cvs2svn Import User [Sat, 27 Feb 1999 22:18:20 +0000 (22:18 +0000)]
This commit was manufactured by cvs2svn to create tag
'release-2-0-3'.
Samba Release Account [Sat, 27 Feb 1999 22:18:19 +0000 (22:18 +0000)]
preparing for release of 2.0.3
John Terpstra [Sat, 27 Feb 1999 22:14:18 +0000 (22:14 +0000)]
A couple of typos.
John Terpstra [Sat, 27 Feb 1999 22:10:57 +0000 (22:10 +0000)]
A tiny typo fix.
Jeremy Allison [Sat, 27 Feb 1999 22:06:38 +0000 (22:06 +0000)]
Added one more bugfix line.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 21:53:24 +0000 (21:53 +0000)]
Tidied up new docs. Put announcement of new docs in WHATSNEW.txt.
Jeremy.
John Terpstra [Sat, 27 Feb 1999 21:45:36 +0000 (21:45 +0000)]
Updated chowning method.
John Terpstra [Sat, 27 Feb 1999 21:36:39 +0000 (21:36 +0000)]
Changed for samba-2.0
John Terpstra [Sat, 27 Feb 1999 21:30:26 +0000 (21:30 +0000)]
Added new documents to reflect auto-FAQs on samba-bugs.
Jeremy Allison [Sat, 27 Feb 1999 21:12:33 +0000 (21:12 +0000)]
Added John's changes to RedHat.
Jeremy.
John Terpstra [Sat, 27 Feb 1999 21:08:26 +0000 (21:08 +0000)]
Modified how we set uid/gid
John Terpstra [Sat, 27 Feb 1999 21:04:36 +0000 (21:04 +0000)]
Changed chown procedure.
John Terpstra [Sat, 27 Feb 1999 20:53:11 +0000 (20:53 +0000)]
Modified attribution info, checked contents.
Jeremy Allison [Sat, 27 Feb 1999 20:13:01 +0000 (20:13 +0000)]
Version changed to 2.0.3 - in preparation for release.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 20:12:25 +0000 (20:12 +0000)]
Removed smbrun binary (left man pages & html docs) from spec files for
Linux systems. smbrun is not now build by default.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 19:37:17 +0000 (19:37 +0000)]
Fixed authentication error messages. This should make it easier
for admins to determine whats wrong with logins.
Upped NT share number to 200. I have *got* to make this dynamic
before 2.0.4.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 03:02:05 +0000 (03:02 +0000)]
Ensure debug levels for user authentication failures are 1 - so admins
can determine *why* users are not being authenticated.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 02:48:03 +0000 (02:48 +0000)]
Updated with smbpasswd fix.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 02:46:58 +0000 (02:46 +0000)]
Fixed bug where setting the password of a completely new user (with
password field set to 'XXXX', but *not* disabled) would set the
disabled flag.
Added code to smbpasswd to print out the state of the disabled and
password not required flags on password change to help admins spot
these problems.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 01:27:50 +0000 (01:27 +0000)]
Updated for 2.0.3.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 00:28:27 +0000 (00:28 +0000)]
Updated html and text docs from yodl changes.
Jeremy.
Jeremy Allison [Sat, 27 Feb 1999 00:07:08 +0000 (00:07 +0000)]
A small raft of fixes... :-).
Added the ability to announce as NT server or NT workstation to the
"announce as" parameter. Updated docs for change.
Added "min passwd length" parameter (and documented it) to allow minimum
unix password length on password change to be set at runtime. Default
is 5 characters set the local.h
Fixed the RPC code that was hardcoding the version at 5.2 and the
announce bitmask parameters.
Jeremy.
Herb Lewis [Fri, 26 Feb 1999 22:56:22 +0000 (22:56 +0000)]
update parameters to proper values for password sync to work
John Terpstra [Fri, 26 Feb 1999 04:46:22 +0000 (04:46 +0000)]
README to clear up the SMBFS misapprehensions.
