Herb Lewis [Fri, 6 Apr 2001 18:37:55 +0000 (18:37 +0000)]
keep cflags from being duplicated if defined in environment
Jeremy Allison [Fri, 6 Apr 2001 18:25:17 +0000 (18:25 +0000)]
Implemented stub function for NET_SHARE_DELETE. Now to implement the real
internals to support server manager.
Jeremy
Jeremy Allison [Fri, 6 Apr 2001 17:41:47 +0000 (17:41 +0000)]
Added stub function for NET_SHARE_ADD. Once this is implemented to call a hook
function (same for NET_SHARE_DELETE and NET_SHARE_SET) we will be able to manage
the shares section in smb.conf via NT server manager........
This should enhance the friendliness of Samba in NT-only shops by an order of
magnitude.
Jeremy.
Jeremy Allison [Fri, 6 Apr 2001 17:13:16 +0000 (17:13 +0000)]
If we can't open the username map file, at least print the strerror.
Jeremy.
Jeremy Allison [Fri, 6 Apr 2001 02:12:07 +0000 (02:12 +0000)]
NT sucks ! :-). We have to prefix the UNIX path with a C: in order to pass the
server manager tests :-). Ensure we don't set a return level on set as
server manager barfs....
Jeremy.
Jeremy Allison [Fri, 6 Apr 2001 01:39:12 +0000 (01:39 +0000)]
We can now use server manager to look at Samba shares. NT still expects a ":"
in the path though.... I'm looking into it.
Jeremy.
Jeremy Allison [Thu, 5 Apr 2001 21:31:12 +0000 (21:31 +0000)]
Setup syslog correctly in swat. Patch from Andrew Bartlett.
Jeremy.
Jeremy Allison [Thu, 5 Apr 2001 20:52:02 +0000 (20:52 +0000)]
Fix from Michael Davidson <md@sco.COM> to merge Solaris and UnixWare ACLs.
Jeremy.
Jeremy Allison [Thu, 5 Apr 2001 20:46:15 +0000 (20:46 +0000)]
Fix from "Romeril, Alan" <a.romeril@ic.ac.uk> to get his NFS quota code
to work on Solaris 2.6.
Jeremy.
Jeremy Allison [Thu, 5 Apr 2001 19:17:54 +0000 (19:17 +0000)]
Tidy up tmp file handling.
Jeremy.
Jeremy Allison [Wed, 4 Apr 2001 23:42:17 +0000 (23:42 +0000)]
I know we're supposed to be feature frozen, but I couldn't resist this... :-).
I worked out and added the per-share get/set RPCs for security descriptors.
Currently this code returns Everyone, full access on get and permission denied
on set, but backending this with a tdb and checking it on tconX (to give full NT
semantics for security on shares) is now an excersise for the reader... :-).
Jeremy.
Jeremy Allison [Wed, 4 Apr 2001 17:20:57 +0000 (17:20 +0000)]
Fix for broken RelientUNIX.
Jeremy.
Jeremy Allison [Tue, 3 Apr 2001 23:09:04 +0000 (23:09 +0000)]
Missed forms update - my CVS mistake.
Jeremy.
Andrew Tridgell [Tue, 3 Apr 2001 22:52:38 +0000 (22:52 +0000)]
test commit for jeremy
Jeremy Allison [Tue, 3 Apr 2001 22:41:31 +0000 (22:41 +0000)]
Sync up with new NT forms code.
Jeremy.
Jeremy Allison [Tue, 3 Apr 2001 00:40:01 +0000 (00:40 +0000)]
Added XFS ACLs on Linux. Code from John Trostel <jtrostel@connex.com>.
Jeremy.
Jeremy Allison [Mon, 2 Apr 2001 23:05:25 +0000 (23:05 +0000)]
IRIX ACLs from Herb.
Jeremy.
Jeremy Allison [Mon, 2 Apr 2001 22:27:40 +0000 (22:27 +0000)]
include/ntdomain.h:
rpc_server/srv_lsa_hnd.c: Remove back pointer from policy handle list as the pipe
that opened the handle may have been closed. We were dereferencing
into something that had been closed.
rpc_parse/parse_spoolss.c: Sync up with Gerald's changes in 2.2.
lib/replace.c: Don't do proto on setlinebuf as it differs between systems.
