1
0
mirror of https://github.com/samba-team/samba.git synced 2025-01-29 21:47:30 +03:00

174 Commits

Author SHA1 Message Date
Jeremy Allison
ddfbcc0581 include/smb.h: Removed dir_ptr. Not needed.
locking/locking.c: First cut a blocking lock code. #ifdef'ed out for now.
locking/locking_shm.c: Removed dir_ptr. Not needed.
smbd/nttrans.c: More work on ChangeNotify - return is not an error and needs
                to be handled as a nttrans with zero params. Removed dir_ptr. Not needed.
smbd/reply.c:
smbd/server.c:
smbd/trans2.c: Removed dir_ptr. Not needed.

Hmmm. At the moment smbclient is broken - doesn't issue prompt correctly.
This needs looking at.

Jeremy.
-
Christopher R. Hertel
53b433e6ab I've come up with a different scheme for doing the DEBUG() and DEBUGADD()
macros that meets the RVALUE requirement and doesn't use the (a?b:c) format
that Andrew called "ugly".  I've added the new macros but kept the old
macros within a #if 0..#else..#endif block in case I've missed somthing.

Basically, I've used

  (void)( (a) && (b) )

or

  (void)( (a) && (b) && (c) )

instead of

  ( (a) ? (void)(b) : (void)(c) )

or similar.  I have this compiled and running now.

Here's the diff:

$ cvs diff smb.h
Enter passphrase for RSA key 'crh@Ruby':
Index: smb.h
===================================================================
RCS file: /data/cvs/samba/source/include/smb.h,v
retrieving revision 1.172
diff -r1.172 smb.h
143a144,145
> #if 0
>
150a153,164
>
> #else
>
> #define DEBUG( level, body ) \
>   (void)( (DEBUGLEVEL >= (level)) \
>        && (dbghdr( level, FILE_MACRO, FUNCTION_MACRO, (__LINE__) )) \
>        && (dbgtext body) )
>
> #define DEBUGADD( level, body ) \
>   (void)( (DEBUGLEVEL >= (level)) && (dbgtext body) )
>
> #endif

Chris -)-----
-
Christopher R. Hertel
efa67d6b2b Removed a stray semicolon in the DEBUGADD() macro. -
Jeremy Allison
9d010e721c Makefile.in: Fixed make clean bug with CVS directory in bin/.
include/smb.h: Fixed bugs in Debug macro's with SGI compiler.
smbd/trans2.c: Fixed bug reported by Zoltan Palmai <ZSPA@chevron.com>
               PR#8903 where get_lanman2_dir_entry() could return a
               mangled name as a resume key and call_trans2findnext()
               wasn't doing the same.
Jeremy.
-
Andrew Tridgell
ab912448c1 removed the if statements from the DEBUG() macro definitions.
Chris, you should never put if statements in macros, use the ugly
(a?b:c) form instead, otherwise you can produce incorrect code when
you have things like:

if (foo)
   DEBUG((blah));
else
   DEBUG((blooh));
-
Jeremy Allison
a0748c3f53 First implementation of ChangeNotify - this version only checks
for changes in the directory modify timestamps. A better version
will look at the requested client flags, and create a hash that
represents the current state of the directory, and check against
this instead.
debug.c: Added lp_timestamp_logs() function.
loadparm.c: Added "change notify timeout" in seconds (default 60)
- this is the scan rate for a directory.
            Added ""timestamp logs" boolean - default True. Turns
off log timestamps (so I can read them :-).
nttrans.c: ChangeNotify implementation.
server.c: ChangeNotify implementation.
shmem_sysv.c: Added exits on shmem errors (without them smbd can
              core dump if some calls fail).
smb.h: Added ChangeNotify flags for future use.
util.c: Tidied up typedef.
Jeremy.
-
Christopher R. Hertel
653c17c1b8 This is the checkin of the debug changes.
Makefile.in:  I've added debug.o.
  proto.h    :  Rebuilt, as is standard for these sorts of things.
  smb.h      :  New macros, etc.
  util.c     :  Debug code removed.

I'll check in debug.c in the next step.

