Tim Potter [Wed, 26 Jan 2005 04:50:04 +0000 (04:50 +0000)]
r5006: Implement parsing of pidl bitmaps in ethereal parsers. This works
well but needs to be stuck in a subtree.
Tim Potter [Wed, 26 Jan 2005 04:34:16 +0000 (04:34 +0000)]
r5005: Add missing size specifiers to various bitmaps.
Andrew Tridgell [Wed, 26 Jan 2005 03:19:42 +0000 (03:19 +0000)]
r5004: add current samba3 nmblookup.c ready for updating to new nbt lib
Andrew Tridgell [Wed, 26 Jan 2005 03:19:05 +0000 (03:19 +0000)]
r5003: delete old nmblookup.c code
Tim Potter [Tue, 25 Jan 2005 23:14:46 +0000 (23:14 +0000)]
r4993: Generate nicer name for enum hf.
Start work on supporting bitmaps.
Tim Potter [Tue, 25 Jan 2005 22:38:57 +0000 (22:38 +0000)]
r4991: Remove debugging statement.
Tim Potter [Tue, 25 Jan 2005 22:33:44 +0000 (22:33 +0000)]
r4990: Generate value strings for pidl enumerations.
Tim Potter [Tue, 25 Jan 2005 11:38:39 +0000 (11:38 +0000)]
r4983: On second thoughts don't include the config.mk file for the swig stuff
just yet. More testing required.
Tim Potter [Tue, 25 Jan 2005 11:21:25 +0000 (11:21 +0000)]
r4982: Start to move swig dependencies into new build system. Unfortunately
I can only get something useful happening by using the BINARY keyword
as nothing else seems to generate dependency lists that can be used
when linking the swig shared libraries. Anyway this is a lot nicer
than having lots of junk in makefile.pm.
Tim Potter [Tue, 25 Jan 2005 10:20:24 +0000 (10:20 +0000)]
r4981: Fix swig build.
Tim Potter [Tue, 25 Jan 2005 10:03:57 +0000 (10:03 +0000)]
r4980: Copy RAP callno constants from Samba 3 and start to use them.
Tim Potter [Tue, 25 Jan 2005 09:46:00 +0000 (09:46 +0000)]
r4979: Return NT_STATUS_INVALID_SYSTEM_SERVICE for unimplemented RAP calls as
this is what win2k3 does.
Tim Potter [Tue, 25 Jan 2005 02:07:27 +0000 (02:07 +0000)]
r4975: Update usage of talloc in rap torture code.
Don't use ZERO_STRUCTP() when creating a new struct rap_call.
Tim Potter [Tue, 25 Jan 2005 01:39:44 +0000 (01:39 +0000)]
r4974: Update location of talloc_guide.txt
Tim Potter [Tue, 25 Jan 2005 01:21:59 +0000 (01:21 +0000)]
r4973: Add a RAP scanner to smbtorture. win2k has call numbers 0-215 although
the cifs tr lists 250-318 also.
Stefan Metzmacher [Mon, 24 Jan 2005 14:44:15 +0000 (14:44 +0000)]
r4962: add infrastructure to use raw krb5 auth in dcerpc client code
Note this doesn't work currently because the gensec_modules are not ready for that yet
metze
Stefan Metzmacher [Mon, 24 Jan 2005 14:31:19 +0000 (14:31 +0000)]
r4959: fix compiler warning
metze
Stefan Metzmacher [Mon, 24 Jan 2005 14:30:22 +0000 (14:30 +0000)]
r4958: fix compiler warnings
metze
Andrew Tridgell [Mon, 24 Jan 2005 03:43:48 +0000 (03:43 +0000)]
r4957: the fetchfile _recv() function was neglecting to steal the data and
free the connection context. This left a whole lot of state hanging
around and didn't give the memory to the caller properly
Andrew Tridgell [Mon, 24 Jan 2005 03:33:36 +0000 (03:33 +0000)]
r4956: - moved the definition of the mangle context structure into a pvfs_shortname
- made the mangle cache size configurable
Andrew Tridgell [Mon, 24 Jan 2005 02:19:57 +0000 (02:19 +0000)]
r4955: fixed a couple of minor memory leaks in the auth_sam code
Andrew Tridgell [Mon, 24 Jan 2005 01:51:45 +0000 (01:51 +0000)]
r4954: we don't need the separate event_remove_*() calls any more, as you now
remove an event by calling talloc_free().
