tprouty/samba.git
16 years agor3438: fix some kerberos-related prototype warnings
Günther Deschner [Mon, 1 Nov 2004 18:44:15 +0000 (18:44 +0000)]
r3438: fix some kerberos-related prototype warnings

Guenther

16 years agor3437: fix the build.
Günther Deschner [Mon, 1 Nov 2004 16:55:40 +0000 (16:55 +0000)]
r3437: fix the build.

Guenther

16 years agor3436: Fix build with recent heimdal-versions (0.6.3) as pointed out by Luke
Günther Deschner [Mon, 1 Nov 2004 16:18:14 +0000 (16:18 +0000)]
r3436: Fix build with recent heimdal-versions (0.6.3) as pointed out by Luke
Mewburn <lukem@NetBSD.org> and close Bugzilla #1661.

Leaving the old define for KRB5_KPASSWD_VERS_SETPW (added by Antti
Andreimann) as fallback when the library does not provide one.

Guenther

16 years agor3420: install smbwrapper and add the include directory in the package
Gerald Carter [Mon, 1 Nov 2004 02:00:09 +0000 (02:00 +0000)]
r3420: install smbwrapper and add the include directory in the package

16 years agor3417: BUG 350: patch from Matt Selsky <selsky@columbia.edu> to use autoconf feature...
Gerald Carter [Sun, 31 Oct 2004 22:06:37 +0000 (22:06 +0000)]
r3417: BUG 350: patch from Matt Selsky <selsky@columbia.edu> to use autoconf feature preopcessing header file only (3 problematic headers on solaris)

16 years agor3408: Another build fix
Volker Lendecke [Sun, 31 Oct 2004 11:26:31 +0000 (11:26 +0000)]
r3408: Another build fix

16 years agor3407: Fix the build
Volker Lendecke [Sun, 31 Oct 2004 11:20:07 +0000 (11:20 +0000)]
r3407: Fix the build

16 years agor3399: a few more packaging fixes
Gerald Carter [Sun, 31 Oct 2004 03:11:41 +0000 (03:11 +0000)]
r3399: a few more packaging fixes

16 years agor3398: fixing up solaris packaging for 3.0.x
Gerald Carter [Sun, 31 Oct 2004 02:32:35 +0000 (02:32 +0000)]
r3398: fixing up solaris packaging for 3.0.x

16 years agor3381: More merging of the #1717 patch. Fixup some erroneous assumptions about
Jeremy Allison [Sat, 30 Oct 2004 01:32:05 +0000 (01:32 +0000)]
r3381: More merging of the #1717 patch. Fixup some erroneous assumptions about
memcpy's into fqdn names. I think the original intent was to create
MYNAME.fqdn.tail.part.
Will need testing to see I haven't broken keytab support.
Jeremy.

16 years agor3379: More merging of kerberos keytab and salting fixes from Nalin Dahyabhai <nalin...
Jeremy Allison [Sat, 30 Oct 2004 00:34:58 +0000 (00:34 +0000)]
r3379: More merging of kerberos keytab and salting fixes from Nalin Dahyabhai <nalin@redhat.com>
(bugid #1717).
Jeremy.

16 years agor3377: Merge in first part of modified patch from Nalin Dahyabhai <nalin@redhat.com>
Jeremy Allison [Fri, 29 Oct 2004 22:38:10 +0000 (22:38 +0000)]
r3377: Merge in first part of modified patch from Nalin Dahyabhai <nalin@redhat.com>
for bug #1717.The rest of the code needed to call this patch has not yet been
checked in (that's my next task). This has not yet been tested - I'll do this
once the rest of the patch is integrated.
Jeremy.

16 years agor3345: More MIT/Heimdal tests for comparing enctypes now.
Jeremy Allison [Fri, 29 Oct 2004 00:02:32 +0000 (00:02 +0000)]
r3345: More MIT/Heimdal tests for comparing enctypes now.
Jeremy.

16 years agor3342: More MIT/Heimdal fixes to allow an enctype to be explicitly set in a krb5_creds
Jeremy Allison [Thu, 28 Oct 2004 23:50:14 +0000 (23:50 +0000)]
r3342: More MIT/Heimdal fixes to allow an enctype to be explicitly set in a krb5_creds
struct.
Jeremy.

