cvs2svn Import User [Thu, 15 Jul 1999 12:39:49 +0000 (12:39 +0000)]
This commit was manufactured by cvs2svn to create tag
'samba_2_0_5_pre1'.
Andrew Tridgell [Thu, 15 Jul 1999 12:39:48 +0000 (12:39 +0000)]
I'm going to do a snapshot release of 2.0.5-pre1. This is just to get
the version number right.
Andrew Tridgell [Thu, 15 Jul 1999 11:42:53 +0000 (11:42 +0000)]
fix from matty for directory creation on read-only shares
Andrew Tridgell [Thu, 15 Jul 1999 10:43:26 +0000 (10:43 +0000)]
add a length to string_sub() and add fstring_sub() and pstring_sub()
Jeremy Allison [Thu, 15 Jul 1999 00:42:40 +0000 (00:42 +0000)]
Moved log message to correct place.
Jeremy.
Jeremy Allison [Wed, 14 Jul 1999 18:55:59 +0000 (18:55 +0000)]
force_group and force_user fixes. Added '+' element to force_group.
Jeremy.
Andrew Tridgell [Wed, 14 Jul 1999 14:16:22 +0000 (14:16 +0000)]
if the share modes fail to initialise and we are root then allow them
to init read-write. This allows people to run smbstatus directly after
a install/upgrade without nasty messages.
Andrew Tridgell [Wed, 14 Jul 1999 13:58:46 +0000 (13:58 +0000)]
removed the HAVE_CRYPT_DECL autoconf stuff as it doesn't work with
glibc 2.1
Andrew Tridgell [Wed, 14 Jul 1999 13:31:20 +0000 (13:31 +0000)]
new prototypes
Andrew Tridgell [Wed, 14 Jul 1999 13:30:34 +0000 (13:30 +0000)]
an update from 2.0
Andrew Tridgell [Wed, 14 Jul 1999 13:29:35 +0000 (13:29 +0000)]
make the default printing model on all Linux systems BSD
Andrew Tridgell [Wed, 14 Jul 1999 13:22:05 +0000 (13:22 +0000)]
the smbd changes from the 2.0 branch. This is going to need a _lot_ of
testing over the next few days.
Andrew Tridgell [Wed, 14 Jul 1999 13:12:13 +0000 (13:12 +0000)]
lib changes from the 2.0 branch
this includes the wildcard fixes, some grotesque reformatting and some
pid_t stuff
Andrew Tridgell [Wed, 14 Jul 1999 13:06:36 +0000 (13:06 +0000)]
include changes to support the new stuff, particularly levelII
oplocks.
all taken from the 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 13:01:47 +0000 (13:01 +0000)]
configure.in changes for .so as a variable
Andrew Tridgell [Wed, 14 Jul 1999 13:00:58 +0000 (13:00 +0000)]
Makefile.in changes for .so as a variable
Andrew Tridgell [Wed, 14 Jul 1999 12:55:25 +0000 (12:55 +0000)]
the locking code from 2.0
I've finally decided that Jeremy's new levelII code should go into
2.0.5, but disabled by default. We have a few days to make sure that
the changes required to support the new code, with the option
disabled, don't break anything. This seemed like the only sane
decision.
