Andrew Tridgell [Fri, 6 Aug 2004 02:32:48 +0000 (02:32 +0000)]
r1662: add a copy of the GPL text to the samba4 svn tree
Andrew Tridgell [Thu, 5 Aug 2004 05:03:04 +0000 (05:03 +0000)]
r1657: fixed the string types of two strings in svcctl.
we still need to fix the handling of subcontexts
Stefan Metzmacher [Wed, 4 Aug 2004 13:23:35 +0000 (13:23 +0000)]
r1654: rename cli_ -> smbcli_
rename CLI_ -> SMBCLI_
metze
Stefan Metzmacher [Wed, 4 Aug 2004 09:33:41 +0000 (09:33 +0000)]
r1649: do a join as dc and run this test as the machine account
metze
Stefan Metzmacher [Wed, 4 Aug 2004 09:30:24 +0000 (09:30 +0000)]
r1648: the endpoints for the DRSUAPI pipe are
lsass,protected_storage not drsuapi:-)
metze
Stefan Metzmacher [Wed, 4 Aug 2004 07:44:14 +0000 (07:44 +0000)]
r1647: destroy mem_ctx
metze
Andrew Tridgell [Wed, 4 Aug 2004 06:01:10 +0000 (06:01 +0000)]
r1646: disable testing of group rename in the RPC-SAMR test, as it leaves the
group in a state where it can't be deleted via samr, which breaks
repeated runs of the test.
Andrew Tridgell [Wed, 4 Aug 2004 06:00:14 +0000 (06:00 +0000)]
r1645: added a debug when a RPC fault is received
Andrew Tridgell [Wed, 4 Aug 2004 05:59:40 +0000 (05:59 +0000)]
r1644: changed the way [relative] pointers work in pidl, making them much
simpler. This doesn't appear to hurt any of the existing uses of
[relative], but fixes its use for svcctl that jelmer is working on.
Jelmer Vernooij [Tue, 3 Aug 2004 23:06:15 +0000 (23:06 +0000)]
r1642: Torture test for EnumServicesStatus, almost finished.
Jelmer Vernooij [Tue, 3 Aug 2004 23:05:43 +0000 (23:05 +0000)]
r1641: Finish IDL for EnumServicesStatus
Andrew Tridgell [Tue, 3 Aug 2004 12:30:17 +0000 (12:30 +0000)]
r1637: - w2k3 can't handle more than 1000 names in a LookupNames request
- use a SID that w2k3 likes in CreateTrustedDomain
Andrew Tridgell [Tue, 3 Aug 2004 08:04:51 +0000 (08:04 +0000)]
r1636: improved the negnowait test so it is standalone (taking advantage of the async APIs)
Andrew Tridgell [Tue, 3 Aug 2004 08:04:11 +0000 (08:04 +0000)]
r1635: when a transport dies, setup errors for all pending sends and recvs, plus disalllow any more sends
Andrew Tridgell [Tue, 3 Aug 2004 07:16:41 +0000 (07:16 +0000)]
r1634: to get signing right for async requests we must send requests in
order. Fixed the linked list add to always add to the end for outgoing
requests.
Andrew Tridgell [Tue, 3 Aug 2004 06:52:06 +0000 (06:52 +0000)]
r1633: fixed a couple of async oplock handling errors
Andrew Tridgell [Tue, 3 Aug 2004 06:50:58 +0000 (06:50 +0000)]
r1632: in case of error don't send uninitialised fnums on the wire
Andrew Tridgell [Tue, 3 Aug 2004 05:59:28 +0000 (05:59 +0000)]
r1631: don't use req->transport after req has been destroyed
Andrew Tridgell [Tue, 3 Aug 2004 05:47:24 +0000 (05:47 +0000)]
r1630: - fixed the replacement timegm() function to work correctly for DST changes
- got rid of global_myname(), using lp_netbios_name() instead
Andrew Tridgell [Tue, 3 Aug 2004 05:44:48 +0000 (05:44 +0000)]
r1629: server_zone needs to be "int" not "int16_t" as it can hold values
larger than 35536 (the max value is actually 12*3600 == 43200)
Andrew Tridgell [Tue, 3 Aug 2004 04:44:27 +0000 (04:44 +0000)]
r1628: - fixed the comment on run_deferopen. (I also think that the
deferopen test should probably be ripped out and redone, as it
tests the wrong thing.)