Andrew Tridgell [Mon, 24 Jan 2005 01:15:30 +0000 (01:15 +0000)]
r4953: - enable easy valgrind use in all our test scripts
- when valgrind isn't set, then setup MALLOC_CHECK_ to abort on error
Andrew Tridgell [Mon, 24 Jan 2005 01:04:15 +0000 (01:04 +0000)]
r4952: removed a bogus talloc_steal() that was trying to cope with the
inverted memory hierarchy. Now the memory hierarchy is logical its not
needed (and can cause a double free in RPC-SCHANNEL)
Andrew Tridgell [Mon, 24 Jan 2005 00:57:14 +0000 (00:57 +0000)]
r4951: some of the code dealing with libcli was getting too complex trying to
handle the inverted memory hierarchy that a normal session
establishment gave. The inverted hierarchy came from that fact that
you first establish a socket, then a transport, then a session and
finally a tree. That leads to the socket being at the top of the
memory hierarchy and the tree at the bottom, which makes no sense from
the users point of view, as they want to be able to free the tree and
have everything disappear.
The core problem was that the libcli interface didn't distinguish
between establishing a primary context and a secondary context. If you
establish a 2nd session on a transport then you want the transport to
be referenced by the session, whereas if you establish a primary
session then you want the transport to be a child of the session.
To fix this I have added "parent_ctx" and "primary" arguments to the
libcli intialisation functions. This makes using the library much
easier, and gives us a memory hierarchy that makes much more sense.
I was prompted to do this by a bug in the cifs backend, which was
caused by the socket not being properly torn down on a disconnect due
to the inverted memory hierarchy.
Andrew Tridgell [Mon, 24 Jan 2005 00:51:06 +0000 (00:51 +0000)]
r4950: removed some excessive debugging messages
Volker Lendecke [Sun, 23 Jan 2005 23:23:26 +0000 (23:23 +0000)]
r4949: First version of a fetchfile composite function which connects to a server and
loads a file. Needs a smb url parsing wrapper.
Volker
Rafal Szczesniak [Sun, 23 Jan 2005 22:14:44 +0000 (22:14 +0000)]
r4948: Typo fixes in comments.
rafal
Andrew Tridgell [Sun, 23 Jan 2005 12:24:36 +0000 (12:24 +0000)]
r4945: the te element isn't needed any more
In general, now that events are children of the structure they are
handling events for, the caller only needs to keep the event handle
around if it plans on changing the event flags later
Andrew Tridgell [Sun, 23 Jan 2005 12:17:45 +0000 (12:17 +0000)]
r4944: every event_add_*() caller was having to call talloc_steal() to take
control of the event, so instead build that into the function. If you
pass NULL as mem_ctx then it leaves it as a child of the events
structure.
Andrew Tridgell [Sun, 23 Jan 2005 11:49:15 +0000 (11:49 +0000)]
r4943: Smplified the events handling code a lot. The first source of
complexity was that events didn't automatically cleanup
themselves. This was because the events code was written before we had
talloc destructors, so you needed to call event_remove_XX() to clean
the event out of the event lists from every piece of code that used
events. I have now added automatic event destructors, which in turn
allowed me to simplify a lot of the calling code.
The 2nd source of complexity was caused by the ref_count, which was
needed to cope with event handlers destroying events while handling
them, which meant the linked lists became invalid, so the ref_count ws
used to mark events for later destruction.
The new system is much simpler. I now have a ev->destruction_count,
which is incremented in all event destructors. The event dispatch code
checks for changes to this and handles it.
Andrew Tridgell [Sun, 23 Jan 2005 11:42:56 +0000 (11:42 +0000)]
r4942: converted the cifs backend to not use event_context_merge(). Instead,
is supplies the server event context during the connect.