Gerald Carter [Thu, 25 Feb 1999 14:57:58 +0000 (14:57 +0000)]
Updated to state that 2.0 does not **officially** suport domain logons
and refer readers to NT Domain FAQ for more info
-jc
Jeremy Allison [Wed, 24 Feb 1999 21:48:05 +0000 (21:48 +0000)]
Attempt to fix smbd always root problem for AIX 4.1.x and 3.x.
Used some of the old code from 1.9.18 to call setpriv and
setuidx/setgidx. This will only work if AIX is correctly
detected as a trapdoor system though.
Jeremy.
Jeremy Allison [Wed, 24 Feb 1999 19:11:28 +0000 (19:11 +0000)]
After a long chat with Luke (a pleasant one actually, thanks for asking :-),
he is happy with me reverting the RPC change.
The reason I need this code in the 2.0 branch (even though it doesn't
work correctly) is I am using it as a stable area that no-one cares
about (as it doesn't work) to write the fully error checked version
of the RPC transport and parsing code.
Having it removed caused me a few problems :-).
Jeremy.
Luke Leighton [Wed, 24 Feb 1999 16:55:53 +0000 (16:55 +0000)]
removed the following dce/rpc code from samba 2.0.x branch:
- smbd (\PIPE\samr)
- smbd (\PIPE\winreg)
- smbd (\PIPE\lsarpc - LsaLookupNames; LsaLookupSids)
- smbd (\PIPE\srvsvc - NetrFileEnum; NetrSessionEnum; NetrConnectionEnum)
- rpcclient
- rpctorture
where parsing and header files were common between rpcclient and smbd
these were also removed. this is one-level only: functions that _these_
functions call, e.g smb_io_unistr3 in rpc_parse/parse_misc.c have _not_
been removed.
lib/util.c had two registry assistance functions removed because they
used #defines in include/rpc_reg.h. this indicates that in main cvs
branch the #defines should actually be in smb.h.
the functionality removed is only that code which is necessary to get
USRMGR.EXE and SRVMGR.EXE to work.
functionality that is unaffected:
- browsing (net share enum)
- nt domain netlogons
functionality that will return to pre-nt-domain code (i.e. SMBtrans2):
- password changing
i am sorely tempted to remove the NETLOGON functionality however jeremy
is guaranteed to go absolutely mental if i do that (pre-existing users,
despite us not telling them 2.0 was ok to use as a PDC) and andrew is
probably going to remove my cvs access for this one as it is.
a good justification for removing NETLOGON functionality from 2.0 is so
that microsoft doesn't put up a KB article advising people not to use
samba because it damages NT clients. with respect to 2.0.2, i agree
with them. the fact that this is actually their fault is neither here
nor there: we shouldn't aggravate bugs in nt clients.
Jeremy Allison [Tue, 23 Feb 1999 18:36:37 +0000 (18:36 +0000)]
Missed FILE_WRITE_EA access :-).
Jeremy.
Jeremy Allison [Tue, 23 Feb 1999 18:34:00 +0000 (18:34 +0000)]
Improved NT open mapping modes - added code to cope with FILE_APPEND_DATA
and FILE_READ_EA access rights.
Jeremy.
Jeremy Allison [Tue, 23 Feb 1999 18:28:27 +0000 (18:28 +0000)]
Added malloc fail check.
Jeremy.
Herb Lewis [Tue, 23 Feb 1999 00:13:34 +0000 (00:13 +0000)]
output [global] section header for testparm and swat.
Jeremy Allison [Mon, 22 Feb 1999 20:03:00 +0000 (20:03 +0000)]
Bugfix from <Robert.Dahlem@ffm2.siemens.de>.
A misplaced extern meant the smbpasswd file locking wasn't begin
done. Also ensure that SIGALRM interrupts system calls.
Jeremy.
Jeremy Allison [Mon, 22 Feb 1999 19:27:05 +0000 (19:27 +0000)]
Fix based on debugging work by "Mattias.Gronlund" <Mattias.Gronlund@sa.erisoft.se>.