16 years agor3339: allow tdbtool to be built
Herb Lewis [Thu, 28 Oct 2004 22:58:21 +0000 (22:58 +0000)]
r3339: allow tdbtool to be built

16 years agor3326: BUG 1782: reorder loadparm.c to prevent testparm from displaying synonyms...
Gerald Carter [Thu, 28 Oct 2004 13:50:13 +0000 (13:50 +0000)]
r3326: BUG 1782: reorder loadparm.c to prevent testparm from displaying synonyms; patch from Luke Mewburn <lukem@NetBSD.org>

16 years agor3298: turning sendfile back off by default
Gerald Carter [Wed, 27 Oct 2004 19:44:12 +0000 (19:44 +0000)]
r3298: turning sendfile back off by default

16 years agor3296: Fix to ensure entries are stored in correct order. Bug #1498. Patch from
Jeremy Allison [Wed, 27 Oct 2004 18:11:41 +0000 (18:11 +0000)]
r3296: Fix to ensure entries are stored in correct order. Bug #1498. Patch from
SATOH Fumiyasu <fumiya@samba.gr.jp>.
Jeremy.

16 years agor3294: Fix for SMB signing with 56-bit DES session keys. From Nalin Dahyabhai <nalin...
Jeremy Allison [Wed, 27 Oct 2004 17:40:28 +0000 (17:40 +0000)]
r3294: Fix for SMB signing with 56-bit DES session keys. From Nalin Dahyabhai <nalin@redhat.com>.
Jeremy.

16 years agor3292: A fix from Narayana Pattipati <narayana[dot]pattipati[at]wipro\dotty/com...
Richard Sharpe [Wed, 27 Oct 2004 17:03:09 +0000 (17:03 +0000)]
r3292: A fix from  Narayana Pattipati <narayana[dot]pattipati[at]wipro\dotty/com> for
Solaris to ensure we distinguish properly between 5.1 and 5.10.

16 years agor3275: Fix from Michael Sweet <mike@easysw.com> for bug #1892.
Jeremy Allison [Wed, 27 Oct 2004 00:52:12 +0000 (00:52 +0000)]
r3275: Fix from Michael Sweet <mike@easysw.com> for bug #1892.
Jeremy.

16 years agor3273: Ensure we're consistent in the use of strchr_m for '@'.
Jeremy Allison [Wed, 27 Oct 2004 00:41:41 +0000 (00:41 +0000)]
r3273: Ensure we're consistent in the use of strchr_m for '@'.
Jeremy.

16 years agor3264: fix lmhosts lookup so that we don't say we found something when we really...
Gerald Carter [Tue, 26 Oct 2004 14:22:37 +0000 (14:22 +0000)]
r3264: fix lmhosts lookup so that we don't say we found something when we really didn't

16 years agor3230: swat/lang/*/include/*html are no installed anymore
Gerald Carter [Tue, 26 Oct 2004 02:35:41 +0000 (02:35 +0000)]
r3230: swat/lang/*/include/*html are no installed anymore

16 years agor3228: Fix for bugzilla #1884 from Brett Funderburg. Use the value of
Tim Potter [Tue, 26 Oct 2004 01:37:19 +0000 (01:37 +0000)]
r3228: Fix for bugzilla #1884 from Brett Funderburg.  Use the value of
desired_access passed in to the lsa open policy routine.  Use a default
value that netapps and windows likes as well.

16 years agor3225: Fix correct use of resume name. Cope with the resume
Jeremy Allison [Mon, 25 Oct 2004 23:39:01 +0000 (23:39 +0000)]
r3225: Fix correct use of resume name. Cope with the resume
name not existing. Found using Samba4 RAW-SEARCH.
Jeremy.

16 years agor3221: Remove check for allow-trusted-domains so that this thing does not throw-up.
John Terpstra [Mon, 25 Oct 2004 19:41:16 +0000 (19:41 +0000)]
r3221: Remove check for allow-trusted-domains so that this thing does not throw-up.

16 years agor3151: Add the "no warranty"-section in the licence header.
Günther Deschner [Sun, 24 Oct 2004 11:50:24 +0000 (11:50 +0000)]
r3151: Add the "no warranty"-section in the licence header.

Guenther

16 years agor3146: Some cleanup for idmap_rid:
Günther Deschner [Sat, 23 Oct 2004 15:16:10 +0000 (15:16 +0000)]
r3146: Some cleanup for idmap_rid:
- fix several memleaks found by valgrind
- turn off support for trusted domains (can be reenabled with
  #define IDMAP_RID_SUPPORT_TRUSTED_DOMAINS 1)
- improve readability

Guenther

16 years agor3145: Add experimental idmap_rid-Plugin.
Günther Deschner [Sat, 23 Oct 2004 13:23:54 +0000 (13:23 +0000)]
r3145: Add experimental idmap_rid-Plugin.

