22 years agoBumped version number in README to avoid confusion with SAMBA_2_0 branch.
Tim Potter [Mon, 7 Jun 1999 23:28:39 +0000 (23:28 +0000)]
Bumped version number in README to avoid confusion with SAMBA_2_0 branch.

22 years agoFixing core dump bug with unix password sync, caused by a NULL
Matthew Chapman [Wed, 2 Jun 1999 04:11:50 +0000 (04:11 +0000)]
Fixing core dump bug with unix password sync, caused by a NULL
connection_struct in a call to OpenDir.
JF, you fixed a similar bug in printing/nt_printing.c, I think your fix
is incorrect as global configuration files should not go through a VFS.

22 years agoSome more BDC-related fixes, mainly to the NET_SAM_SYNC RPC with respect
Matthew Chapman [Wed, 2 Jun 1999 03:19:20 +0000 (03:19 +0000)]
Some more BDC-related fixes, mainly to the NET_SAM_SYNC RPC with respect
to alignment, missing fields, etc. - it should now work correctly.
There is still the problem of decoding the private data field.

22 years agoAvoid a symbol conflict with /usr/include/sys/vfs.h under AIX.
Tim Potter [Tue, 11 May 1999 23:58:20 +0000 (23:58 +0000)]
Avoid a symbol conflict with /usr/include/sys/vfs.h under AIX.
Thanks to Steve Noyd Jones <noyd@u.washington.edu>

22 years agovariation of random SMBtrans2. random delays in between reconnections
Luke Leighton [Fri, 7 May 1999 20:21:30 +0000 (20:21 +0000)]
variation of random SMBtrans2.  random delays in between reconnections
increased.  size of large random buffer decreased.  number of reconnection
attempts increased.

22 years agothe times in sam_user_info_23 may not be in the right order
Luke Leighton [Fri, 7 May 1999 18:21:38 +0000 (18:21 +0000)]
the times in sam_user_info_23 may not be in the right order

22 years agohad to move day display names into lib/util, to get rpctorture to compile.
Luke Leighton [Fri, 7 May 1999 18:20:59 +0000 (18:20 +0000)]
had to move day display names into lib/util, to get rpctorture to compile.

22 years agoFix compiler warning.
Tim Potter [Thu, 6 May 1999 23:52:00 +0000 (23:52 +0000)]
Fix compiler warning.

Added checks to panic if VFS functions are passed NULL pointers.  This
may expose some bugs that have been lurking about.  Checks can be
easily removed later.

22 years agoclean-up of cache-getpw-hash code needed (make proto showed up loads
Luke Leighton [Thu, 6 May 1999 18:13:23 +0000 (18:13 +0000)]
clean-up of cache-getpw-hash code needed (make proto showed up loads
of functions that should be static).

22 years agoJani Jaakkola's "getpwuid() / getpwnam()" hash-cache-hack
Luke Leighton [Thu, 6 May 1999 18:05:45 +0000 (18:05 +0000)]
Jani Jaakkola's "getpwuid() / getpwnam()" hash-cache-hack

22 years agoadded 2 more (unknown) NTLMSSP negotiation bits
Luke Leighton [Thu, 6 May 1999 18:02:39 +0000 (18:02 +0000)]
added 2 more (unknown) NTLMSSP negotiation bits

22 years agoalignment needed after unicode string names in password change.
Luke Leighton [Thu, 6 May 1999 18:02:05 +0000 (18:02 +0000)]
alignment needed after unicode string names in password change.

22 years agolast part of RPC api change.
Jean-François Micouleau [Mon, 3 May 1999 22:04:02 +0000 (22:04 +0000)]
last part of RPC api change.

and of rpcclient eventlog funtion

Jean Francois

22 years agoChanged the RPC api definition now first arg is a pipe_struct
Jean-François Micouleau [Mon, 3 May 1999 22:00:33 +0000 (22:00 +0000)]
Changed the RPC api definition now first arg is a pipe_struct

found some memory leak in the spoolss code.