Chris -)-----
-
Jeremy Allison
3a5eea850b Makefile.in: Moved UBIQX stuff into UTILOBJ.
loadparm.c: Added "ole locking compatibility" option (default "true").
locking.c: Changes to implement union in files_struct.
locking_shm.c: Changes to implement union in files_struct.
nttrans.c: Made opening a directory explicit (we have to).
           Added create directory code for nttrans.
reply.c: Changes to implement union in files_struct.
server.c: Changes to implement union in files_struct. Added create directory code.
trans2.c: Changes to implement union in files_struct.
smb.h: Changes to implement union in files_struct.
util.c: Changed linked list code to UNIQX linked list. This will make
        the other lists I need to implement for ChangeNotify and blocking
        locks easier.
Jeremy.
-
Andrew Tridgell
8b25fe7341 get rid of the runtime test for broken getgroups() and add a compile
time test instead. This also allows us to get rid of the igroups
element of a couple of structures.
-
Andrew Tridgell
3bda7ac417 merge from the autoconf2 branch to the main branch -
Jeremy Allison
ef1af7fe6d chgpasswd.c: Fixed up debug calls to stop crashes if ptsname failed.
local.h: Kept FSTYPE_STRING as Samba for now.
nmbd_browsesync.c: Added bugfix from Matt Chapman mattyc@cyberdude.com
                   - lmb_browserlist is now a struct ubi_dlList not a
                   struct browse_cache_record *.
server.c:
smb.h:
uid.c:
password.c: Removed attrs code - it is not used anywhere.
Jeremy
-
Jeremy Allison
339b102222 Fixed bug found by Richard Sharpe. After increasing files_struct size by
MAX_OPEN_DIRECTORIES for nttrans I forgot to update the code that
enumerates the array.
Created new MAX_FNUMS in local.h, changed all code that iterates
through the files_struct array to use this.
(sorry Richard).
Jeremy.
-
Jeremy Allison
e01ce693f4 Code added to fix the renaming of a directory under NT SMB calls.
local.h: Changed MAXDIR to MAX_OPEN_DIRECTORIES - shmem size also tuned by this.
dir.c: Use MAX_OPEN_DIRECTORIES.
nttrans.c: Allow opening of a directory to succeed. Doesn't actually open
           a file descriptor but takes a files_struct slot marked as an
           fd.
reply.c: Changed to close any outstanding is_directory files.
         reply_close changed to understand directory files.
server.c: Added open_directory(), close_directory() calls.
smb.h: Added is_directory to files_struct.
       Changed OPEN_FNUM to check that target is !is_directory (this
       prevents the normal file calls from processing a directory
       files_struct.
Jeremy.
-
Jeremy Allison
3e5cada988 Makefile: Added nttrans.o
includes.h: Added termios.h for AIX.
nttrans.c: Working NT SMB calls !
pipes.c: Use strequal instead of strcmp.
server.c: Use #defines rather than numbers.
smb.h: Updated NT SMB #defines.
Jeremy.
-
Jeremy Allison
b398f7daf5 nntrans.c: Fully implemented transact rename.
reply.c: Added NT specific rename if exists flag to rename_internals().
smb.h: Added NT rename flag.
Jeremy.
-
Jeremy Allison
556254d725 NT SMB trans reply code.
Jeremy.
-
Jeremy Allison
ce2503fddd includes.h: AIX fix.
nttrans.c: More NT SMB work.
smb.h: More NT SMB defines.
trans2.c: Change call response as I now have docs
          on what the flags mean. #ifdef it with JRATEST
          until I'm sure it's ok though.
Jeremy.
-
Jeremy Allison
e53f4396ea nttrans.c: More code towards NT protocol.
smb.h: More code towards NT protocol.
time.c: Fix for sco bug.
Jeremy.
-
Richard Sharpe
dc9436bae4 Remove the copyright to Canon Information Systems Australia, as we don't
want them to have the copyright.

Added a new DOSERR response code that Win95 returns, unimp, unimplemented.

Added code to ignore errors on setting remote time, as Win 95 does not like
the time being changed on a directory.  Win NT and Samba are OK at this.

This is the next to last clean-ups here. Next is to properly handle restore
times on directories (except for Win95--see above).

Now have Jay's changes in and have fixed a bug reported by Tim Lee.
-
Jeremy Allison
31e768369f Added SSL support from Christian Starkjohann <cs@obdev.at>
This patch may not yet compile with -DUSE_SSL enabled, further
Makefile changes may be needed. But it was important to get
this code in place before I go off to USENIX.
Jeremy.
-
Christopher R. Hertel
0b35065d0c This is the first, small step toward some tweaks that Jeremy and I have
been discussing regarding the debug mechanism.  With this, I've added a
macro that allows syntax like:

  if( DEBUGLVL( 3 ) )
    {
    Debug1( "message" );
    Debug2( "message" );
    }

The next step is to incorporate timestamps.