Written by Sumit Bose <sbose@suse.de> and myself a while ago.

idmap_rid does a direct, static mapping between RIDs and UIDs/GIDs using
the idmap-range as offset. It does thus allow to have a unified mapping
over several winbindd-systems without having the need of a central
LDAP-Server (and all related dependencies and problems this solution can
bring).

Compile:
./configure --with-shared-modules=idmap_rid

Usage:
        idmap backend = idmap_rid

idmp_rid does even allow you to have multiple mappings (for trusted
domains). This is a rather problemtic feature and will be turned off by
default rather soon. The problem is that ranges can quickly overlap when
not measured with caution.

        idmap backend = idmap_rid:"MYDOMAIN=1000-9999 OTHER=10000-19999"

Will valgrind idmap_rid later today and fix a couple of things.

Guenther

16 years agor3143: Allow for multiple DC's to be named as #1c names in lmhosts.
Volker Lendecke [Sat, 23 Oct 2004 09:20:02 +0000 (09:20 +0000)]
r3143: Allow for multiple DC's to be named as #1c names in lmhosts.

Volker

16 years agor3140: * try to ensure consistent usage of the username map.
Gerald Carter [Fri, 22 Oct 2004 20:15:24 +0000 (20:15 +0000)]
r3140: * try to ensure consistent usage of the username map.
  Use the fully qualified DOMAIN\user format for 'security = domain|ads'
  and apply after authentication has succeeded.

* also change fill_domain_username() to only lowercase the username
  and not the domain+username.  This was a cosmetic fix only.
  makes the output more consistent with %D and %U.

16 years agor3138: Fix from Sorin Manolache <sorinm@gmail.com> for memory leak.
Jeremy Allison [Fri, 22 Oct 2004 18:38:57 +0000 (18:38 +0000)]
r3138: Fix from Sorin Manolache <sorinm@gmail.com> for memory leak.
Jeremy.

16 years agor3137: Do not simply ignore failing idmap-module init for remotemaps.
Günther Deschner [Fri, 22 Oct 2004 16:19:31 +0000 (16:19 +0000)]
r3137: Do not simply ignore failing idmap-module init for remotemaps.
(in preparation of adding idmap_rid)

Guenther

16 years agor3121: Bug #1956. Ensure errno is saved and restored consistently on a normal_close.
Jeremy Allison [Thu, 21 Oct 2004 19:51:33 +0000 (19:51 +0000)]
r3121: Bug #1956. Ensure errno is saved and restored consistently on a normal_close.
Jeremy.

16 years agor3120: Fix bug #1955 reported by Love <lha@stacken.kth.se>. Inconsistent error return.
Jeremy Allison [Thu, 21 Oct 2004 18:39:16 +0000 (18:39 +0000)]
r3120: Fix bug #1955 reported by Love <lha@stacken.kth.se>. Inconsistent error return.
Jeremy.

16 years agor3117: Fix from Tom Lackemann <cessnatomny@yahoo.com> for bug #1954.
Jeremy Allison [Thu, 21 Oct 2004 17:22:35 +0000 (17:22 +0000)]
r3117: Fix from Tom Lackemann <cessnatomny@yahoo.com> for bug #1954.
Memory leak in posix acl code.
Jeremy.

16 years agor3088: update nds schema file from Uli Iske <iske@elkb.de>
Gerald Carter [Wed, 20 Oct 2004 15:37:47 +0000 (15:37 +0000)]
r3088: update nds schema file from Uli Iske <iske@elkb.de>

16 years agor3072: Fix for bug #1947 - incorrect use of getpwnam() etc. interface.
Jeremy Allison [Tue, 19 Oct 2004 23:41:26 +0000 (23:41 +0000)]
r3072: Fix for bug #1947 - incorrect use of getpwnam() etc. interface.
Jeremy.