Jean Francois

22 years agoAdded a new command to rpcclient: eventlog <log> where <log> is System,
Jean-François Micouleau [Mon, 3 May 1999 21:52:29 +0000 (21:52 +0000)]
Added a new command to rpcclient: eventlog <log> where <log> is System,
Security or Application. That's a crude equivalent to the eventviewer.

Jean Francois

22 years agoAdding "time" rpcclient command which displays the remote time.
Matthew Chapman [Sat, 1 May 1999 05:56:55 +0000 (05:56 +0000)]
Adding "time" rpcclient command which displays the remote time.
Also added special "now" time to the "at" command, e.g.:
  at now /i cmd   ; pops up a command prompt

22 years agoadded server ntlmv2 false/auto/true parameter, defaults to off.
Luke Leighton [Sat, 1 May 1999 01:41:28 +0000 (01:41 +0000)]
added server ntlmv2 false/auto/true parameter, defaults to off.

22 years agomd5 and hmac_md5
Luke Leighton [Fri, 30 Apr 1999 05:37:55 +0000 (05:37 +0000)]
md5 and hmac_md5

22 years agouse /dev/urandom not /dev/random in head branch.
Andrew Tridgell [Wed, 28 Apr 1999 02:00:38 +0000 (02:00 +0000)]
use /dev/urandom not /dev/random in head branch.

also got rid of /tmp time based random source. I saw a system with a
huge number of files in /tmp and logging in was taking a _long_ time.

22 years agorpc_parse/parse_misc.c : defined a new BUFFER5 struct
Jean-François Micouleau [Tue, 27 Apr 1999 10:43:32 +0000 (10:43 +0000)]
rpc_parse/parse_misc.c : defined a new BUFFER5 struct
include/ntdomain.h     : added rpc_spoolss.h include statement
include/rpc_dce.h      : added definition of RPC_ALTER_CONTEXT request &
param/loadparm.c       : 2 new options for NT printing support and some
                         changes to initial values in the LPRNG case.
rpc_parse/parse_prs.c  : added prs_uint16s()
rpc_parse/parse_rpc.c  : added SYNT_SPOOLSS_V1 and code for the
                         alter-context support.
rpc_server/srv_pipe.c  : alter-context support

Jean Francois

22 years agoAdding RPC spoolss code support and some new low-level printing functions.
Jean-François Micouleau [Tue, 27 Apr 1999 10:13:02 +0000 (10:13 +0000)]
Adding RPC spoolss code support and some new low-level printing functions.

Jean Francois

22 years agoAdding scheduler control pipe (\atsvc), client-side routines, and rpcclient
Matthew Chapman [Fri, 23 Apr 1999 23:49:46 +0000 (23:49 +0000)]
Adding scheduler control pipe (\atsvc), client-side routines, and rpcclient
command "at" (compatible with NT's "at" command - see rpcclient commit) -
useful for remote NT administration.

22 years agoNew rpcclient command "at" for NT scheduler control, a slightly improved
Matthew Chapman [Fri, 23 Apr 1999 15:01:35 +0000 (15:01 +0000)]
New rpcclient command "at" for NT scheduler control, a slightly improved
version of the NT command.

at { time [/INTERACTIVE] [{/EVERY|/NEXT}:5,Sun,...] command | [/DEL] [jobid] }

Examples (options used in abbreviated form):
at                                         ; Shows all jobs
at 1                                       ; Detail on job 1
at /D                                      ; Deletes all jobs
at /D 1                                    ; Deletes job 1
at 11:11:11AM /I /N:1 d:\humour\silly.exe  ; First of next month
at 9:00AM /E:M,T,W,Th,F net send MATTY Hi  ; Each weekday
at 11:00PM /E c:\winnt\backup.exe          ; Every day

22 years agoAdded new define NT_STATUS_MORE_ENTRIES = 0x105.
Matthew Chapman [Fri, 23 Apr 1999 14:48:22 +0000 (14:48 +0000)]
Added new define NT_STATUS_MORE_ENTRIES = 0x105.