Jeremy.
Jeremy Allison [Sun, 1 Apr 2001 19:00:43 +0000 (19:00 +0000)]
Integrated solaris nfs quota code from Alan Romeril <a.romeril@ic.ac.uk>
Jeremy
Jeremy Allison [Sat, 31 Mar 2001 19:06:45 +0000 (19:06 +0000)]
configure configure.in include/config.h.in lib/replace.c: Added test and replacement
for setlinebuf which apparantly doesn't exist on HPUX 11.
include/byteorder.h:
rpc_parse/parse_prs.c:
Ding Dong the witch is dead ! :-). Ok, I'm happy 'cos I've finally deleted
all the *HORRIBLE* DBG_RW_XXX and RW_XXX macros from include/byteorder.h.
They were macros that included macros that had conditional macros included.
No one understood them (they were the cause of most of the bigendian issue
bugs). Finally, I went into parse_prs.c and inlined all of that stuff with
regular function calls. They're understandable, they're easy to edit and
they don't include macros !
JF - please look at the one comment I added (JF PLEASE CHECK). I have
tested this partly with IRIX (a bigendian system) running with AS/U on
a Solaris box in SGI's lab, and I've also confirmed these new changes
work with W2K (vmware) but there may be the odd bug lurking. Herb, if
you could re-checkout and test again with this code that would help.
Extra. Fixed bug spotted by the sharp eyes of JF - big endian unicode
packet would cause a early truncate of string parsing as we were checking for a char *
0, not a uint16 * 0.
Jeremy.
Andrew Tridgell [Sat, 31 Mar 2001 13:48:24 +0000 (13:48 +0000)]
started converting some of the only-ascii code to use srvstr_*
added srvstr_push_ascii() and srvstr_pull_ascii() as convenience
routines to replace the current usage of strncpy() like fns
for packet pull/push. We need to do this in *lots* of places
in Samba in order to get our codepage handling right
Jeremy Allison [Fri, 30 Mar 2001 21:53:33 +0000 (21:53 +0000)]
utils/torture.c: Added one more delete on close test.
smbd/notify_kernel.c: This code was wrong I believe. It was structured to only
return a changenotify event on being called from timeout processing (t != 0).
The kernel changenotify events should fire on *asynchronous* processing (EINTR
return from select caused by the realtime signal delivery) with t == 0.
Reported by Juergen Hasch (Hasch@t-online.de).
ANDREW PLEASE CHECK THIS !
Currently the hash style changenotify is done on async processing as well
as timeout processing. As this is expensive we may want to revisit doing this
and maybe set it to fire only on timeout processing.
Jeremy.
Jeremy Allison [Fri, 30 Mar 2001 18:07:43 +0000 (18:07 +0000)]
Fixed extern ref typo for file generic perms. 2am coding strikes again :-).
Jeremy.
Jeremy Allison [Fri, 30 Mar 2001 08:57:24 +0000 (08:57 +0000)]
This is a big, rather ugly patch. Whilst investigating the files not truncated
when copying to a full disk problem, I discovered that we were not allowing
the delete on close flag to be set properly, this led to other things, and
after investigation of the proper delete on close semantics and their relationship
to the file_share_delete flag I discovered there were some cases where we
weren't doing the deny modes properly. And this after only 5 years working
on them..... :-) :-).
So here's the latest attempt. I realised the delete on close flag needs to
be set across all smbds with a dev/ino pair open - in addition, the delete
on close flag, allow share delete and delete access requested all need to
be stored in the share mode tdb.
The "delete_on_close" entry in the fsp struct is now redundant and should
really be removed. This may also mean we can get rid of the "iterate_fsp"
calls that I didn't like adding in the first place. Whilst doing this patch,
I also discovered we needed to do the se_map_generic() call for file opens
and POSIX ACL mapping, so I added that also.