16 years agor3069: add 'force printername' service parameter for people that want to enforce...
Gerald Carter [Tue, 19 Oct 2004 22:13:08 +0000 (22:13 +0000)]
r3069: add 'force printername' service parameter for people that want to enforce printername == sharename for spoolss printing

16 years agor3068: strip guest mount option off before sending to kernel mount routine to avoid...
Steve French [Tue, 19 Oct 2004 22:05:38 +0000 (22:05 +0000)]
r3068: strip guest mount option off before sending to kernel mount routine to avoid logging spurious message

16 years agor3067: patch based on volker's initial work in trunk that fixes the queu update probl...
Gerald Carter [Tue, 19 Oct 2004 17:05:01 +0000 (17:05 +0000)]
r3067: patch based on volker's initial work in trunk that fixes the queu update problem when using the background daemon

16 years agor3066: BUG 1519: fix segfault caused by double free of a printer
Gerald Carter [Tue, 19 Oct 2004 16:17:23 +0000 (16:17 +0000)]
r3066: BUG 1519: fix segfault caused by double free of a printer

16 years agor3065: BUG 1519 (more): apparently the server_name notify request is used to fill...
Gerald Carter [Tue, 19 Oct 2004 14:45:48 +0000 (14:45 +0000)]
r3065: BUG 1519 (more): apparently the server_name notify request is used to fill in the title bar of the port monitor window and unless we get it right, you cannot open the printer properties from the port monitor window

16 years agor3050: Steal from Samba4 :-). Make us pass most of the new lock tests (except for
Jeremy Allison [Mon, 18 Oct 2004 22:01:10 +0000 (22:01 +0000)]
r3050: Steal from Samba4 :-). Make us pass most of the new lock tests (except for
the cancel lock which I have to add).
Jeremy.

16 years agor3049: fixing some calls in the printing code to stanard_sub_basic(); fix standard_su...
Gerald Carter [Mon, 18 Oct 2004 19:57:03 +0000 (19:57 +0000)]
r3049: fixing some calls in the printing code to stanard_sub_basic(); fix standard_sub_snum() to use the current user's gid;  add some (snum == -1) checks to standard_sub_advanced()

16 years agor3008: when checking for the existance of a lock we are only doing a single
Andrew Tridgell [Sun, 17 Oct 2004 00:22:11 +0000 (00:22 +0000)]
r3008: when checking for the existance of a lock we are only doing a single
tdb call, so there is no need to get the chainlock. This reduces the
number of tdb locking calls made on file IO

16 years agor3002: Fix for bug #1886 - prevent delete on close being set
Jeremy Allison [Sat, 16 Oct 2004 03:04:40 +0000 (03:04 +0000)]
r3002: Fix for bug #1886 - prevent delete on close being set
for readonly files (and return the correct error code).
We now pass the Samba4 test suite on this.
Jeremy.

16 years agor2999: Remove lockedkeys code. Not used.
Jeremy Allison [Fri, 15 Oct 2004 19:45:13 +0000 (19:45 +0000)]
r2999: Remove lockedkeys code. Not used.
Jeremy.

16 years agor2996: readability fix for length check; no semantic changes
Gerald Carter [Fri, 15 Oct 2004 15:04:01 +0000 (15:04 +0000)]
r2996: readability fix for length check; no semantic changes

16 years agor2979: Fix incorrect locks/unlocks in tdb_lockkeys()/tdb_unlockkeys().
Jeremy Allison [Fri, 15 Oct 2004 00:02:55 +0000 (00:02 +0000)]
r2979: Fix incorrect locks/unlocks in tdb_lockkeys()/tdb_unlockkeys().
Spotted by Taj Khattra <taj.khattra@gmail.com>.
Jeremy.

16 years agor2978: Fix #1926 typo in debug. Found by Bill McGonigle <bill+samba@bfccomputing...
Jeremy Allison [Thu, 14 Oct 2004 22:30:33 +0000 (22:30 +0000)]
r2978: Fix #1926 typo in debug. Found by Bill McGonigle <bill+samba@bfccomputing.com>.
Jeremy.

16 years agor2959: If we want to support writes >= 65536 with cli_write, then it had better
Richard Sharpe [Thu, 14 Oct 2004 03:19:57 +0000 (03:19 +0000)]
r2959: If we want to support writes >= 65536 with cli_write, then it had better
return a size_t, not an ssize_t, and we had better left shift the upper
part of the write count, not right shift it.

16 years agor2955: fixing a segfault uncovered by the changes for BUG 1519
Gerald Carter [Wed, 13 Oct 2004 19:40:22 +0000 (19:40 +0000)]
r2955: fixing a segfault uncovered by the changes for BUG 1519

16 years agor2943: solaris packagaing updates from Darren Chew <darrenc@vicscouts.asn.au>
Gerald Carter [Wed, 13 Oct 2004 12:20:07 +0000 (12:20 +0000)]
r2943: solaris packagaing updates from Darren Chew <darrenc@vicscouts.asn.au>

16 years agor2942: Add client-side support of triggering ads printer publishing over msrpc
Günther Deschner [Wed, 13 Oct 2004 01:40:35 +0000 (01:40 +0000)]
r2942: Add client-side support of triggering ads printer publishing over msrpc
setprinter calls inside the net-tool.