- removed some unused code
- add progress printing to the pipe_number test
- fixed the dir1 test to use deltree rather than trying to re-invent it itself (in a broken manner)
- renamed all of the tests into catagories, so you can now run
BASE-* for the base tests, and BENCH-* for the benchmarks.
Andrew Tridgell [Tue, 3 Aug 2004 04:15:05 +0000 (04:15 +0000)]
r1627: make sure we initialise write_time in the deprecated function cli_ctemp()
Andrew Tridgell [Mon, 2 Aug 2004 07:40:55 +0000 (07:40 +0000)]
r1619: - add support for older systems to cli_list*()
- make cli_list_new() use the last_name continue method rather than
the trans2 findnext continue flag, as the continue flag is broken on
win2003 (win2003 sometimes misses up to 1/3 of all files in a directory)
Andrew Tridgell [Mon, 2 Aug 2004 05:51:42 +0000 (05:51 +0000)]
r1618: fixed the receipt of multi-part replies to SMBtrans2
Jelmer Vernooij [Mon, 2 Aug 2004 00:24:04 +0000 (00:24 +0000)]
r1617: Start working on the svcctl pipe
Tim Potter [Fri, 30 Jul 2004 05:30:14 +0000 (05:30 +0000)]
r1611: Put a brace on a new line.
Andrew Tridgell [Thu, 29 Jul 2004 11:55:57 +0000 (11:55 +0000)]
r1606: make the low level socket read/write routines cope properly with non-blocking sockets
Andrew Bartlett [Thu, 29 Jul 2004 10:33:36 +0000 (10:33 +0000)]
r1605: GENSEC krb5 updates - fix a valgrind found uninitialised variable, and
allow tests for 'unwrapped' krb5, allowed by Win2k3.
SPENGO changes, trying to get the logic right (when and what
sub-mechanisms to wrap).
Andrew Bartlett
Andrew Bartlett [Thu, 29 Jul 2004 10:13:34 +0000 (10:13 +0000)]
r1604: Samba4 avoids memcpy() as much as possible - we don't need to make a copy here.
Andrew Bartlett
Andrew Tridgell [Thu, 29 Jul 2004 09:30:54 +0000 (09:30 +0000)]
r1603: fixed in.size to not overstate the packet size by 4 bytes
my apologies to abartlett for thinking this was his bug!
Andrew Tridgell [Thu, 29 Jul 2004 08:14:34 +0000 (08:14 +0000)]
r1602: make sure we honor the use_spnego flag
Jim McDonough [Wed, 28 Jul 2004 13:08:08 +0000 (13:08 +0000)]
r1596: Fix up copyrights.
Tim Potter [Tue, 27 Jul 2004 19:08:34 +0000 (19:08 +0000)]
r1595: Always use a loop variable named $e when iterating over lists of
elements.
Tim Potter [Tue, 27 Jul 2004 18:45:18 +0000 (18:45 +0000)]
r1594: Remove some array checks that we can't do because we don't store state
between scalar and buffer passes.
Fix array handling - most of the samr queryuserinfo calls now parse
properly!
Tim Potter [Tue, 27 Jul 2004 18:43:39 +0000 (18:43 +0000)]
r1593: Fix bug in get_subtree() where we were always looking at the head of
the list of subtrees.
Tim Potter [Tue, 27 Jul 2004 10:56:34 +0000 (10:56 +0000)]
r1592: Fix up subtrees for non-scalar types instead of hardcoding module subtree.
Arrays currently broken though. )-:
Tim Potter [Mon, 26 Jul 2004 04:11:18 +0000 (04:11 +0000)]
r1586: Finally, a technique for putting buffers in scalars in the same
protocol tree that actually works. Instead of trying really hard
to do this in code, just remember a list of strings and subtrees
for every tree. (This does require the addition of a void *user_data
field to ethereal's proto_node structure).
Stefan Metzmacher [Sun, 25 Jul 2004 15:09:48 +0000 (15:09 +0000)]
r1585: add dcpromo.pl start, this should be configure a
samba4 ads domain controller's ldb database
metze
Andrew Tridgell [Fri, 23 Jul 2004 06:40:49 +0000 (06:40 +0000)]
r1578: the first stage of the async client rewrite.
