samba.git
21 years agoCrash fixes:
Andrew Bartlett [Wed, 15 Jan 2003 05:55:22 +0000 (05:55 +0000)]
Crash fixes:
 - fix a crash when a second NTLMSSP session tried to free the first
 - fix a crash due to some NULL pointers in the Add Printer Wizard (or read
   printer code too it appears).  As far as I can tell it's just that the
   GUID just might not exist.

Andrew Bartlett

21 years agoOne more from waitpid() audit: Correctly interpret waitstatus when a
Martin Pool [Wed, 15 Jan 2003 05:27:33 +0000 (05:27 +0000)]
One more from waitpid() audit: Correctly interpret waitstatus when a
child fails.  Possibly this was causing smbmount to exit(0) when it
really should have indicated a failure.

21 years agoRun autoheader
Martin Pool [Wed, 15 Jan 2003 05:23:40 +0000 (05:23 +0000)]
Run autoheader

21 years agoFix bug where the existence of config.cache would cause functions like
Martin Pool [Wed, 15 Jan 2003 05:00:04 +0000 (05:00 +0000)]
Fix bug where the existence of config.cache would cause functions like
yp_get_default_domain to be misdetected.

According to the autoconf manual we need to check for libraries before
looking for functions in them.

21 years agoIRIX uses libns_winbind.so instead of libnss_winbind.so
Herb Lewis [Wed, 15 Jan 2003 01:05:35 +0000 (01:05 +0000)]
IRIX uses libns_winbind.so instead of libnss_winbind.so

21 years ago[merge] removing unused variable and unused file
Gerald Carter [Tue, 14 Jan 2003 23:55:14 +0000 (23:55 +0000)]
[merge] removing unused variable and unused file

21 years agoFix some debug lines, and add a bit more info to help track down ldap
Andrew Bartlett [Tue, 14 Jan 2003 23:45:14 +0000 (23:45 +0000)]
Fix some debug lines, and add a bit more info to help track down ldap
connectivity problems.

Andrew Bartlett

21 years agomerge: disable [u|g]idPool objects
Gerald Carter [Tue, 14 Jan 2003 16:03:27 +0000 (16:03 +0000)]
merge: disable [u|g]idPool objects

21 years agomake -i flag work lik eit did in 2.2
Gerald Carter [Tue, 14 Jan 2003 15:52:45 +0000 (15:52 +0000)]
make -i flag work lik eit did in 2.2

21 years agoCommit my (disabled) fstring/pstring mixup detection code.
Andrew Bartlett [Tue, 14 Jan 2003 06:53:35 +0000 (06:53 +0000)]
Commit my (disabled) fstring/pstring mixup detection code.

We need to fix some 'overmalloc' cases before it can be enabled by default.

Andrew Bartlett

21 years agoThe last of the fstring/pstring mixups, and an extra 'static'.
Andrew Bartlett [Tue, 14 Jan 2003 06:46:14 +0000 (06:46 +0000)]
The last of the fstring/pstring mixups, and an extra 'static'.

21 years agoA couple more pstcpy/fstrcpy mixups - doing an fstrcpy into a pstring is
Andrew Bartlett [Tue, 14 Jan 2003 06:36:40 +0000 (06:36 +0000)]
A couple more pstcpy/fstrcpy mixups - doing an fstrcpy into a pstring is
harmless, but breaks my automated check.

21 years agoAdded comment about a SMB_ASSERT()
Tim Potter [Tue, 14 Jan 2003 05:25:08 +0000 (05:25 +0000)]
Added comment about a SMB_ASSERT()

21 years agoRemove wrapper function _get_trust_account_password() as it doesn't
Tim Potter [Tue, 14 Jan 2003 05:08:42 +0000 (05:08 +0000)]
Remove wrapper function _get_trust_account_password() as it doesn't
seem to do anything useful anymore other than call
secrets_fetch_trust_account_password().

21 years agoAdded a comment in case anyone is thinking of doing a SMB_ASSERT in winbindd.
Tim Potter [Tue, 14 Jan 2003 04:04:13 +0000 (04:04 +0000)]
Added a comment in case anyone is thinking of doing a SMB_ASSERT in winbindd.

21 years agoquery_alt_name takes a forced unicode string in win2000. It is not
Andrew Tridgell [Tue, 14 Jan 2003 02:51:37 +0000 (02:51 +0000)]
query_alt_name takes a forced unicode string in win2000. It is not
null terminated.