Chris -)-----
-
Luke Leighton
2cb94eec59 had what would be a compile error if it wasn't #if 0'd out in passdb_ops -
Luke Leighton
04b8240072 added "domain admin group" and "domain guest group" parameters. this is
because "domain admin users" and "domain guest users" was overloaded.

incorrectly.
-
Jeremy Allison
05c85df3c7 ipc.c: Fix bug where we don't return the correct error code when client
gives a too-small buffer for share info. Fix from Gil Kloepfer <gil@arlut.utexas.edu>
smb.h:
server.c: Fix for a nastly little security problem with multi-user Windows NT
          servers and Samba where the contents of the open-file cache can end
          up being served out to users who shouldn't have access. This is some
          *seriously* ugly code.
Jeremy.
-
Jeremy Allison
42a6551106 smb.h: More NT SMB stuff (just defines).
smbpass.c: Do *NOT* make this function static. It breaks the compile on gcc
util.c: Getting closer to MS wildcard semantics. A trailing '*' matches
        any trailing dot-separated components.
trans2.c: Removed hacks that change multiple '?' -> '*' as this
          breaks things now.
trans2.h: Removed NT_FILE_ATTRIBUTE_NORMAL - now FILE_ATTRIBUTE_NORMAL is defined
          in smb.h.
Jeremy.
-
Jeremy Allison
26c0176e10 Added more NT definitions, getting ready for nttrans code.
Jeremy.
-
Jeremy Allison
cd244b45a5 passdb.c: Fixed typo in coment.
smb.h: Removed comments no longer valid.
smbpass.c: Stopped dummy function from being prototyped.
util.c: Fix for multibyte char problems with strlower, strupper
and string_replace.
Jeremy.
-
Luke Leighton
d906ac5941 removed "default" system for password database api: all functions must
be supported.  there are some stub routines in passdb.c which can be
copied into a password database api which do conversion.  the module
writer can choose which of these to provide full support for instead
of using the conversion routines.
-
Luke Leighton
0193dd21c3 ldap.c :
- added support for some of the new passdb_ops functions.

	- removed functions that are supported "indirectly" through passdb.c

nisppass.c :

	- modified make_nisname_from_xxx() functions to take a "file" arg.

	- turned getnisp21pwuid() into getnisp21pwrid().  getnisp21pwuid()
	  functionality is available through "indirect" support in passdb.c

	- removed functions that are supported "indirectly" through passdb.c

	- added support for some of the new passdb_ops functions.

passdb.c :

	- created getsam21pwrid() function to go alongside getsam21pwuid.
	  it is not expected that getsam21pwuid ever be used, certainly
	  not from the lib/rpc code.

	- created getsamdisprid() and getsamdispent().  these are primarily
	  for support of SamrQueryDisplayInfo, however given that they
	  [struct sam_disp_info] return username, rid and fullname, there may
	  be further instances where these functions will be useful.

	- added support where either the get/add/mod-smb or get/add/mod-sam21
	  functions are optional.  this can be done very easily by checking
	  whether the struct passdb_ops table functions are NULL or not.

	  documented this capability in the notes at the top of the module.

	- where unix uid was referenced, use uid_t.

	- where unix gid was referenced, use gid_t.

smb.h :

	- added sam_disp_info functions to passdb_ops.

	- added getsam21pwrid() function.

smbpass.c :

	- added reference to iterate_getsam21pwrid().

lib/rpc/server/srv_samr.c :

	- removed group rid code added to get_user_info_21() code: this
	  had been added in the wrong place.  the client / server should
	  already know whether it wants to do a lookup by user rid or
	  by group rid.

	  the test of whether the rid is a user or group rid has been left
	  in because this may become useful consistency-check code.