UDP loopback sockets for oplock breaks shouldn't have SO_REUSEADDR set as it
allows two smbd's to bind to the same loopback udp port on Solaris. This
causes problems....
Jeremy.
Michael Warfield [Thu, 18 Feb 1999 04:34:10 +0000 (04:34 +0000)]
Got to the bottom of another weird one...
Piping the output of smbmount back to autofs/automount was causing the
automount process to hang. Reason was that automount was depending
on the pipe to close to continue on, rather than detecting the child
signal. This occured with debug enabled and the daemon process was not
closing the stdout process. Disabling debuging avoids the problem.
Debugging is turned off in the cvs repository and a warning placed over
the debugging option.
This syncs up the 2.0 branch with the main branch...
Jeremy Allison [Thu, 18 Feb 1999 00:49:36 +0000 (00:49 +0000)]
Fixed crash bug where smbd could crash if certain errors occur when
authenticating to secondary or subsequent PDC/BDC's.
Bug was cli_shutdow() was being called and then the cli struct
was being re-used without cli_initialize(). Fixed code stream
so the cli_initialize()/cli_shutdowns are correctly paired.
Jeremy.
Jeremy Allison [Wed, 17 Feb 1999 23:27:29 +0000 (23:27 +0000)]
3 *wonderful* bug fixes from Bob Boehmer <boehmer@worldnet.att.net>
In his own words :
The bugs are:
1. If a file "X" in the tarfile is an exact multiple of 512 bytes in length
then the extraction will fail for the file following "X" in the tarfile.
This is because get_file exits with the tar buffer pointing at the first
block of the next file if the current file was an exact multiple of 512
bytes (the tar buffer should be pointing at the final block of the file just
extracted).
2. If the tarfile is being fed through a pipe to smbclient then next_block
does not always work because fewer than the requested number of bytes may be
returned by the read call in next_block even though the pipe is still open.
3. If a file "X" in the tarfile is 0 bytes in length then the extraction
will fail for the file following "X" in the tarfile because next_block is
called prior to the call to get_file.
This might explain the smbtar problems people have been having
in the 2.x code.
Jeremy.
Jeremy Allison [Wed, 17 Feb 1999 02:37:56 +0000 (02:37 +0000)]
Small raft of changes needed for glibc 2.1 on Linux/Hurd.
From the patch by : Paul <PELaufer@csupomona.edu>.
configure configure.in: Added 64 bit test.
printing/printing.c: Changed strncat to safe_strcat - plus ensure null termination.
smbwrapper/wrapped.c: Added definition of NULL if missing.
client/smbmount.c: Added linux include file <===== MIKE PLEASE CHECK THIS !!!!
Jeremy.
Jeremy Allison [Wed, 17 Feb 1999 02:00:06 +0000 (02:00 +0000)]
Slightly modified version of name mangle patch from Ulrik Dickow <ukd@kampsax.dk>.
Exposes a parameter in name_map_mangle() to allow the caller to decide whether
to add the mapping to the mangle cache or not. This allows some callers not
to add to the cache in circumstances where the client is only sending the
mangled name (NT).
Jeremy.
Jeremy Allison [Tue, 16 Feb 1999 21:55:03 +0000 (21:55 +0000)]
Added file_chmod patch from Francois Gouget fgouget@multimania.com.
Jeremy.
Jeremy Allison [Tue, 16 Feb 1999 21:24:24 +0000 (21:24 +0000)]
Added code to ensure that, even if there is a constant stream of SMB
requests with no echos or pauses, that timeout processing is still
done at SELECT_TIMOUT intervals.
This is done by updating a counter of the number of SMB's processed,
and checking if the time has come for timeout processing every 200
SMB requests.
This should fix the 'smbd is not responding to a HUP signal' that
someone submitted a (rather complex) patch for.
Jeremy.
Jeremy Allison [Tue, 16 Feb 1999 19:20:55 +0000 (19:20 +0000)]
If we cannot open debug file - try /dev/console. If we cannot open
that - then abort as we cannot continue without a debug file handle.