Andrew Tridgell [Wed, 14 Jul 1999 12:48:07 +0000 (12:48 +0000)]
parameter changes from 2.0, except I've made levelII oplocks off by
default
Andrew Tridgell [Wed, 14 Jul 1999 12:38:53 +0000 (12:38 +0000)]
the password database updates from 2.0. This is mostly the fix for the
constant string problem
Andrew Tridgell [Wed, 14 Jul 1999 12:31:17 +0000 (12:31 +0000)]
ACL order change from the 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 12:27:35 +0000 (12:27 +0000)]
smbtar fix from 2.0
Andrew Tridgell [Wed, 14 Jul 1999 12:25:44 +0000 (12:25 +0000)]
- cleanups in namequery code from 2.0
- use only one outstanding read until we find the multiplex bug
- add oplocks to client code (only used by smbtorture)
Andrew Tridgell [Wed, 14 Jul 1999 12:19:13 +0000 (12:19 +0000)]
pid_t and testparm cleanups from the 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 12:18:10 +0000 (12:18 +0000)]
more pid_t cleanups from the 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 06:58:24 +0000 (06:58 +0000)]
swat fixes from 2.0 branch
this cnludes updates to the status page for levelII oplocks. Once we
decide if the levelII code goes into 2.0.5 we will work out if this
stays
Andrew Tridgell [Wed, 14 Jul 1999 06:56:06 +0000 (06:56 +0000)]
transfer some more fixes from the 2.0 branch (mostly smbtorture changes)
Andrew Tridgell [Wed, 14 Jul 1999 06:54:31 +0000 (06:54 +0000)]
transfered nmbd fixes from the 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 06:53:56 +0000 (06:53 +0000)]
transfer pid_t updates from 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 06:53:28 +0000 (06:53 +0000)]
transfer the SIGIOCONF fix from the 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 06:52:48 +0000 (06:52 +0000)]
transfer minor fixes from the 2.0 branch
Andrew Tridgell [Wed, 14 Jul 1999 05:19:19 +0000 (05:19 +0000)]
transfer the 127.0.0.1 behaviour from the 2.0 branch
Andrew Tridgell [Tue, 13 Jul 1999 14:48:51 +0000 (14:48 +0000)]
we don't need this anymore
Andrew Tridgell [Tue, 13 Jul 1999 14:48:37 +0000 (14:48 +0000)]
remove some unused variables
Andrew Tridgell [Tue, 13 Jul 1999 14:47:54 +0000 (14:47 +0000)]
greatly simplify smbmnt, removing the u g f d and P options.
Andrew Tridgell [Tue, 13 Jul 1999 14:47:01 +0000 (14:47 +0000)]
transfer the -b option from the 2.0 branch
Andrew Tridgell [Tue, 13 Jul 1999 14:46:06 +0000 (14:46 +0000)]
removed some meta-data
Andrew Tridgell [Tue, 13 Jul 1999 14:45:30 +0000 (14:45 +0000)]
- fixed some URLs
- made the installation instructions more obvious
Jeremy Allison [Tue, 13 Jul 1999 02:05:05 +0000 (02:05 +0000)]
Looking at string_sub() - which seems to be the cause of Olaf's complaints,
it looks like the lengths which were converted to size_t's should have
been ssize_t's in these two functions. At one point two lengths are
subtracted, and this should be a signed type.
Jeremy.
Jeremy Allison [Tue, 13 Jul 1999 01:42:35 +0000 (01:42 +0000)]
param/loadparm.c: Removed StrCpy from Samba - last use.
lib/util_str.c: Removed StrCpy - added alpha_strcpy().
smbd/message.c: Fix for buffer overrun issues.
include/proto.h: Removed StrCpy.
Jeremy.
Jeremy Allison [Mon, 12 Jul 1999 20:08:26 +0000 (20:08 +0000)]
Purify fixes plus 2 DOS bug fixes found by : Olaf Kirch <okir@monad.swb.de>
Jeremy.
Jeremy Allison [Tue, 29 Jun 1999 16:43:20 +0000 (16:43 +0000)]
Fix for directory copies on Win2k.
typo in oplock.c spotted by J.F.
Jeremy.
Matthew Chapman [Thu, 24 Jun 1999 13:30:55 +0000 (13:30 +0000)]
Adding Win2k browsing fix to release 2.0.4c.
Jeremy Allison [Wed, 23 Jun 1999 01:52:03 +0000 (01:52 +0000)]
Updated with a couple of other changes now I've done the cvs diff against
the 2.0.4b release tag.
Jeremy.
Jeremy Allison [Wed, 23 Jun 1999 01:34:25 +0000 (01:34 +0000)]
Updated for a possible 2.0.4c.
Jeremy.
Jeremy Allison [Wed, 23 Jun 1999 01:06:25 +0000 (01:06 +0000)]
Adding Andrew's new smbmount for Linux and the fixes to the
file_struct leak.
Jeremy.
Jeremy Allison [Fri, 28 May 1999 19:18:23 +0000 (19:18 +0000)]
99.