22 years agoAdding scheduler control pipe (\atsvc), client-side routines, and rpcclient
Matthew Chapman [Fri, 23 Apr 1999 14:47:45 +0000 (14:47 +0000)]
Adding scheduler control pipe (\atsvc), client-side routines, and rpcclient
command "at" (compatible with NT's "at" command - see rpcclient commit) -
useful for remote NT administration.

22 years agoMake proto for VFS init changes and VFS fsync() stuff.
Tim Potter [Tue, 20 Apr 1999 03:44:06 +0000 (03:44 +0000)]
Make proto for VFS init changes and VFS fsync() stuff.

22 years agoRewrite of VFS initialisation routines.
Tim Potter [Tue, 20 Apr 1999 03:42:48 +0000 (03:42 +0000)]
Rewrite of VFS initialisation routines.

22 years agoAdded dummy connect and disconnect VFS operations to avoid the
Tim Potter [Tue, 20 Apr 1999 03:38:22 +0000 (03:38 +0000)]
Added dummy connect and disconnect VFS operations to avoid the
possibilty of NULL function pointers being called.

Changed arguments to fsync() function to break dependency on

22 years agoChanged the way a VFS is initialised. The vfs_init() function is
Tim Potter [Tue, 20 Apr 1999 03:37:11 +0000 (03:37 +0000)]
Changed the way a VFS is initialised.  The vfs_init() function is
executed from the VFS object file and it returns a pointer to a
vfs_ops structure.  If any of the function pointers in vfs_ops are
NULL, then they are replaced with the standard disk functions.  This
should allow disk-related VFS modules to be easily added.  I've
written an auditing VFS module which logs various calls (connect,
disconnect, mkdir, rmdir, open and a few others) to syslog in a couple
of pages of code.

Thanks to tridge for various useful suggestions.

22 years agoBroke out vfs structures to separate header file.
Tim Potter [Tue, 20 Apr 1999 03:31:12 +0000 (03:31 +0000)]
Broke out vfs structures to separate header file.

Some cpp hackery to avoid redefining types used by vfs header file.

Added a vfs_connection_struct to connection_struct so it can be
freed when a VFS share is disconnected.

22 years agoChanged arguments to fsync() function to break dependency on
Tim Potter [Tue, 20 Apr 1999 03:29:05 +0000 (03:29 +0000)]
Changed arguments to fsync() function to break dependency on

22 years agoBroke out of smb.h
Tim Potter [Tue, 20 Apr 1999 03:28:46 +0000 (03:28 +0000)]
Broke out of smb.h

Changed arguments to fsync() function to break dependency on

22 years agoopen_socket_in() takes a different number of parameters in the head
Andrew Tridgell [Sat, 17 Apr 1999 14:13:06 +0000 (14:13 +0000)]
open_socket_in() takes a different number of parameters in the head
branch. We need that merge soon!

22 years agofixed a typo
Andrew Tridgell [Sat, 17 Apr 1999 14:09:15 +0000 (14:09 +0000)]
fixed a typo

22 years agodamn, new files need to be added to the head branch first, I've told
Andrew Tridgell [Sat, 17 Apr 1999 14:06:57 +0000 (14:06 +0000)]
damn, new files need to be added to the head branch first, I've told
other people this before then forgot myself :)

22 years agoRemoved unused vfs_read_data() function at the request of tridge.
Tim Potter [Wed, 14 Apr 1999 07:21:48 +0000 (07:21 +0000)]
Removed unused vfs_read_data() function at the request of tridge.

Removed unnecessary SSL stuff from vfs_write_data().

22 years agoFix compile warning in nametouid().
Tim Potter [Mon, 12 Apr 1999 01:39:46 +0000 (01:39 +0000)]
Fix compile warning in nametouid().