Jeremy.
Jeremy Allison [Tue, 16 Feb 1999 19:14:18 +0000 (19:14 +0000)]
client/client.c client/smbmount.c: -t term page fix from fumiya@cij.co.jp
smbd/trans2.c: Is acls selected make sure filesystem info says we have acls.
Jeremy.
John Terpstra [Mon, 15 Feb 1999 11:03:23 +0000 (11:03 +0000)]
Added file extension. Oops, I left it off.
John Terpstra [Mon, 15 Feb 1999 10:58:48 +0000 (10:58 +0000)]
Added reference to UNIX-INSTALL document as suggested by Volker L.
Michael Warfield [Sun, 14 Feb 1999 23:29:03 +0000 (23:29 +0000)]
Added caching of user password if it was entered via getpass. We were already
caching the password if it came from the command line or from the environment.
This completes the set and deals with the bloody nusance when reconnecting
a connection after we have gone daemon... Grrr...
Added code to clean up the mount point following a catastrophic failure
during reconnect. The smbmount daemon was exiting but leaving the mount
point in an unusable state. If smbmount must exit following a reconnect
failure, we "unmount" the mount point and clean up mnttab. Currently,
the unmount works, fixing some really ugly I/O errors and failure when
trying to remount. The cleanup on mnttab still has problems. This
is better than what it was and doesn't break anything that wasn't broken
before, so I'm committing this in even with the mnttab problem. Will
commit the fix to that when I figure out what is busted there...
This brings the 2.0 branch smbmount program in line with the
main branch...
-mhw-
John Terpstra [Sun, 14 Feb 1999 02:50:00 +0000 (02:50 +0000)]
Updated info only.
John Terpstra [Sun, 14 Feb 1999 02:39:30 +0000 (02:39 +0000)]
Added copy of Win95 PlainPassword stuff - for those who really believe Win98
is not Win95! ;)
Added Registry config file to force Win9X to junk Cached file info on
Opening a file.
John Terpstra [Sun, 14 Feb 1999 02:35:35 +0000 (02:35 +0000)]
Added registry update file to stop all file oplocks and cacheing on NT.
Michael Warfield [Sat, 13 Feb 1999 18:35:54 +0000 (18:35 +0000)]
Fixed a glibc glitch in smbumount.c and reenabled some smbfs
debugging stuff in smbmount.c.
Jeremy Allison [Sat, 13 Feb 1999 01:39:15 +0000 (01:39 +0000)]
Added 'nt acl support' parameter - defaults to off.
smbd/nttrans.c: Added correct mapping from UNIX rwx to NT 'read' 'write' 'execute'.
Jeremy.
Jeremy Allison [Sat, 13 Feb 1999 01:38:10 +0000 (01:38 +0000)]
Added docs for new 'nt acl support' parameter.
Jeremy.
Herb Lewis [Fri, 12 Feb 1999 23:57:16 +0000 (23:57 +0000)]
fix for 5.3 compiles
Jeremy Allison [Fri, 12 Feb 1999 23:17:08 +0000 (23:17 +0000)]
Fix for duplicate name mangle bug (PR#13758). Help from Ulrik Dickow <ukd@kampsax.dk>
was *much* appreciated.
Jeremy.
Jeremy Allison [Fri, 12 Feb 1999 21:41:05 +0000 (21:41 +0000)]
Fixed problem where change to unsigned in dfree caused quota calculations
to screw up.
Jeremy.
Jeremy Allison [Fri, 12 Feb 1999 20:33:26 +0000 (20:33 +0000)]
Fixed bug in swat where remote password change could return a garbage
error string. Fix provided by master@iaas.msu.su.
Jeremy.
Matthew Chapman [Fri, 12 Feb 1999 03:05:06 +0000 (03:05 +0000)]
Don't bother compiling smbrun if we have execl.
Jeremy Allison [Fri, 12 Feb 1999 02:33:58 +0000 (02:33 +0000)]
I just displayed the first UNIX permission set as an NT ACL.......
Now the *real* work starts...