Up to now the client code has had an async API, and operated
asynchronously at the packet level, but was not truly async in that it
assumed that it could always write to the socket and when a partial
packet came in that it could block waiting for the rest of the packet.
This change makes the SMB client library full async, by adding a
separate outgoing packet queue, using non-blocking socket IO and
having a input buffer that can fill asynchonously until the full
packet has arrived.
The main complexity was in dealing with the events structure when
using the CIFS proxy backend. In that case the same events structure
needs to be used in both the client library and the main smbd server,
so that when the client library is waiting for a reply that the main
server keeps processing packets. This required some changes in the
events library code.
Next step is to make the generated rpc client code use these new
capabilities.
Stefan Metzmacher [Wed, 21 Jul 2004 09:57:39 +0000 (09:57 +0000)]
r1558: the unknown_time is the same as the logon_time in the PAC_LOGON_INFO
the account_name is a 'nstring'
metze
Stefan Metzmacher [Tue, 20 Jul 2004 20:30:29 +0000 (20:30 +0000)]
r1552: commit the first version of the pidl generated krb5 PAC parser
NOTE: there a lot of work todo, maybe we need to extent pidl
metze
Stefan Metzmacher [Mon, 19 Jul 2004 10:35:05 +0000 (10:35 +0000)]
r1547: rename 'enum <bla>_level' -> 'enum smb_<bla>_level'
e.g. we now have 'union smb_mkdir' and 'enum smb_mkdir_level' in sync
we may should also rename 'RAW_MKDIR_*' -> 'SMB_MKDIR_*'
metze
Stefan Metzmacher [Mon, 19 Jul 2004 10:15:33 +0000 (10:15 +0000)]
r1546: fix the build the function name has changed to srvsvc_NetRemoteTOD
because I implemented the idl for it
metze
Stefan Metzmacher [Mon, 19 Jul 2004 06:49:53 +0000 (06:49 +0000)]
r1545: add some comments and move a transport function to the transport section
metze
Stefan Metzmacher [Mon, 19 Jul 2004 06:41:34 +0000 (06:41 +0000)]
r1544: add some comments and fix 'return True' -> 'return ret'
metze
Stefan Metzmacher [Mon, 19 Jul 2004 06:14:00 +0000 (06:14 +0000)]
r1543: add srvsvc_NetRemoteTOD idl and torture tests
metze
Tim Potter [Sat, 17 Jul 2004 03:43:15 +0000 (03:43 +0000)]
r1540: - Remove return statements when a parse error is detected as ethereal
will do this for us.
- Fix a format string error
- Use subtrees for scalar and buffer elements
- Add subtree support for function request/responses
- Formatting fixups
Tim Potter [Fri, 16 Jul 2004 12:59:16 +0000 (12:59 +0000)]
r1535: Create a subtree for each array element.
Tim Potter [Fri, 16 Jul 2004 12:58:33 +0000 (12:58 +0000)]
r1534: Alternate method of grouping togethers buffers and scalars which actually
seems to work. We create a subtree one level higher instead of in the
parse function.
Stefan Metzmacher [Fri, 16 Jul 2004 09:51:54 +0000 (09:51 +0000)]
r1530: we don't need to call srvsvc_NetShareGetInfo twice
in NetShareEnum and NetShareEnumAll
metze
Stefan Metzmacher [Fri, 16 Jul 2004 09:36:17 +0000 (09:36 +0000)]
r1529: add NetSrvGetInfo torture test
metze
Stefan Metzmacher [Fri, 16 Jul 2004 09:25:52 +0000 (09:25 +0000)]
r1528: add NetCharDevQ torture tests
metze
Stefan Metzmacher [Fri, 16 Jul 2004 09:10:36 +0000 (09:10 +0000)]
r1527: add NetCharDev torture tests
metze
Stefan Metzmacher [Fri, 16 Jul 2004 08:56:11 +0000 (08:56 +0000)]
r1526: some cleanups and fixes for the srvsvc torture tests
metze
Stefan Metzmacher [Fri, 16 Jul 2004 08:53:20 +0000 (08:53 +0000)]
r1525: share info level 1004 is invalid,
but is not the same as the default case
metze
Stefan Metzmacher [Fri, 16 Jul 2004 07:20:15 +0000 (07:20 +0000)]
r1524: add some comments
metze
Andrew Tridgell [Fri, 16 Jul 2004 03:57:04 +0000 (03:57 +0000)]
r1523: declaring variables "in" and "out" in every C module is a bad idea!