22 years agoAllow "Yes" and "No" as well as "True" and "False" in the three-state
Matthew Chapman [Sun, 11 Apr 1999 11:47:33 +0000 (11:47 +0000)]
Allow "Yes" and "No" as well as "True" and "False" in the three-state

22 years agoTrying to improve DC location & browsing performance for the average user
Matthew Chapman [Sun, 11 Apr 1999 10:23:19 +0000 (10:23 +0000)]
Trying to improve DC location & browsing performance for the average user
who doesn't know what an LMB or DMB is.

* check_master_browser_exists now performs the check the first time
around, so if there is indeed no master browser then Samba takes up the job
much faster.

* Upped default OS level to 32. There is no reason why some stupid little
Windows box should become LMB instead of a Samba *server*.

* "domain master" now defaults to "auto". Currently this attempts to
become DMB iff Samba is the PDC (ala Windows NT). "preferred master" also
defaults to "auto", which enables preferred master iff Samba is DMB.

* lp_server_role now just returns the predetermined role, rather than
working it out each time, since the server role is becoming very heavily
used (esp for the BDC code).

22 years agoFixed rpcclient's samuser command.
Matthew Chapman [Sun, 11 Apr 1999 02:34:44 +0000 (02:34 +0000)]
Fixed rpcclient's samuser command.

22 years agoIncorrect length (number of unicode characters rather than buffer size)
Matthew Chapman [Sun, 11 Apr 1999 01:39:16 +0000 (01:39 +0000)]
Incorrect length (number of unicode characters rather than buffer size)
passed to ascii_to_unibuf in make_buffer2 was breaking User Manager.

22 years agoModification to compile under egcs 2.91.63.
Tim Potter [Sun, 11 Apr 1999 00:27:43 +0000 (00:27 +0000)]
Modification to compile under egcs 2.91.63.

22 years agoPut this back after it dropped off.
John Terpstra [Fri, 9 Apr 1999 13:30:36 +0000 (13:30 +0000)]
Put this back after it dropped off.

22 years agoMainly BDC-related changes.
Matthew Chapman [Thu, 8 Apr 1999 05:37:15 +0000 (05:37 +0000)]
Mainly BDC-related changes.

* Added -b switch to smbpasswd (for a BDC (server) trust account).
* Propagate sec_chan into the various functions which change trust account
  passwords, so they can be used for domain control and inter-domain

22 years agoMainly BDC-related changes.
Matthew Chapman [Thu, 8 Apr 1999 05:36:39 +0000 (05:36 +0000)]
Mainly BDC-related changes.

* Propagate sec_chan into the various functions which change trust account
  passwords, so they can be used for domain control and inter-domain

22 years agoMainly BDC-related changes.
Matthew Chapman [Thu, 8 Apr 1999 05:36:15 +0000 (05:36 +0000)]
Mainly BDC-related changes.

* Propagate sec_chan into the various functions which change trust account
  passwords, so they can be used for domain control and inter-domain
* Fix for endianness problem reported by Edan Idzerda <edan@mtu.edu>. A
  BUFFER2 is really a "unibuf" in my terminology and we should treat it as
* Added some more common NT structures (BIGINT, BUFHDR2, BUFFER4).
* Added NET_SAM_SYNC (-> NetDatabaseSync2) RPC for account replication.
  Still experimental and incomplete, with a few too many NULL security
  descriptors lying around (must go look at Jeremy's SD code). Haven't
  worked out password encryption yet either.
  However, the XXX_INFO structures I've added to rpc_netlogon.h are quite
  nice as they give some insight into how these objects are stored in the

22 years agoMainly BDC-related changes.
Matthew Chapman [Thu, 8 Apr 1999 05:35:53 +0000 (05:35 +0000)]
Mainly BDC-related changes.