	- converted back to getsam21pwrid() not
	  getsam21pwuid(pdb_user_rid_to_uid()).

	  this is because the unix uid to user rid mapping can be non-monotonic
	  in some password database systems, and monotonic in others.  imposing
	  the restriction by converting immediately from rid to uid at this
	  point is inadviseable, and will place this potential restriction on
	  _all_ password database systems, not just some which, for whatever
	  reason, do not support user rids.

	  it should be up to the individual password database writer to
	  convert from user rid to unix uid, should that module not support
	  rids.

lib/rpc/server/srv_util.c :

	- got lookup_user_name() to call getsamdisprid() not getsmbpwuid().
	  a bug was introduced (or at least the bug already there was not
	  fixed) whereby the nt user rid was converted to a unix uid, and
	  then not used.
-
Jeremy Allison
27ca536ad9 Abstracted all the crappy password interfaces through an indirect
function table, selectable at compile time. This should make the
code that implements all the password functions much cleaner, as
it's now very clear exactly what a particular password database
needs to provide to Samba.
Jeremy.
-
Jeremy Allison
7d332b2493 chgpasswd.c: Changed back to getsmb... from getsam...
ldap.c: Stoped dummy_function being prototyped.
loadparm.c: Fixed slprintf sizes.
nisppass.c: Fixed safe_strcpy sizes.
nmbd_processlogon.c: Changed back to getsmb... from getsam...
nttrans.c: Just a dump of new code.
passdb.c: Moved stuff around a lot - stopped any lookups by rid. This
          needs to be indirected through a function table (soon).
password.c: Changed back to getsmb... from getsam...
reply.c: Changed back to getsmb... from getsam...
slprintf.c: Fixed prototype problems.
smb.h: Fixed prototype problems.
smbpass.c: Changed to getsmbfile....
smbpasswd.c: Changed back to getsmb... from getsam...
lib/rpc/server/srv_netlog.c: Changed back to getsmb... from getsam...
lib/rpc/server/srv_samr.c: Fixed rid lookup - use uid or gid lookup.
lib/rpc/server/srv_util.c: Changed back to getsmb... from getsam...
Jeremy.
-
Luke Leighton
f215d375f0 Makefile:
- added nisppass.c and NISPLUS_FLAGS

includes.h:

	- renamed USE_LDAP to USE_LDAP_DB.  renamed NISPLUS to USE_NISPLUS_DB.
	added default define of USE_SMBPASS_DB.

	- removed ldap headers: they are local only to ldap.c

ldap.c :

	- made all ldap-specific functions static.

	- added dummy sam21 functions

loadparm.c :

	- renamed NISPLUS to NISPLUS_HOME


mkproto.awk

	- commented out ldap-specific #ifdef generation code: it's not
	  needed now that ldap-specific functions in ldap.c are static

nisppass.c :

	- first attempt at an add function from
-
Andrew Tridgell
9907296225 we don't need a typedef for string -
Jeremy Allison
2d77445400 This is a security audit change of the main source.
It removed all ocurrences of the following functions :

sprintf
strcpy
strcat

The replacements are slprintf, safe_strcpy and safe_strcat.

It should not be possible to use code in Samba that uses
sprintf, strcpy or strcat, only the safe_equivalents.

Once Andrew has fixed the slprintf implementation then
this code will be moved back to the 1.9.18 code stream.

Jeremy.
-
Luke Leighton
d42dd371fa added sam_passwd structure: don't want smb_passwd to be "polluted". -
Jeremy Allison
61c1dbb978 clitar.c: #ifdef'ed out all the bits that were giving 'defined but not used'
messages.
nttrans.c: More updates.
smb.h: Removed stuff that didn't belong in the smb_passwd struct. Persuaded Luke
       to use a new structure.
web/swat.c: Fixed gcc complaints about shadowing global 'string'.
Jeremy.
-
Luke Leighton
bcafdcda85 added smb_grpid to smb_passwd struct -
Luke Leighton
0b6394c83e added extra fields to struct smb_passwd
the smbpw functions are expected to fill in the blanks by reading the
appropriate smb.conf parameters.

the ldappw functions are expected to fill in the fields from the ldap
database.  a separate utility can be written to fill in the ldap database
fields from the smb.conf parameters, at a later date.
-
Jeremy Allison
94e70edef9 smb.h: Supporting defines for NT trans calls.
trans2.c: Paranoia bugfixes added when studying nttrans.c.
Jeremy.
-
Jeremy Allison
b5187ad6a3 Fixes for the %U and %G problems people have reported.
Essentially, multiple session_setup_and_X's may be done
to an smbd. As there is only one global variable containing
the requested connection name (sessionsetup_user), then any
subsequent sessionsetups overwrite this name (causing %U
and %G to get the wrong name). This is particularly common
when an NT client does a null session setup to get a
browse list after the user has connected, but before
a share has been mounted.