This code, although ugly, now passes the deny mode torture tests plus the
delete on close tests I added. I do need to add one more multiple connection
delete on close test to make sure I got the semantics exactly right, plus we
should also (as Andrew suggested) move to random testing here.
The good news is that NT should now correctly delete the file on disk
full error when copying to a disk :-).
Jeremy.
Jeremy Allison [Fri, 30 Mar 2001 06:31:22 +0000 (06:31 +0000)]
Added one more test - just to make sure flag can be unset.
Jeremy.
Jeremy Allison [Fri, 30 Mar 2001 06:22:14 +0000 (06:22 +0000)]
Expanded delete on close test to 6 cases. I now understand the semantics of
this and the FILE_SHARE_DELETE flag exactly. The bad news is it means our
share mode handling is broken (again :-).
Jeremy.
Jeremy Allison [Fri, 30 Mar 2001 02:48:06 +0000 (02:48 +0000)]
Changed to using GENERIC_XX constants instead of hex values to make
delete test explicit.
Jeremy.
Jeremy Allison [Fri, 30 Mar 2001 00:35:47 +0000 (00:35 +0000)]
Added delete on close test that has allowed me to determine how w2k
behaves in this situation.
Jeremy.
Jeremy Allison [Thu, 29 Mar 2001 02:58:47 +0000 (02:58 +0000)]
Added cli_nt_delete_on_close() call to allow flag to be set for torture tests.
Jeremy.
Jeremy Allison [Thu, 29 Mar 2001 00:58:52 +0000 (00:58 +0000)]
Added cli_nt_create_full() as a way to get at all the ntcreate parameters.
Used in smbtorture mods. Re-cast cli_nt_create() as a call to cli_nt_create_full().
Jeremy.
Jeremy Allison [Wed, 28 Mar 2001 23:20:46 +0000 (23:20 +0000)]
Insure fix. Don't read 4 bytes from data area unless there are 4 bytes to read.
Jeremy.
Jeremy Allison [Wed, 28 Mar 2001 18:12:49 +0000 (18:12 +0000)]
Fixed typo bug spotted by Jim Vopni jimv@metrics.com. Caused no status to
be returned.
Jeremy.
CVS ----------------------------------------------------------------------
Gerald Carter [Wed, 28 Mar 2001 16:08:00 +0000 (16:08 +0000)]
rename of 16 new_smb_io functions to smb_io_* for consistency sake
(merge from 2.2)
Richard Sharpe [Wed, 28 Mar 2001 14:45:57 +0000 (14:45 +0000)]
More memory leaks fixed courtesy of Insure ...
Tim Potter [Wed, 28 Mar 2001 05:01:54 +0000 (05:01 +0000)]
Changed usage message for query_groupmem and query_usergroups as they don't
convert names to rids yet.
Jeremy Allison [Wed, 28 Mar 2001 01:33:57 +0000 (01:33 +0000)]
Fixed the problem Gerald reported. Unfortunately we need to go back to
reporting imaginary "default" inheritable ACLs on directories, otherwise,
when you add an entry and click on apply without noticing there's no
default entry associated with it, it applies a null acl on the files
within the directory (hey, that's what you told NT you wanted, right ! :-).
Also ensure that minimum permissions for a directory are r-x for owner,
not just r--.
Jeremy.
Gerald Carter [Tue, 27 Mar 2001 18:19:21 +0000 (18:19 +0000)]
make proto
Gerald Carter [Tue, 27 Mar 2001 18:19:01 +0000 (18:19 +0000)]
merge from 2.2.
Richard Sharpe [Tue, 27 Mar 2001 12:13:59 +0000 (12:13 +0000)]
One small Insure fix for a memory leak. More fixes to come perhaps ...
Also fixed an error return for smbc_rmdir so that we can distinguish
between EACCES and ENOTEMPTY
Jeremy Allison [Tue, 27 Mar 2001 08:08:04 +0000 (08:08 +0000)]
Removed NFS quotas code for Solaris as Alan wants to re-write it.
Jeremy.