* Fix for endianness problem reported by Edan Idzerda <edan@mtu.edu>. A
  BUFFER2 is really a "unibuf" in my terminology and we should treat it as
* Added some more common NT structures (BIGINT, BUFHDR2, BUFFER4).
* Added NET_SAM_SYNC (-> NetDatabaseSync2) RPC for account replication.
  Still experimental and incomplete, with a few too many NULL security
  descriptors lying around (must go look at Jeremy's SD code). Haven't
  worked out password encryption yet either.
  However, the XXX_INFO structures I've added to rpc_netlogon.h are quite
  nice as they give some insight into how these objects are stored in the

22 years agoFix for endianness problem reported by Edan Idzerda <edan@mtu.edu>. A
Matthew Chapman [Thu, 8 Apr 1999 05:34:11 +0000 (05:34 +0000)]
Fix for endianness problem reported by Edan Idzerda <edan@mtu.edu>. A
BUFFER2 is really a "unibuf" in my terminology and we should treat it as

22 years agoAlways do a compile before CVS commit! )-:
Tim Potter [Thu, 8 Apr 1999 03:13:33 +0000 (03:13 +0000)]
Always do a compile before CVS commit!  )-:

22 years agoForgot about closedir() function for VFS. Hoo embarassing.
Tim Potter [Thu, 8 Apr 1999 03:01:18 +0000 (03:01 +0000)]
Forgot about closedir() function for VFS.  Hoo embarassing.

22 years agoChanged calls to strncat() to safe_strcat(). Fix from SAMBA_2_0 branch.
Tim Potter [Wed, 7 Apr 1999 04:20:00 +0000 (04:20 +0000)]
Changed calls to strncat() to safe_strcat().  Fix from SAMBA_2_0 branch.

22 years agoDejan Ilic: spotted "NETLOGON" bug, failure of NT_STATUS WRONG_PASSWORD.
Luke Leighton [Tue, 6 Apr 1999 16:36:52 +0000 (16:36 +0000)]
Dejan Ilic: spotted "NETLOGON" bug, failure of NT_STATUS WRONG_PASSWORD.

22 years agoUnder IRIX, a void function may return a value, even if that value is
Tim Potter [Mon, 5 Apr 1999 23:35:50 +0000 (23:35 +0000)]
Under IRIX, a void function may return a value, even if that value is

22 years agoFix incorrect conversion of fd_attempt_open() calls to vfs_ops.open().
Tim Potter [Mon, 5 Apr 1999 05:19:34 +0000 (05:19 +0000)]
Fix incorrect conversion of fd_attempt_open() calls to vfs_ops.open().

Use VFS I/O for fd_attempt_open() and check_access_allowed_for_current_user().

22 years agoUse VFS I/O for stat_cache_lookup()
Tim Potter [Mon, 5 Apr 1999 05:18:44 +0000 (05:18 +0000)]
Use VFS I/O for stat_cache_lookup()

22 years agoDisable VFS routines if no libdl available.
Tim Potter [Mon, 5 Apr 1999 05:18:07 +0000 (05:18 +0000)]
Disable VFS routines if no libdl available.

22 years agoDocumentation for VFS construction.
Tim Potter [Mon, 5 Apr 1999 05:14:47 +0000 (05:14 +0000)]
Documentation for VFS construction.

22 years agoFixed spelling typo in Kerberos configure option.
Tim Potter [Sun, 4 Apr 1999 22:58:44 +0000 (22:58 +0000)]
Fixed spelling typo in Kerberos configure option.

22 years agoUpdated for VFS merge.
Tim Potter [Sun, 4 Apr 1999 07:24:14 +0000 (07:24 +0000)]
Updated for VFS merge.

22 years agoFixed up incorrect calls to read_file().
Tim Potter [Sun, 4 Apr 1999 07:18:38 +0000 (07:18 +0000)]
Fixed up incorrect calls to read_file().

22 years agoUse VFS operations for file I/O.
Tim Potter [Sun, 4 Apr 1999 07:05:03 +0000 (07:05 +0000)]
Use VFS operations for file I/O.

Modified args to read_predict to pass in fsp.

Renamed sync_file() function to sys_sync_file().

22 years agoUse VFS operations for file I/O.
Tim Potter [Sun, 4 Apr 1999 07:03:47 +0000 (07:03 +0000)]
Use VFS operations for file I/O.

Modified invocations of fd_attempt_close as it now takes a fsp instead
of a fd.