I actually don't like this idea of sharing the request buffer
structure between client and server, and I'll see if I can convince
abartlet to revert it. It ties the two pieces of code far more
intimately than is justified, and will certainly lead to ugliness as
the client and server take different paths of development. I would far
prefer we just pass 2 arguments (pointer and length) to the signing
calls instead of 1 (the request).
Andrew Bartlett [Fri, 16 Jul 2004 03:10:48 +0000 (03:10 +0000)]
r1522: Oops - I forgot this file in the prvevious commit.
This breaks the request structure out into a common part between the
client and server, which the signing code now uses.
Andrew Bartlett
Andrew Bartlett [Fri, 16 Jul 2004 02:54:57 +0000 (02:54 +0000)]
r1521: Updates to our SMB signing code.
- This causes our client and server code to use the same core code,
with the same debugs etc.
- In turn, this will allow the 'mandetory/fallback' signing algorithms
to be shared, and only written once.
Updates to the SPNEGO code
- Don't wrap an empty token to the server, if we are actually already finished.
Andrew Bartlett
Andrew Tridgell [Thu, 15 Jul 2004 10:16:40 +0000 (10:16 +0000)]
r1520: only call write handler or read handler, not both. This copes with the
connection being removed between calls.
Andrew Tridgell [Thu, 15 Jul 2004 10:08:56 +0000 (10:08 +0000)]
r1519: show the ldb_errstring() value in the log for failed ldb calls
Andrew Tridgell [Thu, 15 Jul 2004 10:07:45 +0000 (10:07 +0000)]
r1518: check for ldb_search giving -1 (indicating db corruption)
Andrew Tridgell [Thu, 15 Jul 2004 10:07:05 +0000 (10:07 +0000)]
r1517: change event_remove_timed() to remove by structure not by handler.
this fixes a crash bug in smbd with multiple RPC clients
Stefan Metzmacher [Thu, 15 Jul 2004 09:43:32 +0000 (09:43 +0000)]
r1516: remove the server_connection from the list on the server_socket
and call talloc_destroy(srv_conn->mem_ctx)
also don't follow NULL pointers
metze
Stefan Metzmacher [Thu, 15 Jul 2004 08:59:07 +0000 (08:59 +0000)]
r1515: move dublicate code to a function
metze
Stefan Metzmacher [Thu, 15 Jul 2004 08:28:19 +0000 (08:28 +0000)]
r1514: close stuff from the server_connection not in the
close_connection fn of a specific service
metze
Stefan Metzmacher [Thu, 15 Jul 2004 08:26:41 +0000 (08:26 +0000)]
r1513: change DEBUG level to 1
metze
Andrew Tridgell [Thu, 15 Jul 2004 07:44:46 +0000 (07:44 +0000)]
r1512: fixed a bug where we could reference the timer event handler after destruction
Andrew Tridgell [Thu, 15 Jul 2004 07:16:15 +0000 (07:16 +0000)]
r1511: fixed a free() that should be ldb_free()
this might explain the tdb corruption that metze found - it caused heap corruption that affected tdb
Andrew Tridgell [Thu, 15 Jul 2004 05:13:08 +0000 (05:13 +0000)]
r1510: add a commented out routine I used to test password change on NT3.
Andrew Tridgell [Thu, 15 Jul 2004 05:11:27 +0000 (05:11 +0000)]
r1509: in order to interoperate with NT3.1 we need to ignore extra data at the end of RPC PDUs.
It turns out that NT3.1 adds junk onto the end of every PDU.
Andrew Tridgell [Thu, 15 Jul 2004 05:04:17 +0000 (05:04 +0000)]
r1508: simple fix for broken server side signing. This may need more work for
SPNEGO, but I'll leave andrew to tackle that.