Jeremy Allison [Tue, 27 Mar 2001 03:16:05 +0000 (03:16 +0000)]
Patch from itegem <J.P.M.v.Itegem@ele.tue.nl> to handle LPRng v3.16 and above.
Jeremy
Jeremy Allison [Tue, 27 Mar 2001 02:43:34 +0000 (02:43 +0000)]
Added NFS quota support for Solaris 5.x from Alan Romeril <a.romeril@ic.ac.uk>.
sun1.samba.org is down at the moment so I can't test the compile on this. I'm
sure Solaris people using quotas will scream if I've meesed anything up :-).
Jeremy.
Jeremy Allison [Tue, 27 Mar 2001 01:41:50 +0000 (01:41 +0000)]
standards.h only exists on IRIX 6.x and above (Herb please check !).
Jeremy.
Jeremy Allison [Tue, 27 Mar 2001 01:19:54 +0000 (01:19 +0000)]
Patch from David Gibson <dgibson@linuxcare.com> to reduce "silent abort"
problems with smbd failing to create a log file. If we can't create a log
file keep using the old file.
Jeremy.
Jeremy Allison [Tue, 27 Mar 2001 00:28:01 +0000 (00:28 +0000)]
Fix for ENOSPC from DCB.
Jeremy.
Jeremy Allison [Tue, 27 Mar 2001 00:24:40 +0000 (00:24 +0000)]
Bail out early if null passwords and lp_null_passwords not set.
Jeremy.
Gerald Carter [Tue, 27 Mar 2001 00:03:49 +0000 (00:03 +0000)]
merge from 2.2
Jeremy Allison [Mon, 26 Mar 2001 23:21:29 +0000 (23:21 +0000)]
Patch from Massimo Sivilotti <mass@tanner.com> to log remote machine/ip on
connection fail.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 23:13:02 +0000 (23:13 +0000)]
Fix from Ryo Kawahara <rkawa@lbe.co.jp> to make SWAT correctly write and smb.conf
file in utf8.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 22:34:25 +0000 (22:34 +0000)]
Fix for smbtar race condition from Glenn Burkhardt <glenn@aoi.ultranet.com>.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 22:28:12 +0000 (22:28 +0000)]
Missed a cli_set_port on message code connect.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 22:20:43 +0000 (22:20 +0000)]
Patch from Sean.Batt@anu.edu.au to ensure error reporting is done before
any code that could modify errno is called.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 22:15:40 +0000 (22:15 +0000)]
Fix from Dave Collier Brown to log a debug at level zero on common misconfigure.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 21:52:43 +0000 (21:52 +0000)]
Patch to make automount lookup fallback to get home directory from getpwnam.
From Robert Montjoy <Rob_Montjoy@ECECS.UC.EDU>.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 19:18:06 +0000 (19:18 +0000)]
smbd/posix_acls.c: Saving and restoring errno here is the wrong place. Moved it
to the places where [f]chmod_acl is called instead.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 17:58:05 +0000 (17:58 +0000)]
Ensure get/set NT ACL code is redirected through vfs.
Jeremy.
Jeremy Allison [Mon, 26 Mar 2001 05:43:04 +0000 (05:43 +0000)]
smbd/posix_acls.c: Sync up with 2.2 changes - don't return deny ACE's.
smbd/vfs.c: Don't call [f]chmod_acl if no acl support.
Jeremy.
Gerald Carter [Sun, 25 Mar 2001 06:05:57 +0000 (06:05 +0000)]
merge from 2.2
Jeremy Allison [Fri, 23 Mar 2001 21:54:30 +0000 (21:54 +0000)]
Added Gerald's fixes.
Jeremy.
Andrew Tridgell [Fri, 23 Mar 2001 21:37:30 +0000 (21:37 +0000)]
a much simpler talloc() implementation. This version has the following
advantages:
- memory is trackable by insure
- a very simple talloc_realloc() is possible (I've added it)
It is slower than the previous talloc code, but I don't think that
is going to be a problem. If it is a problem then there are
some ways we can make it faster but I'd like to leave those
until we have tested this a bit and can see what performance
problems might show up in profiling
Jeremy Allison [Fri, 23 Mar 2001 20:41:22 +0000 (20:41 +0000)]
More unused variables.