This is usefull to mimic the same queries a windows-client does. At
least win2k returns WERR_IO_PENDING when printer is published via
setprinter, samba returns WERR_OK but this does not hurt.

Guenther

16 years agor2935: This is a long-standing one in my patch-queue: A pair of net commands
Volker Lendecke [Tue, 12 Oct 2004 11:58:01 +0000 (11:58 +0000)]
r2935: This is a long-standing one in my patch-queue: A pair of net commands
(usersidlist/allowedusers) to scan a file server's share and list all users
who have permission to connect there.

Volker

16 years agor2924: Another #if that should be an #ifdef.
Tim Potter [Tue, 12 Oct 2004 03:34:01 +0000 (03:34 +0000)]
r2924: Another #if that should be an #ifdef.

16 years agor2923: Fix some obvious copy/paste leftover debug-messages.
Günther Deschner [Mon, 11 Oct 2004 21:09:36 +0000 (21:09 +0000)]
r2923: Fix some obvious copy/paste leftover debug-messages.

Guenther

16 years agor2918: BUG 1907: fix getprinterdriverdir_1(). have to make sure we don't add unneces...
Gerald Carter [Mon, 11 Oct 2004 20:01:01 +0000 (20:01 +0000)]
r2918: BUG 1907: fix getprinterdriverdir_1().  have to make sure we don't add unnecessary double slashes to the servername

16 years agor2917: Fix ip address override in mount.cifs mount helper and clean up warning messag...
Steve French [Mon, 11 Oct 2004 16:30:14 +0000 (16:30 +0000)]
r2917: Fix ip address override in mount.cifs mount helper and clean up warning messages from the sparse tool and expand syntax help

16 years agor2905: Apply the patch for bug #1822. Add -D_REENTRANT to CPPFLAGS and -lthread
Richard Sharpe [Mon, 11 Oct 2004 05:07:31 +0000 (05:07 +0000)]
r2905: Apply the patch for bug #1822. Add -D_REENTRANT to CPPFLAGS and -lthread
to LDFLAGS.

Patch supplied by Sean McGrath ...

16 years agor2899: Change some #if DEBUG_PASSWORD's to #ifdef DEBUG_PASSWORD.
Tim Potter [Mon, 11 Oct 2004 00:32:31 +0000 (00:32 +0000)]
r2899: Change some #if DEBUG_PASSWORD's to #ifdef DEBUG_PASSWORD.

Bugzilla #1903.

16 years agor2894: Certain versions of GNU ld the default is not to have the
Tim Potter [Sun, 10 Oct 2004 12:01:14 +0000 (12:01 +0000)]
r2894: Certain versions of GNU ld the default is not to have the
--allow-shlib-undefined flag defined.  This causes a stackload of
warnings when building modules.

Fix by Michel Gravey which closes bugzilla #1776.

16 years agor2868: Well, I'm not quite sure what I'm doing back in Samba 3.0, but anyway...
Andrew Bartlett [Sat, 9 Oct 2004 01:44:05 +0000 (01:44 +0000)]
r2868: Well, I'm not quite sure what I'm doing back in Samba 3.0, but anyway...

I've been grumbling about under-efficient calls in SAMR, and finally
got around to fixing some of them.

We now call sys_getgroups() (which in turn calls initgroups(), until
glibc 3.4 is released) to figure out a user's group membership.  This
is far, far more efficient than scanning all the groups looking for a
match, and is still the 'posix way', just using an effiecient call.

The seperate issue of 'who is in this group' remains, but this one has
been biting some people.

I need to talk to VL about how best to exersise nasty corner cases,
but my initial tests hold strong.  (The code is also much simpiler
than before, which has to count for something :-)

Andrew Bartlett

16 years agor2865: Add static and remove unused functions that only cload the blame-game
Andrew Bartlett [Fri, 8 Oct 2004 13:00:47 +0000 (13:00 +0000)]
r2865: Add static and remove unused functions that only cload the blame-game
in finding out who is causing the massive performance problems with
large LDAP directories.