Andrew Tridgell [Thu, 15 Jul 2004 02:11:03 +0000 (02:11 +0000)]
r1507: fixed the handling of SMB chaining with the new server structure. You
must think carefully about packet chaining when dealing with any
authentication or SMB parsing issues. The particular problem here was
that a chained tconX didn't get the req->session setup after an
initial sesstion setup call, so the tconx used a bogus VUID.
Stefan Metzmacher [Wed, 14 Jul 2004 12:44:31 +0000 (12:44 +0000)]
r1499: combine struct user_struct and struct smbsrv_user
to a struct smbsrv_session that the same as cli_session for the client
we need a gensec_security pointer there
(spnego support will follow)
prefix some related functions with smbsrv_
metze
Andrew Bartlett [Wed, 14 Jul 2004 12:14:07 +0000 (12:14 +0000)]
r1498: (merge from 3.0)
Rework our random number generation system.
On systems with /dev/urandom, this avoids a change to secrets.tdb for every fork().
For other systems, we now only re-seed after a fork, and on startup.
No need to do it per-operation. This removes the 'need_reseed'
parameter from generate_random_buffer().
This also requires that we start the secrets subsystem, as that is
where the reseed value is stored, for systems without /dev/urandom.
In order to aviod identical streams in forked children, the random
state is re-initialised after the fork(), at the same point were we do
that to the tdbs.
Andrew Bartlett
Stefan Metzmacher [Wed, 14 Jul 2004 11:28:34 +0000 (11:28 +0000)]
r1497: add forward declaration for struct auth_session_info
(fix compiler warning)
metze
Stefan Metzmacher [Wed, 14 Jul 2004 09:00:28 +0000 (09:00 +0000)]
r1494: fix debug message
metze
Stefan Metzmacher [Tue, 13 Jul 2004 21:04:56 +0000 (21:04 +0000)]
r1486: commit the start of the generic server infastructure
the idea is to have services as modules (smb, dcerpc, swat, ...)
the process_model don't know about the service it self anymore.
TODO:
- the smbsrv should use the smbsrv_send function
- the service subsystem init should be done like for other modules
- we need to have a generic socket subsystem, which handle stream, datagram,
and virtuell other sockets( e.g. for the ntvfs_ipc module to connect to the dcerpc server
, or for smb or dcerpc or whatever to connect to a server wide auth service)
- and other fixes...
NOTE: process model pthread seems to be broken( but also before this patch!)
metze
Stefan Metzmacher [Tue, 13 Jul 2004 18:41:08 +0000 (18:41 +0000)]
r1483: build dynconfig.c also with PICFLAGS
metze
Stefan Metzmacher [Tue, 13 Jul 2004 18:10:11 +0000 (18:10 +0000)]
r1482: today I saw DCERPC_AUTH_TYPE 16(0x10)
and it seems to be raw krb5, but I need to do some tests
metze
Stefan Metzmacher [Tue, 13 Jul 2004 18:05:02 +0000 (18:05 +0000)]
r1481: add idl file and torture test dummies
for DRSUapi (the Active Directory Replication Protocol)
I'll try to fill the idl file as part of a study project
together with some other students...
metze
Stefan Metzmacher [Tue, 13 Jul 2004 17:52:29 +0000 (17:52 +0000)]
r1480: gwsam has unresolved symbols in it
(on my SuSE 9.1)
so I disable it for now
metze
Stefan Metzmacher [Tue, 13 Jul 2004 17:40:28 +0000 (17:40 +0000)]
r1479: print out domain too
Andrew Bartlett [Tue, 13 Jul 2004 06:39:55 +0000 (06:39 +0000)]
r1476: Don't print messages about the CCACHE not being found - this is normal.
Andrew Bartlett
Andrew Bartlett [Tue, 13 Jul 2004 05:14:59 +0000 (05:14 +0000)]
r1475: More kerberos work
- We can now connect to hosts that follow the SPNEGO RFC, and *do not*
give us their principal name in the mechListMIC.
- The client code now remembers the hostname it connects to
- We now kinit for a user, if there is not valid ticket already
- Re-introduce clock skew compensation
TODO:
- See if the username in the ccache matches the username specified
- Use a private ccache, rather then the global one, for a 'new' kinit
- Determine 'default' usernames.