Thanks to Metze for finding this.

21 years agoclearer debug message when the user is already in the ldap db
Andrew Tridgell [Tue, 14 Jan 2003 00:39:18 +0000 (00:39 +0000)]
clearer debug message when the user is already in the ldap db

21 years agoadd help text for 'net ads lookup'
Andrew Tridgell [Tue, 14 Jan 2003 00:35:38 +0000 (00:35 +0000)]
add help text for 'net ads lookup'

21 years agoMissing indirect in final free.
Jeremy Allison [Mon, 13 Jan 2003 23:07:28 +0000 (23:07 +0000)]
Missing indirect in final free.
Jeremy.

21 years agoAlways initialise this variable - and don't set the 'must change now' if it was
Andrew Bartlett [Mon, 13 Jan 2003 21:49:49 +0000 (21:49 +0000)]
Always initialise this variable - and don't set the 'must change now' if it was
last changed at '0'.

We need to actually change this password sometime...

Andrew Bartlett

21 years agoFix typo, and the build.
Andrew Bartlett [Mon, 13 Jan 2003 21:48:52 +0000 (21:48 +0000)]
Fix typo, and the build.

21 years agoMake sure that those cleanups actually went in.
Richard Sharpe [Mon, 13 Jan 2003 20:30:28 +0000 (20:30 +0000)]
Make sure that those cleanups actually went in.

21 years agoPatch from metze to add what he feels is the correct semantics for a Domain
Andrew Bartlett [Mon, 13 Jan 2003 13:11:36 +0000 (13:11 +0000)]
Patch from metze to add what he feels is the correct semantics for a Domain
Controller.  As we have had a number of attempts at this over the last little
while, I need to get my test rig going, and give this whole area a poke...

Meanwhile, if you want to use this, just adjust your 'auth methods' line to use
samstrict_dc...

Andrew Bartlett

21 years agoPatch from metze to to make testparm show values for 'workgroup', 'netbios
Andrew Bartlett [Mon, 13 Jan 2003 13:03:25 +0000 (13:03 +0000)]
Patch from metze to to make testparm show values for 'workgroup', 'netbios
name' and 'netbios scope'.  Probably has a similar effect on SWAT.

Also adds '-V' to testparm.

Andrew Bartlett

21 years agoPatch from ab, to make findsmb omit -r most of the time.
Andrew Bartlett [Mon, 13 Jan 2003 12:54:59 +0000 (12:54 +0000)]
Patch from ab, to make findsmb omit -r most of the time.

(original mail follows)
--
From: Alexander Bokovoy <a.bokovoy@sam-solutions.net>
Greetings!

I propose to make an optional argument '-r' for findsmb utility to pass
'-r' option to nmblookup optionally and default to omit it. The reason is
to have findsmb working more user-friendly in contemporary office
environments where Windows 95/98 boxes are rare and also to allow usage of
the utility under non-priviledged accounts when Windows95/98 boxes are in
minority or even absent in network neighbourhood.

An attached patch provides this change for HEAD together with
documentation changes.

--
/ Alexander Bokovoy

21 years agoUpdates to our NTLMSSP code:
Andrew Bartlett [Mon, 13 Jan 2003 12:48:37 +0000 (12:48 +0000)]
Updates to our NTLMSSP code:

This tries to extract our server-side code out of sessetup.c, and into a more
general lib.  I hope this is only a temporay resting place - I indend to
refactor it again into an auth-subsystem independent lib, using callbacks.

Move some of our our NTLMSSP #defines into a new file, and add two that I found
in the COMsource docs - we seem to have a double-up, but I've verified from
traces that the NTLMSSP_TARGET_TYPE_{DOMAIN,SERVER} is real.

This code also copes with ASCII clients - not that we will ever see any here,
but I hope to use this for HTTP, were we can get them.  Win2k authenticates
fine under forced ASCII, btw.

Tested with Win2k, NTLMv2 and Samba's smbclient.

Andrew Bartlett

21 years agoAccessing data after it's been free()ed really is a no-no...
Andrew Bartlett [Mon, 13 Jan 2003 12:42:20 +0000 (12:42 +0000)]
Accessing data after it's been free()ed really is a no-no...

