4443f3002c310d3ba29d8c343ad03907.
Metze, please check!
Michael
Stefan Metzmacher [Wed, 25 Jun 2008 17:14:34 +0000 (19:14 +0200)]
mkversion.sh: printf %s is more portable than echo -e
metze
Andrew Bartlett [Fri, 27 Jun 2008 02:56:20 +0000 (12:56 +1000)]
Fix installation of Samba4 into an empty tree.
Jelmer's previous patch failed because sh is very fussy about it's
syntax, but only compaled in this case if you take the 'no existing
binary' path.
Andrew Bartlett
Andrew Bartlett [Fri, 27 Jun 2008 02:03:23 +0000 (12:03 +1000)]
Merge branch 'v4-0-test' of ssh://git.samba.org/data/git/samba into 4-0-local
Jelmer Vernooij [Thu, 26 Jun 2008 09:51:22 +0000 (11:51 +0200)]
Use make magic for installing/uninstalling binaries.
Jelmer Vernooij [Thu, 26 Jun 2008 09:39:59 +0000 (11:39 +0200)]
Use make template for installing binaries.
Jelmer Vernooij [Thu, 26 Jun 2008 08:56:59 +0000 (10:56 +0200)]
Move blackbox tests closer to what they're testing.
Jelmer Vernooij [Thu, 26 Jun 2008 08:11:49 +0000 (10:11 +0200)]
Only try to detect the events lib dir if it wasn't set earlier.