- The default for Krb5 is the one in the ccache, then $USER
- For NTLMSSP, it's just $USER
Andrew Bartlett
Andrew Bartlett [Tue, 13 Jul 2004 03:23:55 +0000 (03:23 +0000)]
r1474: It is useful if talloc_strdup() behaves like strdup()
- NULL in, NULL out
Andrew Bartlett
Volker Lendecke [Mon, 12 Jul 2004 16:35:48 +0000 (16:35 +0000)]
r1470: Get the smb_trans2 structure out of the rap_cli_call struct.
Initial attempt at RAP server infrastructure. Look at rap_server.c for the
dummy functions that are supposed to implement the core functionality.
ipc_rap.c contains all the data shuffling. _rap_shareenum and _rap_serverenum2
in ipc_rap.c are (I think) regular enough to be auto-generated.
I did not test all the corner cases yet, but nevertheless I would like some
comments on the general style.
Volker
P.S: samba-3 smbclient now doesn't freak out anymore, although the results are
not entirely correct :-)
Stefan Metzmacher [Mon, 12 Jul 2004 15:34:34 +0000 (15:34 +0000)]
r1469: fix a segfault and compiler warning,
introduced by the "compiler warning fix" in rev 1460...
metze
Stefan Metzmacher [Mon, 12 Jul 2004 13:53:35 +0000 (13:53 +0000)]
r1467: disable gensec_krb5 by default till abartlet
add the kinit code
metze
Andrew Tridgell [Mon, 12 Jul 2004 13:23:01 +0000 (13:23 +0000)]
r1466: the name "oid" is taken by some silly system headers - avoid it in our code
Andrew Tridgell [Mon, 12 Jul 2004 13:22:26 +0000 (13:22 +0000)]
r1465: always do a full C prototype, even if its only (void).
this declaration:
int foo();
is *not* the same as this one:
int foo(void);
the first means "I don't know what arguments it takes". The second
means "it takes no arguments"
Andrew Tridgell [Mon, 12 Jul 2004 13:20:50 +0000 (13:20 +0000)]
r1464: the recent build changes completely lost the speed advantage of using
PCH (in fact, it meant that PCH was a slowdown, not a speedup). To
gain speed with PCH you must ensure that the .gch file is compiled
with _exactly_ the same options as the normal object files.
this fixes the .gch build options
Stefan Metzmacher [Mon, 12 Jul 2004 13:15:53 +0000 (13:15 +0000)]
r1463: fix the krb5 build
metze
Andrew Bartlett [Mon, 12 Jul 2004 09:11:13 +0000 (09:11 +0000)]
r1462: GENSEC Kerberos and SPENGO work:
- Spelling - it's SPNEGO, not SPENGO
- SMB signing - Krb5 logins are now correctly signed
- SPNEGO - Changes to always tell GENSEC about incoming packets, empty or not.
Andrew Bartlett
Andrew Bartlett [Mon, 12 Jul 2004 09:07:10 +0000 (09:07 +0000)]
r1461: ntlm_check.c is a server-side peice of code, so it belongs in AUTH.
Andrew Bartlett
Andrew Bartlett [Mon, 12 Jul 2004 09:02:09 +0000 (09:02 +0000)]
r1460: Avoid a compile warning.
Andrew Bartlett
Andrew Bartlett [Mon, 12 Jul 2004 05:50:29 +0000 (05:50 +0000)]
r1458: Add a new configure option, to make it possible to both find errors,
and compile with gtk.
The --enable-developer option was just too noisy with buggy GTK headers.
Andrew Bartlett
Andrew Bartlett [Mon, 12 Jul 2004 04:26:50 +0000 (04:26 +0000)]
r1457: Add the GSSAPI layer to our gensec_krb5 code.
Andrew Bartlett
Andrew Bartlett [Mon, 12 Jul 2004 02:40:38 +0000 (02:40 +0000)]
r1456: Rename this parameter to avoid shadowing a badly-named GTK global.
Andrew Bartlett
Jelmer Vernooij [Sun, 11 Jul 2004 20:16:02 +0000 (20:16 +0000)]
r1455: More Gtk+ updates:
- Start working on 'gwsam'
- Add GtkSelectDomainDialog and GtkSelectHostDialog