22 years ago#ifdef'ed out functions obsoleted by VFS:
Tim Potter [Sun, 4 Apr 1999 06:53:46 +0000 (06:53 +0000)]
#ifdef'ed out functions obsoleted by VFS:


Added warning comments to dos_mkdir() as it's used in client code.

22 years agoAdded smbd/vfs.o and smbd/vfs-wrap.o to object list for smbd.
Tim Potter [Sun, 4 Apr 1999 06:47:57 +0000 (06:47 +0000)]
Added smbd/vfs.o and smbd/vfs-wrap.o to object list for smbd.

22 years agoWrote VFS initialisation routines.
Tim Potter [Sun, 4 Apr 1999 06:39:45 +0000 (06:39 +0000)]
Wrote VFS initialisation routines.

Converted some useful routines from doscalls.c and changed them to use
VFS I/O functions:


Some of these functions have been #ifdef'ed out of doscalls.c as they
are not used anywhere else.  Not sure whether they should be deleted
or not.  The remaining dos_* calls seem to be used by clients and for
locking stuff.  This should be cleaned up sometime.

22 years agoWrapped up all VFS disk I/O functions for portability. I remember
Tim Potter [Sun, 4 Apr 1999 06:37:07 +0000 (06:37 +0000)]
Wrapped up all VFS disk I/O functions for portability.  I remember
tridge telling me why this needed to be done but I have since
forgotten.  (-:

22 years agoUse VFS operations for file I/O.
Tim Potter [Sun, 4 Apr 1999 06:25:13 +0000 (06:25 +0000)]
Use VFS operations for file I/O.

22 years agoCall VFS initialisation, share connect and share disconnect functions.
Tim Potter [Sun, 4 Apr 1999 06:22:22 +0000 (06:22 +0000)]
Call VFS initialisation, share connect and share disconnect functions.

Initialise VFS function pointers with symbols from the 'vfs object'
file, use disk pointers otherwise.

Added a hook to check for a 'default' VFS share (like the [homes]
section).  Currently empty - I'm unsure whether this will be useful or

22 years agoUse VFS operations for file I/O.
Tim Potter [Sun, 4 Apr 1999 05:57:16 +0000 (05:57 +0000)]
Use VFS operations for file I/O.

Changed arg to fd_attempt_close() from fd to fsp_struct.

22 years agoUse VFS operations for file I/O.
Tim Potter [Sun, 4 Apr 1999 05:46:40 +0000 (05:46 +0000)]
Use VFS operations for file I/O.

Pass files_struct and connection_struct to read_predict() and
do_read_prediction() functions, respectively.

22 years agoUse VFS operations for file I/O.
Tim Potter [Sun, 4 Apr 1999 05:40:28 +0000 (05:40 +0000)]
Use VFS operations for file I/O.

Changed calls to dos_{opendir,readdir} to vfs_{opendir,readdir}

22 years agoUse VFS operations for file I/O.
Tim Potter [Sun, 4 Apr 1999 05:32:42 +0000 (05:32 +0000)]
Use VFS operations for file I/O.

22 years agoAdded struct vfs_ops describing virtual filesystem operations.
Tim Potter [Sun, 4 Apr 1999 05:29:41 +0000 (05:29 +0000)]
Added struct vfs_ops describing virtual filesystem operations.

Created constants for pstring and fstring lengths.

22 years agoAdded configuration parameter 'vfs object' to specify loadable object
Tim Potter [Sun, 4 Apr 1999 04:54:50 +0000 (04:54 +0000)]
Added configuration parameter 'vfs object' to specify loadable object
file containing a VFS implementation.

22 years agonew files needed by profiling code
Andrew Tridgell [Thu, 1 Apr 1999 05:35:22 +0000 (05:35 +0000)]
new files needed by profiling code

22 years agoKen McDonell from SGI was interested in adding some profiling
Andrew Tridgell [Thu, 1 Apr 1999 05:22:58 +0000 (05:22 +0000)]
Ken McDonell from SGI was interested in adding some profiling
capabilities to Samba so that Samba could talk to the SGI PCP
(Performance Co-Pilot) apps.