Andrew Bartlett

16 years agor2852: Oh. Allow to migrate win2k3/xp-drivers as well.
Günther Deschner [Thu, 7 Oct 2004 14:47:53 +0000 (14:47 +0000)]
r2852: Oh. Allow to migrate win2k3/xp-drivers as well.
Thanks to Bjoern Jacke for his moral support :)

Guenther

16 years agor2837: Fix printer-migration w.r.t. to new naming-convention for
Günther Deschner [Thu, 7 Oct 2004 11:01:13 +0000 (11:01 +0000)]
r2837: Fix printer-migration w.r.t. to new naming-convention for
policy-handles. Also remove some unused vars.

Guenther

16 years agor2835: Since we always have -I. and -I$(srcdir) in CFLAGS, we can get rid of
Tim Potter [Thu, 7 Oct 2004 04:01:18 +0000 (04:01 +0000)]
r2835: Since we always have -I. and -I$(srcdir) in CFLAGS, we can get rid of
'..' from all #include preprocessor commands.   This fixes bugzilla #1880
where OpenVMS gets confused about the '.' characters.

16 years agor2834: Netapps can return NT_STATUS_ACCESS_DENIED when trying to return the
Tim Potter [Thu, 7 Oct 2004 03:55:39 +0000 (03:55 +0000)]
r2834: Netapps can return NT_STATUS_ACCESS_DENIED when trying to return the
security descriptor for a file.  Return an error in this case instead
of panicing trying to unpack a zero length buffer.  Found by Brett
Funderburg.

16 years agor2832: Readd WKGUID-binding to match the correct default-locations of new
Günther Deschner [Wed, 6 Oct 2004 16:21:35 +0000 (16:21 +0000)]
r2832: Readd WKGUID-binding to match the correct default-locations of new
User-, Group- and Machine-Accounts in Active Directory (this got lost
during the last trunk-merge).

This way we match e.g. default containers moved by redircmp.exe and
redirusr.exe in Windows 2003 and don't blindly default to cn=Users or
cn=Computers.

Further wkguids can be examied via "net ads search wellknownobjects=*".
This should still keep a samba3-client joining a samba4 dc. Fixes
Bugzilla #1343.

Guenther

16 years agor2828: Fix for bugzilla #1864 from Brett again.
Tim Potter [Wed, 6 Oct 2004 02:05:39 +0000 (02:05 +0000)]
r2828: Fix for bugzilla #1864 from Brett again.

Add sd->type field to security descriptor Python representation.

16 years agor2826: Complain if 'password chat' doesn't contain the %u variable. based on a patch...
Jelmer Vernooij [Tue, 5 Oct 2004 22:18:32 +0000 (22:18 +0000)]
r2826: Complain if 'password chat' doesn't contain the %u variable. based on a patch by Ronan Waide

16 years agor2824: restored the is_case_sensitive option to ms_fnmatch() in Samba3. It is
Andrew Tridgell [Tue, 5 Oct 2004 03:26:02 +0000 (03:26 +0000)]
r2824: restored the is_case_sensitive option to ms_fnmatch() in Samba3. It is
very rarely used, but we sohuldn't be removing a feature in a minor
release of this kind.

16 years agor2823: Patch from Brett Funderburg to pass create options parameter to
Tim Potter [Tue, 5 Oct 2004 01:36:37 +0000 (01:36 +0000)]
r2823: Patch from Brett Funderburg to pass create options parameter to
nt_create_andx() function.

16 years agor2822: Fix parameter confusion in priming of name-to-sid cache. Found by
Tim Potter [Tue, 5 Oct 2004 00:48:56 +0000 (00:48 +0000)]
r2822: Fix parameter confusion in priming of name-to-sid cache.  Found by
Qiao Yang.

16 years agor2821: Adding "Windows x64" as architecture string and driverdir "x64" for the
Günther Deschner [Mon, 4 Oct 2004 22:13:57 +0000 (22:13 +0000)]
r2821: Adding "Windows x64" as architecture string and driverdir "x64" for the
64bit AMD platform.

(This used to be "Windows AMD64" and "AMD64" in one of the release
candidates of SP2 for Windows XP. AMD64 is obviously still supported but
not documented.)

Guenther

16 years agor2819: Make 'password history'-behaviour in ldapsam more consistent.
Günther Deschner [Mon, 4 Oct 2004 15:53:33 +0000 (15:53 +0000)]
r2819: Make 'password history'-behaviour in ldapsam more consistent.

Currently we cannot store more then 15 password history entries (windows
NT4 allows to store 24) in ldapsam. When choosing more then "15" with
pdbedit -P "password history", we fail to initialize the password
history upon password change and overwrite the history, effectively
using a password history of "1". We do already decrease any
history-policy larger then 15 to 15 while storing the password history
list attribute in ldap.