These changes store the requested_name in the vuid structure
(so this only really works for user level and above security)
and copies this name back into the global variable before
the standard_sub call.

Jeremy.
-
Jeremy Allison
c6c22df201 clientgen.c: Fixed null session setup bug.
password.c: Stopped cli_nt_logout call (we don't have it correct yet).
            Added Luke object-orientation fix :-).
smb.h: Added clnt_name_slash to cli_state.
lib/rpc/client/cli_login.c: Changed global_myname to clnt_name_slash where needed.
lib/rpc/client/cli_netlogon.c: Fixed debug messages, don't check creds on error.
lib/rpc/client/cli_pipe.c: Fixed debug messages, Added Luke object-orientation fix.
lib/rpc/parse/parse_misc.c: Fixed STRING2 linearization bug that was adding 1.
Jeremy.
-
Jeremy Allison
17b94a7084 This is the checkin that adds the security=domain functionality.
WARNING - so far this has only been tested against a Samba PDC
(still waiting for IS to add me the machine accounts :-).

Still missing is the code in smbpasswd that will add a machine
account password and change it on the domain controller, but
this is not hard, and I will check it in soon.

Jeremy.
-
Jeremy Allison
866406bfe3 This looks like a big change but really isn't.
It is changing the global variables "myname" and "myworkgroup"
to "global_myname" and "global_myworkgroup" respectively.

This is to make it very explicit when we are messing
with a global (don't ask - it makes the domain client
code much clearer :-).

Jeremy.
-
Jeremy Allison
899a9f0dce Makefile: Added nterr.c into the mix.
clientgen.c: Added nt_error as an entry in the struct client_state.
password.c: Open the netlogon pipe.
smb.h: Added nt_error as an entry in the struct client_state.
lib/rpc/parse/parse_net.c: Added comments on net logon.
lib/rpc/server/srv_netlog.c: Added comments on net logon.
Jeremy.
-
Jeremy Allison
e6bda112eb genrand.c: Improved generation of random values, more secure.
loadparm.c: Started add of 'security=domain' code.
password.c: Fix for security=server NT bugs.
reply.c: Started add of 'security=domain' code.
server.c: Started add of 'security=domain' code.
smb.h: Started add of 'security=domain' code.
Jeremy.
-
Jeremy Allison
eb4fe9ecdf ipc.c: Fix for printer queue spinning with Win95.
nmbd.c: Fix for always overwriting log despite append setting.
smb.h: Addition of last time password changed entry to account info.
smbpass.c: Changes to support last time changed field in smbpasswd file.
smbpasswd.c: Changes to support last time changed field in smbpasswd file.
util.c: Fix for always overwriting log despite append setting.
Jeremy.
-
Jeremy Allison
cd2613c572 Changes to allow Samba to be compiled with -Wstrict-prototypes
with gcc. (Not a big change although it looks like it :-).

Jeremy.
-
Jeremy Allison
fc2295e0f5 Makefile, loadparm.c, server.c, smb.h, util.c: Patch from
stn@techfak.uni-kiel.de (Stefan Nehlsen) to get homes from
the NIS+ map.
smbpasswd.c: Tidy up of cli_state structure.
Jeremy.
-
Jeremy Allison
b6c965c396 Added codepage 936 (simplified Chineses).
In doing so I realized that much code was being
duplicated between Hangul, Big5 and Simplified
Chinese - so I re-arranged kanji.[ch] to go
through generic functions for all multibyte
characters that can be identified by a single
code range (not Kanji - but all the others).

Jeremy.
-
Jeremy Allison
f9dacd1d8b Patch from Chris Maltby <chris@softway.com.au>. His comments follow:
+ improvement to smbtar to allow exclusion/inclusion of system and
    hidden files, and to generate a listing of what has been archived
    in a format useful for automated backup systems.

  + add the "Softq" spooling system to samba's printing capabilities.

  + I have "fixed" the intrusion of US style dates into samba reporting
    as well. The format yyyy/mm/dd is not only uunambiguous, but also
    has the benefit of making lexicographic sorts work correctly.

Jeremy.
-