This change adds a profiling shared memory area and uses it to count
two fairly trivial things, the number of uid switches and the number
of SMB packets processes. To add more just edit include/profile.h and
then increment it at the right place.

I've also added a -P switch to smbstatus to dump the profile area.

22 years agodeclare ttarf as double, as in client.c
Alexandre Oliva [Tue, 30 Mar 1999 10:41:07 +0000 (10:41 +0000)]
declare ttarf as double, as in client.c

22 years agouse double instead of SMB_BIG_UINT for dir_total and ttarf
Alexandre Oliva [Tue, 30 Mar 1999 10:25:20 +0000 (10:25 +0000)]
use double instead of SMB_BIG_UINT for dir_total and ttarf
(by tridge, merged from branch)

22 years agoadded SamrQueryDomainInfo call info level 0x3 for kix32.exe support
Luke Leighton [Mon, 29 Mar 1999 17:24:46 +0000 (17:24 +0000)]
added SamrQueryDomainInfo call info level 0x3 for kix32.exe support

22 years ago"User Manager" - create user + change password now work.
Luke Leighton [Thu, 25 Mar 1999 21:32:04 +0000 (21:32 +0000)]
"User Manager" - create user + change password now work.
next problem: user group adding not supported so an "access denied"
message is reported instead of "ok" when a new user is created.

22 years agoSMB_BIG_INT dir_total incompatible with %d: must use %llu (line 446)
Luke Leighton [Thu, 25 Mar 1999 21:01:08 +0000 (21:01 +0000)]
SMB_BIG_INT dir_total incompatible with %d: must use %llu (line 446)

22 years agofixed issues with "Welcome to SAMBA Domain" for when admin user/pass is
Luke Leighton [Thu, 25 Mar 1999 20:56:28 +0000 (20:56 +0000)]
fixed issues with "Welcome to SAMBA Domain" for when admin user/pass is
used to add workstation to domain.  unix account db not modified: only
SAM password db is used.

22 years ago* client/client.c (dir_total): use SMB_BIG_UINT
Alexandre Oliva [Thu, 25 Mar 1999 20:21:01 +0000 (20:21 +0000)]
* client/client.c (dir_total): use SMB_BIG_UINT
* client/clitar.c (ttarf): ditto
* * lib/snprintf.c: support long longs; adapted from Cloyce D. Spradling's
patch <cloyce@headgear.org>

22 years agoadded SAMR_USER_INFO_24 info level to samr_set_userinfo. this is used
Luke Leighton [Thu, 25 Mar 1999 15:14:30 +0000 (15:14 +0000)]
added SAMR_USER_INFO_24 info level to samr_set_userinfo.  this is used
by "Welcome to SAMBA Domain".

22 years agoSAM database "set user info".
Luke Leighton [Thu, 25 Mar 1999 13:54:31 +0000 (13:54 +0000)]
SAM database "set user info".

- removed DOM_RID4

- removed SAMR_UNKNOWN_32

- added SAMR_SET_USERINFO (opcode 0x32)

- added level 0x1 to SAMR_QUERY_DOM_INFO (needed for create user)

- fixed pwdb_gethexpwd() it was failing on XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

- added mod_sam21pwd_entry()

- preparing to call mod_sam21pwd_entry()

- added "user session key" to user_struct.dc.  this is md4(nt#) and is
  needed to decode user's clear-text passwords in SAMR_SET_USERINFO.

- split code out in chgpasswd.c to decode 516 byte password buffers.

22 years agoNULL pointer handling in nt_lm_owf_gen
Luke Leighton [Wed, 24 Mar 1999 21:23:39 +0000 (21:23 +0000)]
NULL pointer handling in nt_lm_owf_gen

22 years agomatt, you had removed a "pointer" from DOM_R_REF structure which made it
Luke Leighton [Wed, 24 Mar 1999 21:09:34 +0000 (21:09 +0000)]
matt, you had removed a "pointer" from DOM_R_REF structure which made it
possible to fix lsa_r_lookup_names, but forgot to add that pointer in
to lsa_r_lookup_sids, where DOM_R_REF was also being used.