Guenther

16 years agor2814: Exactly the same as "main" build fix.
Rafal Szczesniak [Sun, 3 Oct 2004 20:25:34 +0000 (20:25 +0000)]
r2814: Exactly the same as "main" build fix.

rafal

16 years agor2813: Fix the build.
Rafal Szczesniak [Sun, 3 Oct 2004 19:53:20 +0000 (19:53 +0000)]
r2813: Fix the build.
At least temporarily, since I've got the impression that _real_ fix
is more complex...

rafal

16 years agor2779: Some fixes to pam_winbind.c.:
Andrew Bartlett [Sat, 2 Oct 2004 11:11:49 +0000 (11:11 +0000)]
r2779: Some fixes to pam_winbind.c.:

Allow 'require_membership_of' and 'require-membership-of'.

Really use a different struct for the SID->Name lookup.

Andrew Bartlett

16 years agor2778: merged the new samba4 ms_fnmatch code to Samba3. Thanks to Rusty
Andrew Tridgell [Sat, 2 Oct 2004 09:14:43 +0000 (09:14 +0000)]
r2778: merged the new samba4 ms_fnmatch code to Samba3. Thanks to Rusty
Russel for some help in designing the new algorithm.

16 years agor2772: Check correct string length when verifying password-policies. Do not
Günther Deschner [Sat, 2 Oct 2004 00:13:21 +0000 (00:13 +0000)]
r2772: Check correct string length when verifying password-policies. Do not
allow e.g. two umlauts and one ascii char to comply with account-policy
"min password length" of 5.

Thanks to Uwe Morgenroth from CC Compunet and Volker.

TODO: we do check the length against AP_MIN_PASSWORD_LEN *and*
lp_min_passwd_length() - both can have differing values.

16 years agor2771: Second (and last) part of Swat-i18n-Patch from Björn Jacke
Günther Deschner [Fri, 1 Oct 2004 22:24:44 +0000 (22:24 +0000)]
r2771: Second (and last) part of Swat-i18n-Patch from Björn Jacke
<bjacke@sernet.de>

"Do not use display charset for swat output. In HTML we do not care
about the "locale charmap" because HTML code is UTF-8 only now.
Additionally take care that we convert files from statuspage from unix
charset to UTF-8. Thus we have correct HTML output under all
circumstances. We now also convert the share names correctly from unix
encoding to web encoding and vice vera.  "

Guenther

16 years agor2770: oops; internal_resolve_name() should stay static in 3.0
Gerald Carter [Fri, 1 Oct 2004 20:52:50 +0000 (20:52 +0000)]
r2770: oops; internal_resolve_name() should stay static in 3.0

16 years agor2768: BUG 1519: save the hostname used in the open_printer_ex() for later reuse...
Gerald Carter [Fri, 1 Oct 2004 20:34:12 +0000 (20:34 +0000)]
r2768: BUG 1519: save the hostname used in the open_printer_ex() for later reuse when filling in the spolss replies (also gets rid of get_called_name()

16 years agor2762: Remove silly conversion to and from UTF8 on the winbind pipe. Fix the
Andrew Bartlett [Fri, 1 Oct 2004 03:28:39 +0000 (03:28 +0000)]
r2762: Remove silly conversion to and from UTF8 on the winbind pipe.  Fix the
naming of the require_membership_of parameter in pam_winbind and fix
the error code for 'you didn't specify a domain' in ntlm_auth.

Andrew Bartlett

16 years agor2761: Print the decrypted, not encrypted key.
Andrew Bartlett [Fri, 1 Oct 2004 03:14:57 +0000 (03:14 +0000)]
r2761: Print the decrypted, not encrypted key.

Andrew Bartlett

16 years agor2760: Another patch from The Written Word. Don't declare function prototypes
Tim Potter [Fri, 1 Oct 2004 02:59:43 +0000 (02:59 +0000)]
r2760: Another patch from The Written Word.  Don't declare function prototypes
inside a function.  Bugzilla #1762.

16 years agor2759: Fix for winbindd on AIX 5.1. Apparently it doesn't have as many methods
Tim Potter [Fri, 1 Oct 2004 02:57:10 +0000 (02:57 +0000)]
r2759: Fix for winbindd on AIX 5.1.  Apparently it doesn't have as many methods
in struct secmethod_table as AIX 5.2.  Patch from The Written Word.