9999999% sure this is the correct fix for the directory open bugs.
I finally noticed in the Lotus Notes capture file that the "FILE_NON_DIRECTORY"
flag was being set in create options for a open on a directory that should
fail. All the other heuristics were incorrect.
Jeremy.
Herb Lewis [Wed, 26 May 1999 21:00:09 +0000 (21:00 +0000)]
add missing cvs.log file
Jeremy Allison [Wed, 26 May 1999 18:45:11 +0000 (18:45 +0000)]
Fixed the sense of the fcntl_lock* tests using WIFEXITED.
Don't check for fcntl64 if the glibc2.1 broken test has already
failed. This *should* fix the RedHat 6.0 glibc2.1 bugs.
Jeremy.
Jeremy Allison [Tue, 25 May 1999 23:11:12 +0000 (23:11 +0000)]
New broken 64 bit locking test.
Jeremy.
Jeremy Allison [Tue, 25 May 1999 23:10:04 +0000 (23:10 +0000)]
(Hopefully) correct fix for a glibc2.1 x86 Linux system w.r.t. the 64
bit locking issue. Added a test to check for broken 64 bit locks
and to use the 64 -> 32 bit mapping code in reply.c if detected.
Jeremy.
Jeremy Allison [Tue, 25 May 1999 00:16:36 +0000 (00:16 +0000)]
Fixed core dump causing bug from cr/lf translation. Fix from raymond@orion.no
Jeremy.
Herb Lewis [Tue, 25 May 1999 00:15:23 +0000 (00:15 +0000)]
examples/autofs/mount.smb
examples/printer-accounting/hp5-redir
examples/printer-accounting/lp-acct get rid of $Id $Source and $Log
packaging/SGI/idb.pl remove files from include/.cvsignore and add files
from the SCRIPTS target that were missed.
packaging/SGI/mkman don't make man pages for smbmount, smbumount, smbrun
Jeremy Allison [Mon, 24 May 1999 22:18:36 +0000 (22:18 +0000)]
Fixed typo in configure.
Jeremy.
Jeremy Allison [Mon, 24 May 1999 22:06:06 +0000 (22:06 +0000)]
Bugfix for pending_modtime getting set to -1 from David N. Worrell
(dworrell@wfubmc.edu).
Jeremy.
Jeremy Allison [Thu, 20 May 1999 00:30:53 +0000 (00:30 +0000)]
Fix for parent smbd dieing in sigterm. This was previously
committted and now seems reverted.... hmmmmm.
Jeremy.
Samba Release Account [Thu, 20 May 1999 00:12:35 +0000 (00:12 +0000)]
preparing for release of 2.0.4b
Jeremy Allison [Thu, 20 May 1999 00:04:55 +0000 (00:04 +0000)]
Added the NT_Security file to the packaging for RedHat.
Jeremy Allison [Thu, 20 May 1999 00:03:12 +0000 (00:03 +0000)]
For some reason this html file didn't get correctly tagged and added to
the release branch...
Jeremy.
Samba Release Account [Wed, 19 May 1999 22:13:03 +0000 (22:13 +0000)]
preparing for release of 2.0.4b
Jeremy Allison [Wed, 19 May 1999 21:02:37 +0000 (21:02 +0000)]
Stop rcs complaining about log.
Jeremy.
Jeremy Allison [Wed, 19 May 1999 19:19:35 +0000 (19:19 +0000)]
Fixing Word bug in 2.0.x release branch.
Jeremy.
cvs2svn Import User [Tue, 18 May 1999 17:49:53 +0000 (17:49 +0000)]
This commit was manufactured by cvs2svn to create branch
'SAMBA_2_0_RELEASE'.
Herb Lewis [Tue, 18 May 1999 17:49:51 +0000 (17:49 +0000)]
cast ptsname to get rid of warning message
Herb Lewis [Tue, 18 May 1999 16:41:10 +0000 (16:41 +0000)]
remove html file after copy
Jeremy Allison [Tue, 18 May 1999 16:35:34 +0000 (16:35 +0000)]
Changed version to 2.0.4a as I am adding the NT_Security docs into
the tarball. I checked in the NT_Security files in the 2.0 branch
only and got bitten by a cvs 'feature' :-).