Andrew Bartlett

21 years agoFix to debian bug #171071 - we had the wrong dereference on the pointer to be
Andrew Bartlett [Mon, 13 Jan 2003 12:21:07 +0000 (12:21 +0000)]
Fix to debian bug #171071 - we had the wrong dereference on the pointer to be
Realloc()ed, causing it to fail.

Big thanks to Sandor Sonfeld <sonf@linuxmail.org> for the debug, stack and
valgrind traces!

Andrew Bartlett

21 years agoDon't force the DOS password into a 14 char space, as this would imply null
Andrew Bartlett [Mon, 13 Jan 2003 08:53:54 +0000 (08:53 +0000)]
Don't force the DOS password into a 14 char space, as this would imply null
termination - the password will not be null terminated before hashing if
len >= 14.

related to debian bug #157432

Andrew Bartlett

21 years agoEnsure we do not pass uninitialised data to the kernel. (Picked up by
Andrew Bartlett [Mon, 13 Jan 2003 08:09:32 +0000 (08:09 +0000)]
Ensure we do not pass uninitialised data to the kernel.  (Picked up by
valgrind).

Andrew Bartlett

21 years agoFix typo in comment.
Martin Pool [Mon, 13 Jan 2003 04:57:21 +0000 (04:57 +0000)]
Fix typo in comment.

21 years agoMerge of remove "winbindd holding pattern" from appliance. It's a bit
Tim Potter [Mon, 13 Jan 2003 03:45:31 +0000 (03:45 +0000)]
Merge ... remove "winbindd holding pattern" from appliance.  It's a bit
simpler as the rescanning of trusted domains helps us out a bit.

21 years agoKeep all the const warnings in one place, by adding a utility function to
Andrew Bartlett [Sun, 12 Jan 2003 12:23:37 +0000 (12:23 +0000)]
Keep all the const warnings in one place, by adding a utility function to
make the TDB_DATA.

Andrew Bartlett

21 years agoOops, this is the change to use an fstring for the incoming service buffer -
Andrew Bartlett [Sun, 12 Jan 2003 11:31:03 +0000 (11:31 +0000)]
Oops, this is the change to use an fstring for the incoming service buffer -
the commit to reply.c just matches a pstrcpy for the pstring.  (harmless, fixes
it for the automated test).

Andrew Bartlett

21 years agoMake the 'service' in make_connection() use an fstrcpy(), and an fstring,
Andrew Bartlett [Sun, 12 Jan 2003 11:29:02 +0000 (11:29 +0000)]
Make the 'service' in make_connection() use an fstrcpy(), and an fstring,
becouse that is what it's input (reply_tcon_and_x) uses, and becouse we really
don't want supprises for service names.

Also remove a legacy #define, in favor of the lp_ equiv.

Andrew Bartlett

21 years agoWhile the usage is most bisarre, this fixes up another fstring/pstring mixup.
Andrew Bartlett [Sat, 11 Jan 2003 12:33:00 +0000 (12:33 +0000)]
While the usage is most bisarre, this fixes up another fstring/pstring mixup.
(In this case, an fstrcpy() into a pstring, ie harmless).

Andrew Bartlett

21 years agoUse size_t for the counter vars, to match the type they are assigned from
Andrew Bartlett [Sat, 11 Jan 2003 12:24:19 +0000 (12:24 +0000)]
Use size_t for the counter vars, to match the type they are assigned from
(signed/unsigned mixup).

Andrew Bartlett

21 years agoA couple more fstring/pstring issues - and move from sizeof(fstring) to
Andrew Bartlett [Sat, 11 Jan 2003 12:18:57 +0000 (12:18 +0000)]
A couple more fstring/pstring issues - and move from sizeof(fstring) to
sizeof(var) in slprintf statements, so as to ensure we don't get mixups there.

Also remove fstrcpy() into a variable that is about to be printed into anyway.

Andrew Bartlett

21 years agoRemove an unused paramter for our old LM-only password change code, and fix a
Andrew Bartlett [Sat, 11 Jan 2003 12:13:03 +0000 (12:13 +0000)]
Remove an unused paramter for our old LM-only password change code, and fix a
(harmless) fstring/pstring mixup.

Andrew Bartlett

21 years agoFix a number of client-side fstring/pstring mixups.
Andrew Bartlett [Sat, 11 Jan 2003 12:04:14 +0000 (12:04 +0000)]
Fix a number of client-side fstring/pstring mixups.