Stefan Metzmacher [Sun, 23 Jan 2005 11:12:42 +0000 (11:12 +0000)]
r4941: - all needed data is now in sam.ldb and hacked.ldb is not needed anymore by the hacked ldap backend
- readd the schema naming context container object as it's needed for a w2k3 dc join
metze
Tim Potter [Sun, 23 Jan 2005 11:03:20 +0000 (11:03 +0000)]
r4940: Add a variable to hold a list of typedefs for which we shouldn't generate
a dissector for. A hand written dissector needs to be added to eparser.c
for the plugin to link.
Andrew Tridgell [Sun, 23 Jan 2005 09:03:05 +0000 (09:03 +0000)]
r4939: make a few more private pointers type safe (this might help abartlet
track down a bug he just told me about)
Andrew Tridgell [Sun, 23 Jan 2005 09:01:46 +0000 (09:01 +0000)]
r4938: allow the caller to supply an existing event_context if they want to
in smb_composite_connect_send(). This makes doing parallel calls much
easier.
Andrew Tridgell [Sun, 23 Jan 2005 08:19:38 +0000 (08:19 +0000)]
r4937: simplify the connect code in the same way
Andrew Tridgell [Sun, 23 Jan 2005 08:16:16 +0000 (08:16 +0000)]
r4936: moved to a convention where the completion function is only called in
one place. This makes the code more robust, and simpler (it would have
prevented the error that volker found).
Andrew Tridgell [Sun, 23 Jan 2005 00:51:20 +0000 (00:51 +0000)]
r4935: fixed a bug where "c->status = xxx_handler(x);" could write to c after
it is freed. The problem is that the handler might complete the
request, and called the c->async.fn() async handler. That handler
might free the request handle.
Tim Potter [Sat, 22 Jan 2005 08:09:10 +0000 (08:09 +0000)]
r4930: Update my copyright.
Andrew Tridgell [Sat, 22 Jan 2005 05:36:32 +0000 (05:36 +0000)]
r4927: parse the NBT session request in the smb server. This gets rid of that
annoying "not parsing session request" message on each SMB connection
Andrew Tridgell [Sat, 22 Jan 2005 02:51:39 +0000 (02:51 +0000)]
r4924: continue the effort to simplify and generalise the composite
interface. This patch removes the "stage" variable, which is really
better suited to the backend state structures
Andrew Bartlett [Sat, 22 Jan 2005 02:39:05 +0000 (02:39 +0000)]
r4923: Avoid using krb5-config in the path if a krb5 prefix was already set.
Andrew Bartlett
Andrew Tridgell [Sat, 22 Jan 2005 02:08:30 +0000 (02:08 +0000)]
r4922: fixed an infinite loop in the name resolve code when handling a method
in smb.conf that isn't implemented in the library
Andrew Tridgell [Sat, 22 Jan 2005 01:37:48 +0000 (01:37 +0000)]
r4919: if a caller doesn't provide an event context to the resolver library,
then create one. This fixes a crash in the RAW-NEGNOWAIT test for
'host' resolution.
Andrew Tridgell [Sat, 22 Jan 2005 00:52:54 +0000 (00:52 +0000)]
r4916: added "host" name resolution using fork() per gethostbyname()
comments welcome, but please think about the alternatives first :-)
Andrew Tridgell [Fri, 21 Jan 2005 23:53:10 +0000 (23:53 +0000)]
r4915: free temp context _before_ the async callback, as the async callback might destroy our top level context
leaving the tmp context freed (so a double free)
Andrew Bartlett [Fri, 21 Jan 2005 23:15:18 +0000 (23:15 +0000)]
r4914: Fill in the realm for the self-join.
Andrew Bartlett
Andrew Tridgell [Fri, 21 Jan 2005 22:01:57 +0000 (22:01 +0000)]
r4911: make sure we fill in the transport called name on port 445 as well
(thanks to abartlet for spotting this bug)
Andrew Tridgell [Fri, 21 Jan 2005 20:38:33 +0000 (20:38 +0000)]
r4909: fixed name_trn_id generation (thanks to metze for spotting the bug!)