Jeremy.
Matthew Chapman [Tue, 18 May 1999 16:02:22 +0000 (16:02 +0000)]
Adding missing html and txt versions of NT_Security.yo
Herb Lewis [Tue, 18 May 1999 15:45:37 +0000 (15:45 +0000)]
fix typo
Samba Release Account [Tue, 18 May 1999 00:39:41 +0000 (00:39 +0000)]
preparing for release of 2.0.4
Jeremy Allison [Mon, 17 May 1999 23:37:24 +0000 (23:37 +0000)]
Fixed bug where copymap was referenced if malloc fails.
Jeremy.
Jeremy Allison [Mon, 17 May 1999 22:28:11 +0000 (22:28 +0000)]
Removed the sockethack code (sorry Andrew :-). Added the recursive directory
smbclient bugfix (sorely needed).
Jeremy.
Andrew Tridgell [Sun, 16 May 1999 12:14:31 +0000 (12:14 +0000)]
added a --with-sockethack option that should give us an _enormous_
boost in netbench performance on Linux/SMP if the corresponding 4 line
change is made in the kernel. Note: this is a experimental performance
hack only at the moment. Do _not_ enable it until we find a way to
make it safe in the kernel, which may not be for a while.
Andrew Tridgell [Sat, 15 May 1999 06:20:33 +0000 (06:20 +0000)]
added full netbench simulation in smbtorture. It now reads a 4MB load
description file that was derived from a packet capture of a real
netbench run. That file describes every SMB operation that a netbench
client performs on the wire and smbtorture uses that to simulate any
number of simultaneous netbench clients, giving results on the same
scale that netbench does.
the nice thing is that it all seems to work. I'm actually getting
results pretty close to what a real netbench lab gets.
Samba Release Account [Sat, 15 May 1999 02:03:14 +0000 (02:03 +0000)]
Updated Roadmap as we have met one our our objectives
(the "NT Security release" *is* the 2.0.4 release).
Jeremy.
Samba Release Account [Sat, 15 May 1999 02:00:00 +0000 (02:00 +0000)]
Fixed typo.
Jeremy.
Jeremy Allison [Sat, 15 May 1999 00:06:34 +0000 (00:06 +0000)]
Fixes due to DEBUG format %d needing (int) casting for
arguments on AIX. This fixes many cases where we are
printing a size_t and using a format of %d.
Jeremy.
Jeremy Allison [Thu, 13 May 1999 18:03:58 +0000 (18:03 +0000)]
Updated for 2.0.4
Jeremy.
John Terpstra [Wed, 12 May 1999 20:11:29 +0000 (20:11 +0000)]
Modified post install script so xtoi is run only if /etc/xinetd.conf exists.
Jeremy Allison [Wed, 12 May 1999 19:48:04 +0000 (19:48 +0000)]
Explicit cast to int before mask...
Jeremy.
Jeremy Allison [Wed, 12 May 1999 19:34:09 +0000 (19:34 +0000)]
Ensure that when a pipe number is extracted from the incoming buffer not
using SVAL, that we mask off the lower 16 bits to ensure correct comparisons.
Jeremy.
John Terpstra [Tue, 11 May 1999 22:53:21 +0000 (22:53 +0000)]
Added new package components (not for general distribution).
Andrew Tridgell [Mon, 10 May 1999 23:19:16 +0000 (23:19 +0000)]
use a separate directory for each cleint in netbench tests
Andrew Tridgell [Sun, 9 May 1999 11:07:47 +0000 (11:07 +0000)]
print a + at the start of the unlink phase, so we can see it more
clearly
Andrew Tridgell [Sun, 9 May 1999 09:55:24 +0000 (09:55 +0000)]
fixed typo
Andrew Tridgell [Sun, 9 May 1999 09:53:08 +0000 (09:53 +0000)]
set a 2 minute timeout on SMB packets in the torture code.