Andrew Bartlett

21 years agoPatch from Nik Conwell <nik@bu.edu>. Don't reference free()ed data when trying
Andrew Bartlett [Sat, 11 Jan 2003 03:29:31 +0000 (03:29 +0000)]
Patch from Nik Conwell <nik@bu.edu>.  Don't reference free()ed data when trying
to figure out if we have got our ticket yet.

Andrew Bartlett

21 years agoReturn the 'freindly' NT error message if at all possible.
Andrew Bartlett [Sat, 11 Jan 2003 03:25:31 +0000 (03:25 +0000)]
Return the 'freindly' NT error message if at all possible.

Andrew Bartlett

21 years agoNo point having this for both parts of the 'if' statement.
Andrew Bartlett [Sat, 11 Jan 2003 03:23:16 +0000 (03:23 +0000)]
No point having this for both parts of the 'if' statement.

Andrew Bartlett

21 years agotypo from merge
Gerald Carter [Sat, 11 Jan 2003 02:36:13 +0000 (02:36 +0000)]
typo from merge

21 years agofix tdb_append() prototype
Gerald Carter [Sat, 11 Jan 2003 02:30:34 +0000 (02:30 +0000)]
fix tdb_append() prototype

21 years ago[merge] make sure to updatre print queue cache during timeout_processing() to send...
Gerald Carter [Sat, 11 Jan 2003 02:29:23 +0000 (02:29 +0000)]
[merge] make sure to updatre print queue cache during timeout_processing() to send notify events; CR 1491

21 years agoAdded new message_send_pid() code that uses tdb append to reduce locking
Jeremy Allison [Sat, 11 Jan 2003 00:17:37 +0000 (00:17 +0000)]
Added new message_send_pid() code that uses tdb append to reduce locking
contention on the messaging tdb.
Jeremy.

21 years agoAdded tdb_append() call. Efficiently adds to an entry. Used by new messaging
Jeremy Allison [Sat, 11 Jan 2003 00:07:40 +0000 (00:07 +0000)]
Added tdb_append() call. Efficiently adds to an entry. Used by new messaging
code. Also added torture tests for it.
Jeremy.

21 years agoFirst part of efficiency fixes for message sending to pid's (cutting down
Jeremy Allison [Fri, 10 Jan 2003 20:17:06 +0000 (20:17 +0000)]
First part of efficiency fixes for message sending to pid's (cutting down
the amount of time we hold tdb locks). Gulp down all messages at once rather
than reading/re-writing one at a time. NOTE: All dispatch routines *must*
be able to cope with incoming message on *odd* byte boundaries (all current
handlers do).
Jeremy.

21 years agoFix lpq_cache time check (missed from yesterday).
Jeremy Allison [Fri, 10 Jan 2003 19:59:18 +0000 (19:59 +0000)]
Fix lpq_cache time check (missed from yesterday).
Jeremy.

21 years agoa 0 length printer data value is not a memory allocation error; fix CR601
Gerald Carter [Thu, 9 Jan 2003 19:51:28 +0000 (19:51 +0000)]
a 0 length printer data value is not a memory allocation error; fix CR601

21 years agotdb_chainlock_with_timeout: Add TODO: If we time out waiting for a
Martin Pool [Thu, 9 Jan 2003 07:23:23 +0000 (07:23 +0000)]
tdb_chainlock_with_timeout: Add TODO: If we time out waiting for a
lock, it might be nice to use F_GETLK to get the pid of the process
currently holding the lock and print that as part of the debugging
message.

I'd like to have this in appliance_head, but the code is too different
so I won't worry for now.

21 years agoRemove read_with_timeout() and replaced its only caller (the passwd chat
Tim Potter [Thu, 9 Jan 2003 06:58:07 +0000 (06:58 +0000)]
Remove read_with_timeout() and replaced its only caller (the passwd chat
stuff) with a call to read_socket_with_timeout() which does the same thing.

Passwd chat still works but I couldn't figure out the right arguments to
passwd chat to get it to work right but data was definitely getting
through.

21 years agoRemove dead function.
Tim Potter [Thu, 9 Jan 2003 05:50:15 +0000 (05:50 +0000)]
Remove dead function.