Jeremy.
Jeremy Allison [Fri, 23 Mar 2001 19:01:27 +0000 (19:01 +0000)]
Fix insure-found parameter size missmatch.
Jeremy.
Jeremy Allison [Fri, 23 Mar 2001 18:14:22 +0000 (18:14 +0000)]
Made the fixes Andrew suggested to asci string code to use srvstr_push.
Jeremy.
Jeremy Allison [Fri, 23 Mar 2001 03:12:58 +0000 (03:12 +0000)]
Two OS/2 printer fixes from Jim McDonough @ IBM.
First one adds a new info level into the lanman printing and an ioctl to the trans2 code.
Andrew - this uses ASCII only. It looks ok to me but please check !
Second one adds a parameter "os2 driver map" that allows OS/2 driver names to be mapped.
Jeremy.
Jeremy Allison [Fri, 23 Mar 2001 02:14:08 +0000 (02:14 +0000)]
groupdb/mapping.c:
include/proto.h: Fix missing (void) in proto.
rpc_server/srv_samr_nt.c: Fix user private group problem by filtering out groups that
clash with users.
smbd/posix_acls.c: Ensure default ACE's are sensible.
utils/pdbedit.c: Fix from Simo Sorce.
Jeremy.
Jean-François Micouleau [Fri, 23 Mar 2001 01:19:21 +0000 (01:19 +0000)]
can delete mapping from command line now.
J.F.
Jeremy Allison [Fri, 23 Mar 2001 00:54:55 +0000 (00:54 +0000)]
Sync up with 2.2 ACL code.
Jeremy.
Jean-François Micouleau [Fri, 23 Mar 2001 00:50:31 +0000 (00:50 +0000)]
first pass of the new group mapping code
J.F.
Tim Potter [Thu, 22 Mar 2001 22:39:37 +0000 (22:39 +0000)]
Updated broken links.
Jeremy Allison [Thu, 22 Mar 2001 01:26:37 +0000 (01:26 +0000)]
New POSIX ACL mapping code. Works with UNIX permissions, now for testing
with real ACLs...
Jeremy.
Jeremy Allison [Wed, 21 Mar 2001 23:50:26 +0000 (23:50 +0000)]
Fix memory leak when doing DLIST_REMOVE(head,head).
Jeremy.
Andrew Tridgell [Wed, 21 Mar 2001 06:44:49 +0000 (06:44 +0000)]
added -E option (for exit on errors)
forced exclusion of filenames containing only '.' as NT rejects those
Andrew Tridgell [Wed, 21 Mar 2001 05:54:09 +0000 (05:54 +0000)]
latest attempt at lanman1 wildcard matching
Andrew Tridgell [Wed, 21 Mar 2001 04:14:02 +0000 (04:14 +0000)]
added "enhanced browsing" option docs
Andrew Tridgell [Wed, 21 Mar 2001 04:12:36 +0000 (04:12 +0000)]
added option "enhanced browsing"
this allows users to disable the cross-subnet browse extensions that I
added to Samba a couple of years ago. This may be useful for getting
rid of empty workgroups.
Jeremy Allison [Tue, 20 Mar 2001 23:29:36 +0000 (23:29 +0000)]
When calling DLIST_REMOVE(x,x) (from lib/interface.c) ensure that the pointer
is still valid before setting prev and next to null.
Jeremy.
Jeremy Allison [Tue, 20 Mar 2001 23:07:36 +0000 (23:07 +0000)]
Fix for crash when doing name lookup with a quoted string. Part of
lookup_name was expecting to be able to write to the string. Changed
lookup_name to use const.
Jeremy.
Gerald Carter [Mon, 19 Mar 2001 18:14:09 +0000 (18:14 +0000)]
reverted the rename of new_spoolss_io_r_enumprinterdrivers()
Gerald Carter [Mon, 19 Mar 2001 17:23:26 +0000 (17:23 +0000)]
merge from 2.2 (reorder commands in array)
Jeremy Allison [Mon, 19 Mar 2001 07:08:02 +0000 (07:08 +0000)]
Added sys_dlopen/sys_dlsym/sys_dlclose.