Andrew Tridgell [Sun, 9 May 1999 08:40:25 +0000 (08:40 +0000)]
better filename choice
Andrew Tridgell [Sun, 9 May 1999 08:30:13 +0000 (08:30 +0000)]
use a better synchronisation method at startup
Andrew Tridgell [Sun, 9 May 1999 05:02:14 +0000 (05:02 +0000)]
don't delete files till all are written. This stresses the disk
subsystem in a more similar way to a netbench run.
Andrew Tridgell [Sun, 9 May 1999 04:45:09 +0000 (04:45 +0000)]
added synchronisation code at startup so all clients start at the same
time. That gives a more accurate result.
Andrew Tridgell [Sun, 9 May 1999 03:51:44 +0000 (03:51 +0000)]
improved the netbench simulator a bit. give results as throughput in
MB/sec
Andrew Tridgell [Sun, 9 May 1999 03:00:10 +0000 (03:00 +0000)]
added a very simple netbench simulator to smbtorture. The tests are
called NBW95 and NBWNT.
I don't have the right sort of hardware (100MBit etc) to test whether
the results of this similator are anything like what netbench sees,
but I'm hoping Jeremy will run that test for me.
A typical run would be:
smbtorture //server/share -Uuser%pass -o 10 -N 20 NBWNT NBW95
that would run a NT and a 95 client test with 20 clients and 10 files
per client. The result is given as seconds for the run.
I think that for best results you would use a few copies of smbtorture
running on different clients and each similating maybe 10 clients to
build up a large client load with a small number of client machines.
Andrew Tridgell [Sun, 9 May 1999 00:37:56 +0000 (00:37 +0000)]
modified smbtorture so you can run individual tests using command line
parameters. For example:
smbtorture //fjall/tmp RANDOMIPC LOCK2 ATTR
would run those 3 tests. ALL will run all tests. The default is ALL.
Matthew Chapman [Sat, 8 May 1999 06:23:11 +0000 (06:23 +0000)]
The install instructions file is UNIX_INSTALL.txt not UNIX-INSTALL.txt.
Never underestimate the stupidity of SysAdmins.
Jeremy Allison [Wed, 5 May 1999 18:32:52 +0000 (18:32 +0000)]
Fix check return for sys_open.
Jeremy.
Jeremy Allison [Wed, 5 May 1999 17:06:54 +0000 (17:06 +0000)]
Tidyups caught by running gcc with -Wall. It gets a different set from the IRIX compiler.
Jeremy.
Jeremy Allison [Tue, 4 May 1999 01:05:08 +0000 (01:05 +0000)]
Size bugfix for smbsh (patch from Dave Smith dsmith@tqs.com).
Jeremy.
Jeremy Allison [Mon, 3 May 1999 19:31:25 +0000 (19:31 +0000)]
Removed the null packet send. This code worked perfectly with IRIX
and kills Win9x clients with a Linux server. I need to go back and
check the WinNT traces to make sure of this.....
Jeremy.
Jeremy Allison [Mon, 3 May 1999 17:33:10 +0000 (17:33 +0000)]
Added "oplock break wait time" parameter to help with Win9x oplock related
problems. Default set to 10ms.
Also documented it :-).
Jeremy.
Jeremy Allison [Thu, 29 Apr 1999 01:53:15 +0000 (01:53 +0000)]
Fixed (I think) the last outstanding bug before 2.0.4. This was the lotus
notes client problem where we were too generous in allowing a fallback
to opening a directory as a file (PR#15717). I added code to ensure that only non
oplock opens with no read/write access requested fall back to a directory
open.
This seems to work with the NT rename problem as well as correctly
stopping "type \directory".
Jeremy.
Jeremy Allison [Wed, 28 Apr 1999 22:21:43 +0000 (22:21 +0000)]
Re-wrote prs_force_grow() as my fist implementation was crap. Now only
grow by the requested amount in this case.
Jeremy.
Jeremy Allison [Wed, 28 Apr 1999 21:44:59 +0000 (21:44 +0000)]
Fixed the security=domain which seemed to have been broken. Hmmm. Wonder
when. The code in rpc_client/cli_pipe.c wasn't coping with a ERRmoredata
error correctly, and the rpc_read code wasn't growing the parse buffer
correctly either.....
Also made smbd and nmbd ignore SIGFPE for systems that have it.
Jeremy.