21 years agoMoved smbd process counting functions from smbd/server.c to smbd/process.c
Tim Potter [Thu, 9 Jan 2003 05:49:51 +0000 (05:49 +0000)]
Moved smbd process counting functions from smbd/server.c to smbd/process.c
so that bin/vfstest can link.

21 years agoCorrect misleading comment: winbindd_lookupname gets the SID for a
Martin Pool [Thu, 9 Jan 2003 01:16:12 +0000 (01:16 +0000)]
Correct misleading comment: winbindd_lookupname gets the SID for a
name, not vice versa.

21 years agoEnsure we return disk full by default on short writes.
Jeremy Allison [Wed, 8 Jan 2003 23:49:37 +0000 (23:49 +0000)]
Ensure we return disk full by default on short writes.
Jeremy.

21 years agoEnsure we don't get an invalid number for total smbd's if the tdb update
Jeremy Allison [Wed, 8 Jan 2003 21:42:43 +0000 (21:42 +0000)]
Ensure we don't get an invalid number for total smbd's if the tdb update
fails.
Jeremy.

21 years agoreran autoconf
Herb Lewis [Wed, 8 Jan 2003 18:33:16 +0000 (18:33 +0000)]
reran autoconf

21 years agoSome fixes to editreg. Clientgen should not be touched, and an update to configure...
Richard Sharpe [Wed, 8 Jan 2003 17:59:52 +0000 (17:59 +0000)]
Some fixes to editreg. Clientgen should not be touched, and an update to configure that Herb needs.

21 years agoFix erroneous free of SAM_ACCOUNT (make clearer when allocated
Jeremy Allison [Wed, 8 Jan 2003 07:02:18 +0000 (07:02 +0000)]
Fix erroneous free of SAM_ACCOUNT (make clearer when allocated
memory is being returned).
Jeremy

21 years agoRemoved pointless null statement in signal handler.
Tim Potter [Wed, 8 Jan 2003 04:14:12 +0000 (04:14 +0000)]
Removed pointless null statement in signal handler.

21 years agoDisplay an error message when the server cannot be resolved when running at
Tim Potter [Wed, 8 Jan 2003 02:54:42 +0000 (02:54 +0000)]
Display an error message when the server cannot be resolved when running at
debug level 0.

21 years agoFix open problem with changing attributes on an existing file - based
Jeremy Allison [Wed, 8 Jan 2003 02:09:14 +0000 (02:09 +0000)]
Fix open problem with changing attributes on an existing file - based
on work by  <steve@griffin.sio2.nl>.
Jeremy.

21 years agoFix memory leak. Thanks, Herb!
Volker Lendecke [Tue, 7 Jan 2003 20:54:31 +0000 (20:54 +0000)]
Fix memory leak. Thanks, Herb!

Volker

21 years agothis makes it compile, but I'm not sure if it is correct. Are we really
Herb Lewis [Tue, 7 Jan 2003 17:55:14 +0000 (17:55 +0000)]
this makes it compile, but I'm not sure if it is correct. Are we really
testing a pointer value to be equal to 0x19 ???

21 years agocannot make assignments to const values.
Herb Lewis [Tue, 7 Jan 2003 17:43:37 +0000 (17:43 +0000)]
cannot make assignments to const values.

21 years agoadd and escape before $( in regex string so it is not interpreted as the
Herb Lewis [Tue, 7 Jan 2003 17:33:30 +0000 (17:33 +0000)]
add and escape before $( in regex string so it is not interpreted as the
perl variable for "real gid of this process"

21 years agoCreate a machine account correctly if the unix account exists.
Volker Lendecke [Tue, 7 Jan 2003 10:38:33 +0000 (10:38 +0000)]
Create a machine account correctly if the unix account exists.

With this, my test for W2kSP3 join after pdbedit -a -m works fine.

Volker

21 years agoAdded code to the handling of the -M option to ensure that we default to
Christopher R. Hertel [Tue, 7 Jan 2003 04:33:43 +0000 (04:33 +0000)]
Added code to the handling of the -M option to ensure that we default to
port 139 for that option.  The NetBIOS Messenger Service doesn't run on
a non-NBT port.

21 years agoRename a local variable to avoid a warning.
Tim Potter [Tue, 7 Jan 2003 03:39:22 +0000 (03:39 +0000)]
Rename a local variable to avoid a warning.