Jeremy.
Jeremy Allison [Fri, 12 Feb 1999 01:44:11 +0000 (01:44 +0000)]
Re-wrote the rpc_parse/parse_sec.c code to do true allocation on demand.
Luke - this is the new rpc malloc code I described to you.
Look at how it is used for guidence in re-writing the other
rpc calls.
The nt security descriptor stuff is now re-written to use this,
but doesn't give any advantage over the current SD stuff.
I am in the process of adding the mapping from UNIX perms
to displayed NT ACLs.
Jeremy.
Luke Leighton [Tue, 9 Feb 1999 21:41:48 +0000 (21:41 +0000)]
when multiple (separate) large rpc responses come in on a pipe,
prev_pdu_file_offset was not being re-initialised to zero.
Jeremy Allison [Tue, 9 Feb 1999 19:22:36 +0000 (19:22 +0000)]
Fixed crash bug when rotating logs. Also tidied up signal handler
to prevent looping bug. Crash bug was error in checking getuid()
rather than geteuid().
Jeremy.
Jeremy Allison [Tue, 9 Feb 1999 02:22:20 +0000 (02:22 +0000)]
Force Samba to return the correct NT error code if a :AFP_xxx resource or
data fork is opened. With NT4.0 SP3 this doesn't seem to be needed (the
file copies work correctly anyway) but it's exactly the same error that
NT generates when writing such a file to a FAT partition, so it can't
hurt....
Jeremy.
Jeremy Allison [Tue, 9 Feb 1999 00:29:15 +0000 (00:29 +0000)]
Fixed the little-endian unicode issue....
Jeremy.
Jeremy Allison [Mon, 8 Feb 1999 21:09:18 +0000 (21:09 +0000)]
configure configure.in: Fixed bugs in the SSL configure changes.
lib/util_unistr.c:
libsmb/smbencrypt.c:
rpc_parse/parse_misc.c:
rpc_parse/parse_prs.c:
rpc_parse/parse_reg.c:
rpc_parse/parse_rpc.c: Ensured that unicode strings are always little-endian
and are not re-converted for transport over the wire.
Jeremy.
Matthew Chapman [Mon, 8 Feb 1999 03:36:32 +0000 (03:36 +0000)]
NetBIOS scopes have been broken for a long time in Samba, which is some
indication of the number of people who use them!
(A loop in name_mangle requires len = 0 at the beginning of each name
component; it was being reset to 0 on a '.', which meant that the for loop
then incremented it to 1 by the beginning of the next component.)
Matthew Chapman [Mon, 8 Feb 1999 03:34:24 +0000 (03:34 +0000)]
Name types in the lmhosts file should be read in hexadecimal.
Jeremy Allison [Mon, 8 Feb 1999 01:43:07 +0000 (01:43 +0000)]
Ensure correct version is in cvs...
Jeremy.
John Terpstra [Sun, 7 Feb 1999 10:41:52 +0000 (10:41 +0000)]
Modified usage line to match man page.
John Terpstra [Sun, 7 Feb 1999 10:37:13 +0000 (10:37 +0000)]
Added more comment to Test 7 - re username%password
John Terpstra [Sun, 7 Feb 1999 10:24:44 +0000 (10:24 +0000)]
Added commentry to Test 8 - re: username and password on Samba server.
Samba Release Account [Sat, 6 Feb 1999 22:57:24 +0000 (22:57 +0000)]
preparing for release of 2.0.2
Andrew Tridgell [Sat, 6 Feb 1999 22:42:45 +0000 (22:42 +0000)]
fixed two compile warnings/errors
(not critical, just sloppy)
Jeremy Allison [Sat, 6 Feb 1999 22:10:41 +0000 (22:10 +0000)]
Fixed looping on SIGCLD problem. Missing break & missing CatchChild...
DAMN !!!!!
Jeremy.
