Tim Potter [Fri, 31 Oct 2003 02:56:11 +0000 (02:56 +0000)]
Merge from 3.0:
Revision 1.45.2.23:
Changed output so all debug output goes to stderr, and all file processing
goes to stdout.
Note: This change permits use of testparm processing of smb.conf to be
redirected into a file that can be used as an smb.conf file. ie: All
information that should not be in smb.conf will be on stderr, all pertinent
smb.conf info will go to stdout.
Example of use:
A fully documented smb.conf.master file can be maintained.
To create smb.conf do:
testparm -s > smb.conf
Jim McDonough [Thu, 30 Oct 2003 18:13:25 +0000 (18:13 +0000)]
Last of the GUID->struct uuid changes.
GUID has been removed. If you are dealing with a flattened version of a
[gu]uid, use UUID_FLAT. smb_pack_uuid and smb_unpack_uuid will switch
between the two.
I'm not sure exactly what the genparse stuff is doing here, so I just
switched it to a UUID_FLAT (no functional change), but I suspect this
may not be the right way to go.
Volker Lendecke [Thu, 23 Oct 2003 16:51:08 +0000 (16:51 +0000)]
Merge from 3_0:
After a phonecall with jra finally commit this.
This changes our behaviour when the setresuid call is available. We now not
only change the effective uid but also the real uid when becoming
unprivileged. This is mainly for improved AFS compatibility, as AFS selects
the token to send to the server based on the real uid of the process.
I tested this with a W2k server with two non-root 'runas' sessions. They come
in via a single smbd as two different users using two session setups. Samba on
Linux can still switch between the two uids, proved by two different files
created via those sessions.
Volker Lendecke [Thu, 23 Oct 2003 13:46:32 +0000 (13:46 +0000)]
Merge from 3_0:
According to Ethereal we have a 32-Bit quantity here. And with SSVAL valgrind
reports an unitialized read which is obviously correct. And I hate valgrind
errors ;-)
Jeremy Allison [Wed, 22 Oct 2003 23:38:18 +0000 (23:38 +0000)]
Put strcasecmp/strncasecmp on the banned list (except for needed calls
in iconv.c and nsswitch/). Using them means you're not thinking about multibyte at
all and I really want to discourage that.
Jeremy.
Volker Lendecke [Wed, 22 Oct 2003 07:31:17 +0000 (07:31 +0000)]
Merge from 3_0:
In cli_lsa_lookup_sids don't leave the domain field uninitialized if
some sid could not be mapped. Otherwise this call is unnecessarily
complicated to call.
Jeremy Allison [Mon, 20 Oct 2003 19:02:04 +0000 (19:02 +0000)]
Merge Volker's fix.
We are doing NT error codes now.... If we have an NT error, report that
back the same way we handle the DOS error. Although I don't see why
BUFFER_TOO_SMALL should not be handled as an error, simply copy the logic.
Jim McDonough [Mon, 20 Oct 2003 18:35:12 +0000 (18:35 +0000)]
Several updates:
- add support for named pipe and netbios queries in parse code
- fix map request structure...unknown byte was alignment
- add sample of named pipe over netbios query in rpcclient (comment only)
Jim McDonough [Fri, 17 Oct 2003 19:47:06 +0000 (19:47 +0000)]
uuid itself might not be aligned (as is the case in epm map requests), so
it needs to be aligned outside the smb_io_rpc_uuid() call if a specific
rpc or struct needs it that way.
Simo Sorce [Wed, 15 Oct 2003 17:41:25 +0000 (17:41 +0000)]
created a new target: genparse
this target will build parse file with genstruct
it is the duty of the developer to commit updated files
this is made to make build platforms independent of a working perl
installation as always been with samba
so currently you need to run:
make genparse
and commit:
cvs ci include/tdbsam2_parse_info.h
if you change anything in genparse/genstruct code or tdbsam2 code.
Simo Sorce [Mon, 13 Oct 2003 21:26:10 +0000 (21:26 +0000)]
So here it is a non-intrusive patch with my latest work on gums (the
laternative to the current passdb).
Currently it is run through a comatibility module in the passdb layer, with
a subset of the functionality it may provide.
It is still work in progress, but as someone asked me about it, and as it
should make no difference to the normal code, I tought it was a good idea to
put it into.
It adds a dependency on perl. I know it is not very nice, but I'm sure we
will work out a solution for that.
As always blame me if I break something, but try to fix yourself, as I am
busy-busy-busy :-)
Jelmer Vernooij [Thu, 9 Oct 2003 22:56:08 +0000 (22:56 +0000)]
We need to make sure that the files that contain the static_init_$subsystem;
macro get recompiled after configure has ran.
This was done by touching the source file. This patch changes it to removing
the object file, so that you don't need write-access to the source directory.