21 years ago'index' is the name of a libc function. Rename to 'ndx' to avoid compiler
Tim Potter [Tue, 7 Jan 2003 03:36:39 +0000 (03:36 +0000)]
'index' is the name of a libc function.  Rename to 'ndx' to avoid compiler
warnings.

21 years agoRemove unused variable.
Tim Potter [Tue, 7 Jan 2003 03:31:52 +0000 (03:31 +0000)]
Remove unused variable.

21 years agoFix memory leaks in pdb_ code.
Jeremy Allison [Mon, 6 Jan 2003 21:48:43 +0000 (21:48 +0000)]
Fix memory leaks in pdb_ code.
Jeremy.

21 years agoadd $(LDFLAGS) to libsmbclient
Gerald Carter [Mon, 6 Jan 2003 18:00:18 +0000 (18:00 +0000)]
add $(LDFLAGS) to libsmbclient

21 years agoFix from "Ken Cross" <kcross@nssolutions.com> for enumerating
Jeremy Allison [Mon, 6 Jan 2003 07:29:34 +0000 (07:29 +0000)]
Fix from "Ken Cross" <kcross@nssolutions.com> for enumerating
groups.
Jeremy.

21 years agoFix a segfault when we don't correctly load a VFS module (don't keep it in
Andrew Bartlett [Mon, 6 Jan 2003 06:15:34 +0000 (06:15 +0000)]
Fix a segfault when we don't correctly load a VFS module (don't keep it in
the loaded list on error).  Also change some of the error returns, becouse
NT_STATUS_UNSUCCESSFUL gives a most useless error message on the client.

As for which error, my logic is that a share without a valid VFS module is
not a valid share, and therefore should return the same error as a non-existant
share.

Andrew Bartlett

21 years agoClear up the auth_sam password checking code (the core of our password checking
Andrew Bartlett [Sun, 5 Jan 2003 07:32:08 +0000 (07:32 +0000)]
Clear up the auth_sam password checking code (the core of our password checking
routines).  In particular, we now better support the NT# in LM feild, and the
LMv2 password scheme.  (LMv2 is basicly NTLMv2 capped at 24 bytes, slightly
more secure, and in the LM feild for compatiblity).

Thanks to the Samba-TNG team and Luke Leighton for various descriptions of this
algorithm, and to MS for a solution that seems to actually make sense for once
:-).

Andrew Bartlett

21 years agoEven when I can't manage to do QA, at least the build farm will catch some of
Andrew Bartlett [Sun, 5 Jan 2003 04:59:42 +0000 (04:59 +0000)]
Even when I can't manage to do QA, at least the build farm will catch some of
the silly stuff.

(fixes password changes from RAP based clients like smbpasswd)

Andrew Bartlett

21 years agoFix problem with "hide unreadable". stat file opens are baaack :-).
Jeremy Allison [Fri, 3 Jan 2003 19:07:26 +0000 (19:07 +0000)]
Fix problem with "hide unreadable". stat file opens are baaack :-).
Jeremy.

21 years agoregenerate man pages
Gerald Carter [Fri, 3 Jan 2003 17:35:18 +0000 (17:35 +0000)]
regenerate man pages

21 years agopatch to include support for daemontools from Michael Handler
Gerald Carter [Fri, 3 Jan 2003 17:32:11 +0000 (17:32 +0000)]
patch to include support for daemontools from Michael Handler

21 years agomake sure we create the PIDDIR and LOCKDIR from installdirs
Gerald Carter [Fri, 3 Jan 2003 15:31:46 +0000 (15:31 +0000)]
make sure we create the PIDDIR and LOCKDIR from installdirs

21 years agooops. Use -r, not -f
Gerald Carter [Fri, 3 Jan 2003 15:21:01 +0000 (15:21 +0000)]
oops.  Use -r, not -f

21 years agopatch from Andrew Bird to correct a few file/directory checks when building
Gerald Carter [Fri, 3 Jan 2003 15:19:51 +0000 (15:19 +0000)]
patch from Andrew Bird to correct a few file/directory checks when building

21 years agoFixed some simple typos, including one that would have resulted in a
Christopher R. Hertel [Fri, 3 Jan 2003 04:32:23 +0000 (04:32 +0000)]
Fixed some simple typos, including one that would have resulted in a
DEBUG() message printing the wrong value.