Richard Sharpe [Sat, 6 Feb 1999 14:46:35 +0000 (14:46 +0000)]
Another quick fix for SSL support. Make sure that ssl.o is linked in with
everything :-)
Now seems to link :-)
Richard Sharpe [Sat, 6 Feb 1999 14:42:05 +0000 (14:42 +0000)]
Updates to make SSL stuff work.
Modified configure.in to ensure that SSL related stuff gets into
configure. Still a bit tacky, but a beginning. We check to
ensure that if -with-ssl is called, that the SSL includes and libraries
are on the system or we fall over.
It seems to build OK under Linux. Pity I could not have got it into
2.0.1
Samba Release Account [Sat, 6 Feb 1999 02:34:27 +0000 (02:34 +0000)]
preparing for release of 2.0.1
Jeremy Allison [Sat, 6 Feb 1999 02:09:59 +0000 (02:09 +0000)]
Updated for 2.0.1 release.
Jeremy.
Jeremy Allison [Sat, 6 Feb 1999 00:56:26 +0000 (00:56 +0000)]
Preparing for 2.0.1 - updated rpm packaging now we can do
--with-lockdir and --with-privatedir.
JEremy.
Jeremy Allison [Fri, 5 Feb 1999 21:32:16 +0000 (21:32 +0000)]
lib/util_sock.c: Added debug print if core read or write fails to give UNIX errno.
tests/* - Added #if defined(HAVE_UNISTD_H) to help with large file stuff.
Jeremy.
Jeremy Allison [Fri, 5 Feb 1999 02:19:07 +0000 (02:19 +0000)]
Added updates to help users of "bind interfaces only".
Jeremy.
Jeremy Allison [Fri, 5 Feb 1999 01:59:08 +0000 (01:59 +0000)]
Directory tidyups to help me track any problems.
Upped MAX_OPEN_DIRECTORIES to 256.
Ensured new directory handles are only allocated between 255 - MAX.
Stopped re-open of a directory about to be closed.
Log a message if a directory is to be closed and the bitmap
is not set.
Jeremy.
Jeremy Allison [Fri, 5 Feb 1999 00:49:27 +0000 (00:49 +0000)]
Allow large file support to succeed even if ino_t isn't 64 bits and there's
no ino64_t. AIX has large file support and doesn't use a 64 bit ino_t.
Jeremy.
Jeremy Allison [Thu, 4 Feb 1999 21:29:18 +0000 (21:29 +0000)]
Fixes for HPUX.
I have added a new define for configure - HAVE_EXPLICIT_LARGEFILE_SUPPORT.
This is tested for at the end of configure by the following check :
"As a gating factor for large file support, in order to
use <4GB files we must have the following minimal support
available.
long long, a 64 bit off_t or off64_t, a 64 bit ino_t or
ino64_t. If we don't have all of these then disable large
file support."
I have to do this as HPUX will detect the open64 etc.
functions in libc but *THEY DON'T WORK* before HPUX11.x.
The definitions in include/includes.h that set up large
file defines, and the calls in lib/system.c that use the explicit
large file functions are now depended on HAVE_EXPLICIT_LARGEFILE_SUPPORT
also being defined.
include/rpc_lsa.h: Upped number of groups to 96 as a temporary
measure.
Jeremy.
Richard Sharpe [Thu, 4 Feb 1999 06:10:43 +0000 (06:10 +0000)]
Minor wording change to the usage message for nmblookup.
Now makes more sense, at least to my sleep deprived brain.
Jeremy Allison [Thu, 4 Feb 1999 02:40:35 +0000 (02:40 +0000)]
Fixed nmblookup docs to match new code from Richard.
Jeremy.
Jeremy Allison [Wed, 3 Feb 1999 22:30:38 +0000 (22:30 +0000)]
Someone sent me a trace showing a client doing a readraw request
with a completely bogus FID (0xFFFF). This was causing the new
2.0 code to crash by indirecting through a null pointer, as
the old debug code wanted to print this fid value, and was
using fsp->fnum to do it - erroneously assuming that fsp
was non-null.
Jeremy.
Jeremy Allison [Wed, 3 Feb 1999 21:04:38 +0000 (21:04 +0000)]
A couple of malloc fail catches.