Andrew Tridgell [Fri, 21 Jan 2005 13:39:33 +0000 (13:39 +0000)]
r4901: a bit more info on nbt packets under high debug level
Andrew Tridgell [Fri, 21 Jan 2005 13:29:24 +0000 (13:29 +0000)]
r4900: build fix for IRIX 6.5
Andrew Tridgell [Fri, 21 Jan 2005 13:16:33 +0000 (13:16 +0000)]
r4899: fixed build
Andrew Tridgell [Fri, 21 Jan 2005 13:13:24 +0000 (13:13 +0000)]
r4898: - removed the unused wins_srv_*() code
- expanded the generic async name resolver to try multiple methods
- added wins resolutions to the list of methods tried
- fixed up the random trn id generation to use the good random generator
Andrew Bartlett [Fri, 21 Jan 2005 12:45:11 +0000 (12:45 +0000)]
r4897: Unbreak the LDAP server. Somehow the generic service structures
(which seem just a little too complex) changed, but this code was not
updated or tested.
Also clarify the existing code, by not reusing variables.
Andrew Bartlett
Andrew Tridgell [Fri, 21 Jan 2005 11:41:48 +0000 (11:41 +0000)]
r4896: make sure the event context doesn't go away while waiting for event completion
Andrew Bartlett [Fri, 21 Jan 2005 11:31:35 +0000 (11:31 +0000)]
r4895: I missed this in my previous 'use secrets.ldb' commit.
Andrew Bartlett
Andrew Tridgell [Fri, 21 Jan 2005 11:25:33 +0000 (11:25 +0000)]
r4894: namecache.c is not used any more either
Andrew Bartlett [Fri, 21 Jan 2005 11:23:11 +0000 (11:23 +0000)]
r4893: Move to using secrets.ldb for the Kerberos verify, instead of
secrets.tdb from Samba3.
Andrew Bartlett
Andrew Tridgell [Fri, 21 Jan 2005 11:23:11 +0000 (11:23 +0000)]
r4892: we don't need nameserv.h any more
Andrew Tridgell [Fri, 21 Jan 2005 11:18:56 +0000 (11:18 +0000)]
r4891: - added a generic resolve_name() async interface in libcli/resolve/,
which will eventually try all resolution methods setup in smb.conf
- only resolution backend at the moment is bcast, which does a
parallel broadcast to all configured network interfaces, and takes
the first reply that comes in (this nicely demonstrates how to do
parallel requests using the async APIs)
- converted all the existing code to use the new resolve_name() api
- removed all the old nmb code (yay!)
Andrew Bartlett [Fri, 21 Jan 2005 11:10:03 +0000 (11:10 +0000)]
r4890: Try to cope with mechanism mismatch in the client speaks first version
of the SPNEGO state-machine. (Such as on LDAP and HTTP)
Andrew Bartlett
Andrew Tridgell [Fri, 21 Jan 2005 06:58:16 +0000 (06:58 +0000)]
r4889: make sure ndr print flags are initialised in ndrdump
Andrew Tridgell [Fri, 21 Jan 2005 06:56:57 +0000 (06:56 +0000)]
r4888: use the neater calling convention
Andrew Tridgell [Fri, 21 Jan 2005 06:56:13 +0000 (06:56 +0000)]
r4887: removed a bogus cast
Andrew Tridgell [Fri, 21 Jan 2005 06:55:33 +0000 (06:55 +0000)]
r4886: fixed two places where we process the send side of a socket after the
recv side in the same event. That's a bad idea, as the first callback
could decide to destroy the socket.