Jeremy.
Andrew Tridgell [Mon, 19 Mar 2001 00:32:16 +0000 (00:32 +0000)]
fixed rpcclient readline code
Andrew Tridgell [Mon, 19 Mar 2001 00:22:52 +0000 (00:22 +0000)]
added basic command completion support
Andrew Tridgell [Sun, 18 Mar 2001 23:42:30 +0000 (23:42 +0000)]
fixed unused variable
Andrew Tridgell [Sun, 18 Mar 2001 23:41:53 +0000 (23:41 +0000)]
much simpler readline code
should work with readline 2.x
Andrew Tridgell [Sun, 18 Mar 2001 22:47:17 +0000 (22:47 +0000)]
fixed some compilation errors with IRIX cc
Andrew Tridgell [Sun, 18 Mar 2001 13:24:57 +0000 (13:24 +0000)]
much better readline support from Simo Sorce, with some mods from me
to make it a bit simpler
Andrew Tridgell [Sat, 17 Mar 2001 14:07:10 +0000 (14:07 +0000)]
fix this in rpc calls as well
- ADMIN$ is an IPC share, not a disk share
Andrew Tridgell [Sat, 17 Mar 2001 14:02:44 +0000 (14:02 +0000)]
ADMIN$ is an IPC share, not a disk share
Andrew Tridgell [Sat, 17 Mar 2001 06:23:09 +0000 (06:23 +0000)]
removed useless debug msg
Jeremy Allison [Sat, 17 Mar 2001 03:36:38 +0000 (03:36 +0000)]
lib/system.c (Finally) fixed all insure errors in password caching code. We can't
stop libc routines from calling getpwXXX functions, so caching a pointer to them
is impossible. This new code now makes two copies of the returned struct passwd
struct - one used as a cache, one returned to allow the caller to modify. When
doing a lookup we compare against the cached copy. Code is now easier to understand
also.
smbd/posix_acls.c: If we move the head of the linked list, remember to pass a
reference to that pointer.....
Jeremy.
Jeremy Allison [Sat, 17 Mar 2001 02:06:16 +0000 (02:06 +0000)]
I'm happy with the mapping for NT getACL, now for NT setACL.
Jeremy.
Jeremy Allison [Fri, 16 Mar 2001 19:17:04 +0000 (19:17 +0000)]
Updates for CUPS 1.1 from Michael Sweet.
Jeremy.
Jeremy Allison [Fri, 16 Mar 2001 19:10:55 +0000 (19:10 +0000)]
Updated defaults for CUPS printing from Michael Sweet.
Jeremy.
Andrew Tridgell [Fri, 16 Mar 2001 06:17:45 +0000 (06:17 +0000)]
don't need to force unicode strings in flg2 anymore
Andrew Tridgell [Fri, 16 Mar 2001 06:17:11 +0000 (06:17 +0000)]
fixed volume_name(). It used dos_to_unix() which uses a static buffer,
and that buffer got clobbered in the trans2 QFS_VOLUME_INFO call, so
the volume label always showed as "." from NT clients.
Jeremy Allison [Fri, 16 Mar 2001 05:55:30 +0000 (05:55 +0000)]
Added Michael Sweet's CUPS patch to call directly into the CUPS backend.
Parameterises the printing functions so other backends can be plugged
directly in (this is a good thing :-).
Jeremy.
Andrew Tridgell [Fri, 16 Mar 2001 04:09:47 +0000 (04:09 +0000)]
finished the conversion to unicode of the last of the trans2 fns
Andrew Tridgell [Fri, 16 Mar 2001 03:53:59 +0000 (03:53 +0000)]
converted the nttrans code to unicode on the wire
Andrew Tridgell [Fri, 16 Mar 2001 03:43:14 +0000 (03:43 +0000)]
converted reply_printqueue
Andrew Tridgell [Fri, 16 Mar 2001 03:25:33 +0000 (03:25 +0000)]
converted reply_search