16 years agor2755: Fix NTLMv2 for use with pam_winbind, the plaintext ntlm_auth modes,
Andrew Bartlett [Thu, 30 Sep 2004 00:49:41 +0000 (00:49 +0000)]
r2755: Fix NTLMv2 for use with pam_winbind, the plaintext ntlm_auth modes,
and the wbinfo -a test tool.

If 'client ntlmv2 auth' is set, then we will send an NTLMv2, rather
than an NT/LM response to the server.

Andrew Bartlett

16 years agor2753: Workaround for the (rather broken) _samr_query_useraliases rpc-call.
Günther Deschner [Wed, 29 Sep 2004 15:26:38 +0000 (15:26 +0000)]
r2753: Workaround for the (rather broken) _samr_query_useraliases rpc-call.

_samr_query_useraliases shows up with all kind of very weird memberships
(global-groups, machine-accounts, etc.). Sometimes even if there is no
alias-membership at all.

One of the biggest mistakes is to convert any unix-group the user is a
member of, into an alias by default in get_group_from_gid.

get_alias_user_groups should be rewritten to use
pdb_enum_alias_memberships.

Guenther

16 years agor2752: Fix the paranoia-check to ensure the ldap-attribute and the
Günther Deschner [Wed, 29 Sep 2004 13:34:41 +0000 (13:34 +0000)]
r2752: Fix the paranoia-check to ensure the ldap-attribute and the
smb.conf-parameter for samba's "algorithmic rid base" in ldapsam are
identical.

It tried to get the value of LDAP_ATTR_ALGORITHMIC_RID_BASE via
get_userattr_key2string() for a very long time now. This just can not
work because LDAP_ATTR_ALGORITHMIC_RID_BASE is neither in attrib_map_v22
nor in attrib_map_v30. Instead, get it directly from dominfo_attr_list.

Ldapsam will now correctly refuse to initialize when admins tried
manually to have differing values for "algorithmic rid base" in ldap and
smb.conf. idmap_ldap is another story...

Guenther

16 years agor2746: Fix typos in net's usage-output.
Günther Deschner [Wed, 29 Sep 2004 09:56:35 +0000 (09:56 +0000)]
r2746: Fix typos in net's usage-output.

Guenther

16 years agor2736: Fix bug in Python printerdata wraper found by Daniel Jarboe.
Tim Potter [Tue, 28 Sep 2004 21:30:54 +0000 (21:30 +0000)]
r2736: Fix bug in Python printerdata wraper found by Daniel Jarboe.

16 years agor2732: Fixed typo.
Günther Deschner [Tue, 28 Sep 2004 14:13:46 +0000 (14:13 +0000)]
r2732: Fixed typo.

Guenther

16 years agor2729: Fix ldapsam_compat homeDrive. Thanks to jason@env.leeds.ac.uk
Volker Lendecke [Tue, 28 Sep 2004 13:24:38 +0000 (13:24 +0000)]
r2729: Fix ldapsam_compat homeDrive. Thanks to jason@env.leeds.ac.uk

Volker

16 years agor2715: Swat-i18n-Patch from Björn Jacke <bjacke@sernet.de>:
Günther Deschner [Tue, 28 Sep 2004 10:46:40 +0000 (10:46 +0000)]
r2715: Swat-i18n-Patch from Björn Jacke <bjacke@sernet.de>:

"Unify charset-handling in Content-Type:-headers to UTF-8 (fixes #1766),
making UTF-8 the only supported (but fully functional) charset.  Also
remove html-header-files for ja and tr that have become obsolete now.
Reformat msgstr in msg-files to UTF-8."

Guenther

16 years agor2708: Don't look for bzero, we don't use it.
Jeremy Allison [Tue, 28 Sep 2004 05:33:30 +0000 (05:33 +0000)]
r2708: Don't look for bzero, we don't use it.
Samba4 torture tester treates maxentries = 0 as maxentries ==1.
Jeremy.

16 years agor2706: The code that prints the remote MAC address (from the statistics section
Christopher R. Hertel [Tue, 28 Sep 2004 02:13:53 +0000 (02:13 +0000)]
r2706: The code that prints the remote MAC address (from the statistics section
of the response packet) was outside of the if() that determined whether
the query was successful or not.  As a result, the MAC address would print
out even if there was no MAC address.  At least the garbage it printed was
relatively consistent.  :)

I moved the MAC printing into the if() block and added an 'else' to print
"No reply..." if the query fails.

Chris -)-----