Andrew Tridgell [Fri, 21 Jan 2005 06:54:10 +0000 (06:54 +0000)]
r4885: added a new NBT client library. Features include:
- structures defined using IDL in nbt.idl
- build around our events structure, and talloc
- fully async
- supports all NBT packet fields as per rfc1002
- easy interfaces for name query and status
For the moment there are just a couple of test functions in
namequery.c, test_name_query() and test_name_status(). These will be
removed when we hook the new library into libcli/ fully
The new library will also be a fairly good basis for a nbt
server. Although it can't be a server as-is, I wrote it with the needs
of a server in mind (for example, extremely scalable idtree based
packet handling)
Andrew Tridgell [Fri, 21 Jan 2005 06:46:07 +0000 (06:46 +0000)]
r4884: - 2nd part of support ndr_size_ generation on unions as well as structures
- added "nopull" and "nopush" flags, to allow for externally written parsers for sub-structures
Andrew Tridgell [Fri, 21 Jan 2005 06:44:52 +0000 (06:44 +0000)]
r4883: support ndr_size_ generation on unions as well as structures
Tim Potter [Thu, 20 Jan 2005 05:57:05 +0000 (05:57 +0000)]
r4865: Fix up subtree name when dissecting unions. E.g in LSA, The subtree
used to be called something like "Level, R->out.info" but now is called
"Lsa PolicyInformation".
Andrew Tridgell [Thu, 20 Jan 2005 00:46:04 +0000 (00:46 +0000)]
r4863: schema_find_attribute() should be static
Simo Sorce [Wed, 19 Jan 2005 22:56:14 +0000 (22:56 +0000)]
r4862: - better structure schema tests
- fix check for deletion of required attributes on modify
removed by mistake with the previous commits
Simo Sorce [Wed, 19 Jan 2005 22:26:14 +0000 (22:26 +0000)]
r4859: changed a numbe rof things:
- Try to no tcopy memory if not necessary
- let the searches hang around until we finish the operation
- greatly simplify the modify operation
still todo:
- check attributes syntax
- check objectclasses type, we cannot allow to add a new structural
objectclass to an exixting entry unless the existing one is a parent
- more tests
Andrew Tridgell [Wed, 19 Jan 2005 21:38:27 +0000 (21:38 +0000)]
r4858: a better fix for alignment of enumerated types (I'm not even sure why
the last fix worked at all)
Simo Sorce [Wed, 19 Jan 2005 13:54:10 +0000 (13:54 +0000)]
r4844: - Remove the unused attrsyn structure
- Change 0 to NULL when checking allocations
- Introduce the schema_attr_cmp hepler function
- Do not allow auxiliary classes to be missing
- Try to ease code readability and try to get
main code out of loops when possibile.
Andrew Tridgell [Wed, 19 Jan 2005 11:58:40 +0000 (11:58 +0000)]
r4843: fixed the alignment handling of enumerated types
Tim Potter [Wed, 19 Jan 2005 11:24:29 +0000 (11:24 +0000)]
r4842: Set the end of a structure so that when you click on the proto tree,
the hex display window highlights the part of the buffer occupied by
the structure. Cool!
Tim Potter [Wed, 19 Jan 2005 10:17:46 +0000 (10:17 +0000)]
r4841: Handle levels that are signed integer types.
Andrew Tridgell [Wed, 19 Jan 2005 03:34:49 +0000 (03:34 +0000)]
r4833: added LOCAL-SOCKET to the list of tests that are expected to pass
Andrew Tridgell [Wed, 19 Jan 2005 03:33:49 +0000 (03:33 +0000)]
r4832: added simple testing of tcp sockets to LOCAL-SOCKET test
Andrew Tridgell [Wed, 19 Jan 2005 03:20:20 +0000 (03:20 +0000)]
r4831: added udp support to our generic sockets library.
I decided to incorporate the udp support into the socket_ipv4.c
backend (and later in socket_ipv6.c) rather than doing a separate
backend, as so much of the code is shareable. Basically this adds a
socket_sendto() and a socket_recvfrom() call and not much all.
For udp servers, I decided to keep the call as socket_listen(), even
though dgram servers don't actually call listen(). This keeps the API
consistent.
I also added a simple local sockets testsuite in smbtorture,
LOCAL-SOCKET
Andrew Tridgell [Tue, 18 Jan 2005 21:11:19 +0000 (21:11 +0000)]
r4828: don't apply the schema until we get it working properly
Andrew Tridgell [Tue, 18 Jan 2005 10:10:35 +0000 (10:10 +0000)]
r4819: its just not my day today ....