21 years agopstring/fstring missmatches found by Andrew Bartlett.
Jeremy Allison [Thu, 2 Jan 2003 20:43:27 +0000 (20:43 +0000)]
pstring/fstring missmatches found by Andrew Bartlett.
Jeremy.

21 years agoWITH_SENDFILE #ifdef merged from SAMBA_2_2
Gerald Carter [Thu, 2 Jan 2003 18:20:42 +0000 (18:20 +0000)]
WITH_SENDFILE #ifdef merged from SAMBA_2_2

21 years agoWe still have 'ldap server' and 'ldap port'. Could somebody please
Volker Lendecke [Thu, 2 Jan 2003 16:12:39 +0000 (16:12 +0000)]
We still have 'ldap server' and 'ldap port'. Could somebody please
regen the docs?

Thanks,

Volker

21 years agoWe already have one function to move unistr2 -> multibyte-static, so we
Andrew Bartlett [Thu, 2 Jan 2003 13:10:25 +0000 (13:10 +0000)]
We already have one function to move unistr2 -> multibyte-static, so we
don't need a second just for pdb.

Also, remove magic 'is lp_guest_account' test - the magic RID should be
up to the passdb backend to set.

Andrew Bartlett

21 years agoBecouse these functions return a size_t, they can't return negitive numbers.
Andrew Bartlett [Thu, 2 Jan 2003 13:01:44 +0000 (13:01 +0000)]
Becouse these functions return a size_t, they can't return negitive numbers.
So a < 0 check is pointless, instead check against -1, which will be cast to
unsigned.

Andrew Bartlett

21 years agodebugparse is a seperate utility, no need to include it in the main ubiqx
Andrew Bartlett [Thu, 2 Jan 2003 12:58:23 +0000 (12:58 +0000)]
debugparse is a seperate utility, no need to include it in the main ubiqx
objects.

Andrew Bartlett

21 years agoAdd PRINTF_ATTRIBUTE() to a few more printf() style functions. Aids in
Andrew Bartlett [Thu, 2 Jan 2003 12:54:02 +0000 (12:54 +0000)]
Add PRINTF_ATTRIBUTE() to a few more printf() style functions.  Aids in
compiler-based argument checking.

21 years agoCall me parinoid, but I don't like the idea that we could ever have dbf point
Andrew Bartlett [Thu, 2 Jan 2003 12:53:04 +0000 (12:53 +0000)]
Call me parinoid, but I don't like the idea that we could ever have dbf point
to a closed file.  Swap the assign and the close around.

Andrew Bartlett

21 years agoAdd a dash of static.
Andrew Bartlett [Thu, 2 Jan 2003 12:39:36 +0000 (12:39 +0000)]
Add a dash of static.

21 years agoOne more const. Andrew, you seem to have krb5 :-)
Volker Lendecke [Thu, 2 Jan 2003 09:40:10 +0000 (09:40 +0000)]
One more const. Andrew, you seem to have krb5 :-)

Volker

21 years agoBIG patch...
Andrew Bartlett [Thu, 2 Jan 2003 09:07:17 +0000 (09:07 +0000)]
BIG patch...

This patch makes Samba compile cleanly with -Wwrite-strings.
 - That is, all string literals are marked as 'const'.  These strings are
always read only, this just marks them as such for passing to other functions.

What is most supprising is that I didn't need to change more than a few lines of code (all
in 'net', which got a small cleanup of net.h and extern variables).  The rest
is just adding a lot of 'const'.

As far as I can tell, I have not added any new warnings - apart from making all
of tdbutil.c's function const (so they warn for adding that const string to
struct).

Andrew Bartlett

21 years agoDocument the size constraints on a write request.
Richard Sharpe [Wed, 1 Jan 2003 21:54:00 +0000 (21:54 +0000)]
Document the size constraints on a write request.

21 years agoMove our password change code along a little - use NTSTATUS, and implmenet
Andrew Bartlett [Wed, 1 Jan 2003 04:19:34 +0000 (04:19 +0000)]
Move our password change code along a little - use NTSTATUS, and implmenet
minimum password age and min password length for all password changes.

Andrew Bartlett

21 years agofix for bad check spotted by Ray Simard <ray@sylvan-glade.com>
Simo Sorce [Tue, 31 Dec 2002 10:23:37 +0000 (10:23 +0000)]
fix for bad check spotted by Ray Simard <ray@sylvan-glade.com>