Jeremy.
Jeremy Allison [Wed, 3 Feb 1999 03:18:38 +0000 (03:18 +0000)]
Fixed bug where cli->timeout was in milliseconds, but open_sockets_out
was in seconds. This meant that PDC -> BDC failover took so long it essentiually
didn't happen.
include/client.h: Added explicit comment that cli->timeout should be milliseconds.
include/local.h: Changed LONG_CONNECT_TIMEOUT & SHORT_CONNECT_TIMEOUT to be milliseconds.
lib/smbrun.c: Tidied up fork failure catching.
lib/util_sock.c: Ensured timeout treated as milliseconds.
libsmb/clientgen.c: Added comment.
rpc_server/srv_pipe_hnd.c: Luke's changes.
smbd/chgpasswd.c: Paranoia code for EINTR around sys_waitpid.
Jeremy.
Herb Lewis [Wed, 3 Feb 1999 01:30:01 +0000 (01:30 +0000)]
add ISA define
Jeremy Allison [Wed, 3 Feb 1999 01:23:12 +0000 (01:23 +0000)]
Applying Luke's fix for the multiple DCE/RPC PDU problem to 2.0.
Jeremy.
Herb Lewis [Wed, 3 Feb 1999 01:17:11 +0000 (01:17 +0000)]
allow specifying different instriction set arch. for regular and -32
builds (needed for IRIX to build smbsh on mips4 ISA machines)
Richard Sharpe [Tue, 2 Feb 1999 23:01:08 +0000 (23:01 +0000)]
Minor changes to the usage message in neblookup to explain what
nmblookup -M -
does.
Also added a -T flag, which forces the translation of IP addresses to
names. Not [yet] the default.
Jeremy Allison [Tue, 2 Feb 1999 19:25:32 +0000 (19:25 +0000)]
Fixed bug for Turbo Pascal users where
echo >test.~pa
would create a file called : test.~pa.~pa.~pa.~pa
Jeremy.
Jeremy Allison [Tue, 2 Feb 1999 00:47:53 +0000 (00:47 +0000)]
Fixed devide by zero on SIGHUP bug if no stat cache lookups
have been done.
Jeremy.
Herb Lewis [Mon, 1 Feb 1999 23:37:14 +0000 (23:37 +0000)]
Turns out that Win98 *can* display large volume sizes, even with
the non-trans2 call - if you set the return parameters right.
Jeremy.
Michael Warfield [Mon, 1 Feb 1999 22:03:11 +0000 (22:03 +0000)]
`
Jeremy Allison [Mon, 1 Feb 1999 21:03:43 +0000 (21:03 +0000)]
locking/locking_shm.c: Added printf to help track down negative share mode problem.
printing/print_svid.c: Fix for DG/UX lpstat.
smbd/dfree.c: Added 'max_four_gig' parameter to normalize to 4GB when asked
(used to be the default).
smbd/open.c smbd/reply.c smbd/trans2.c:
Changed to allow a <4GB volume size to be returned on a trans2
qfsinfo level 1 call. NT clients refuse to do the correct call to Samba servers
(Herb recons it's the unicode bit) - this allows even a level 1 to return a
volume size up to
9444732961341243916800 bytes (should be enough for now,
until we get the unicode support added :-).
Jeremy.
Matthew Chapman [Sun, 31 Jan 1999 22:04:08 +0000 (22:04 +0000)]
Be very careful allowing "utility" functions to modify input strings.
check_mangled_cache was not always restoring the filename correctly
before returning False.
Thanks to Christoph Scheeder <christoph.scheeder@Scheeder.de> for tracking
this down.
Jeremy Allison [Fri, 29 Jan 1999 02:43:23 +0000 (02:43 +0000)]
Fixed large (<4GB) filesystem size reporting.
Jeremy.
Jeremy Allison [Fri, 29 Jan 1999 02:40:54 +0000 (02:40 +0000)]
Fixed large (>4GB) filesystem size reporting.
Jeremy.