22 years agoremoved 2 unnecessary args from make_uni_hdr.
Luke Leighton [Wed, 24 Mar 1999 17:26:08 +0000 (17:26 +0000)]
removed 2 unnecessary args from make_uni_hdr.

22 years agoFixed LSA Lookup Names. There were a few too many NULL pointers in a
Matthew Chapman [Wed, 24 Mar 1999 12:18:28 +0000 (12:18 +0000)]
Fixed LSA Lookup Names. There were a few too many NULL pointers in a
negative response, which tended to crash lsass.exe.

22 years agodynamic memory allocation in samr enum dom users. works with 849 entries now.
Luke Leighton [Tue, 23 Mar 1999 20:43:44 +0000 (20:43 +0000)]
dynamic memory allocation in samr enum dom users.  works with 849 entries now.

22 years agoFixed a typo where the RPC header mem_buffer was initialised as 0x8 bytes long
Matthew Chapman [Tue, 23 Mar 1999 15:01:37 +0000 (15:01 +0000)]
Fixed a typo where the RPC header mem_buffer was initialised as 0x8 bytes long
rather than 0x18. Rather nasty, I doubt the client ever worked for multiple

22 years agoERRmoredata is informational and should not be treated as a hard error
Matthew Chapman [Tue, 23 Mar 1999 14:58:26 +0000 (14:58 +0000)]
ERRmoredata is informational and should not be treated as a hard error

22 years agoSame infinite loop bug as I fixed in SAMBA_2_0, but I just spent half
Matthew Chapman [Tue, 23 Mar 1999 14:56:25 +0000 (14:56 +0000)]
Same infinite loop bug as I fixed in SAMBA_2_0, but I just spent half
an hour tracking it down in HEAD so I'm going to commit it here as well.

22 years ago#if defined(HAVE_MYSQL_H) || defined(WITH_MYSQLSAM)
Matthew Chapman [Tue, 23 Mar 1999 14:14:12 +0000 (14:14 +0000)]
#if defined(HAVE_MYSQL_H) || defined(WITH_MYSQLSAM)
should instead have &&.

22 years agoAnother UNICODE issue - this time BUFFER2 was being transmitted incorrectly.
Matthew Chapman [Tue, 23 Mar 1999 13:45:42 +0000 (13:45 +0000)]
Another UNICODE issue - this time BUFFER2 was being transmitted incorrectly.

22 years agoIf a buffer pointer is NULL then its length_is() attribute should not be
Matthew Chapman [Tue, 23 Mar 1999 02:31:34 +0000 (02:31 +0000)]
If a buffer pointer is NULL then its length_is() attribute should not be

Such a problem was crashing Server Manager.

22 years agoThe line:
Matthew Chapman [Sun, 21 Mar 1999 13:01:31 +0000 (13:01 +0000)]
The line:

    this_pdu_data_pos = data_pos - p->prev_pdu_file_offset;

is correct for first two PDU's only, after that it results in extra
garbage after each fragment and hence "Protocol Error" messages from
NT4 SP4. Changed to:

    this_pdu_data_pos = (pdu_data_sent == 0) ? 0 : (pdu_data_sent - 0x18);

22 years agoEnsuring return status is aligned in SAMR_QUERY_DISPINFO response.
Matthew Chapman [Sun, 21 Mar 1999 12:55:18 +0000 (12:55 +0000)]
Ensuring return status is aligned in SAMR_QUERY_DISPINFO response.

22 years agoadded a "createuser" command.
Luke Leighton [Fri, 19 Mar 1999 21:20:19 +0000 (21:20 +0000)]
added a "createuser" command.

22 years agovalue->enc_secret not value->secret
Luke Leighton [Fri, 19 Mar 1999 21:15:03 +0000 (21:15 +0000)]
value->enc_secret not value->secret