Andrew Tridgell [Tue, 18 Jan 2005 10:06:37 +0000 (10:06 +0000)]
r4818: missed version.h here
Andrew Tridgell [Tue, 18 Jan 2005 09:30:43 +0000 (09:30 +0000)]
r4817: ccache was being made ineffective on all the build farm machines
because the version number was being auto-updated and included in all
C files. With this change it is only included where needed.
Tim Potter [Tue, 18 Jan 2005 06:38:11 +0000 (06:38 +0000)]
r4816: Note that NTTIME is 4-byte aligned which makes the comment about alignment
for NTTIME_hyper make a bit more sense.
Tim Potter [Tue, 18 Jan 2005 05:32:35 +0000 (05:32 +0000)]
r4815: Handle uint8 and uint16 bitmap types.
Break out arguments to proto_register_protocol() so they can be renamed
easily. Unfortunately the "pidl_" prefix will have to stick around if/when
the old dissectors are removed because the filter and short names clash.
Andrew Tridgell [Tue, 18 Jan 2005 04:06:58 +0000 (04:06 +0000)]
r4814: cope with perl not being in /usr/bin for idl building
Andrew Tridgell [Tue, 18 Jan 2005 03:27:12 +0000 (03:27 +0000)]
r4813: this is a temporary solution to a link problem we have on some
platforms. The problem is that some loaders (for example the IRIX 6.5
machine us4 in the build farm) must have libs listed after object
files that depend on them on the link line. If you put the libs first
then all the symbols in the libs remain unresolved.
I think the correct fix for this is to separate xxx_LINK_LIBS out from
the current xxx_LINK_FLAGS, and use xxx_LINK_FLAGS followed by
xxx_LINK_LIST followed by xxx_LINK_LIBS. I'm hoping metze or jelmer,
as our build wizards, might take a look at this when they get time.
This interim fix should work fine, but its rather ugly, as it lists
the flags and libs twice in each link.
Andrew Tridgell [Tue, 18 Jan 2005 02:04:13 +0000 (02:04 +0000)]
r4812: removed dependence on Data::Dumper
if you need Dumper for debugging (and it is damn useful!) then please
use the require trick in MyDumper() from pidl so we don't end up
depending on it. Too many systems don't have it.
Andrew Tridgell [Mon, 17 Jan 2005 22:53:37 +0000 (22:53 +0000)]
r4811: now that the event context is at the socket level, the event cleanup
should be there too
Andrew Tridgell [Mon, 17 Jan 2005 22:53:00 +0000 (22:53 +0000)]
r4810: fixed anonymous connections with smbclient. Thanks to jbm for pointing this out.
Stefan Metzmacher [Mon, 17 Jan 2005 16:18:48 +0000 (16:18 +0000)]
r4806: - add some data to the configuration naming context
- add some stuff to make w2k3 dc join to get the correct
values
metze
Stefan Metzmacher [Mon, 17 Jan 2005 14:59:21 +0000 (14:59 +0000)]
r4804: add more emuns and bitmaps
metze
Andrew Tridgell [Mon, 17 Jan 2005 11:54:22 +0000 (11:54 +0000)]
r4801: remove the two bogus ctype.h includes
Andrew Tridgell [Mon, 17 Jan 2005 11:45:12 +0000 (11:45 +0000)]
r4800: proper fix for the _GNU_SOURCE problem with comparison_fn_t
Alexander Bokovoy [Mon, 17 Jan 2005 09:32:11 +0000 (09:32 +0000)]
r4799: comparison_fn_t is under __USE_GNU on GNU systems, therefore, we need _GNU_SOURCE defined in the test
Tim Potter [Mon, 17 Jan 2005 06:37:47 +0000 (06:37 +0000)]
r4798: When dissecting structures, name the protocol tree after the field name,
not the field type (i.e DACL and SACL instead of dom_sid).
Tim Potter [Mon, 17 Jan 2005 06:21:40 +0000 (06:21 +0000)]
r4797: Add the name of what we are pointing to when dissecting pointers.
Tim Potter [Mon, 17 Jan 2005 05:41:18 +0000 (05:41 +0000)]
r4796: Get rid of unecessary #include