IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
This macro is not setting torture failure, leading to errors instead
of failures. Use torture_assert_ntstatus_(ok|equal)* macros.
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
sizeof(buf) is 5. On FreeBSD10/clang this overwrites "ret". Not good.
Same as 574750777a
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Dec 9 19:43:31 CET 2014 on sn-devel-104
sizeof(buf) is 5. On FreeBSD10/clang this overwrites "ret". Not good.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Fri Nov 28 13:30:18 CET 2014 on sn-devel-104
Shows attribute(stat) access open can create a file,
and subsequent attribute(stat) opens don't break oplocks.
Can be extended to explore more varients.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
The issue is that previously bad_ea_name[5] was the last element on
the array, and so when we later did a strlen() on it, we read past the
end of the stack array. We need bad_ea_name[5] to be the second-last
element, followed by the \0 placed there by the strlcpy().
Found by AddressSanitizer
Change-Id: I871c08200aa2591c612dfa44da92b83132f83d88
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Kamen Mazdrashki <kamenim@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Thu Sep 11 08:50:16 CEST 2014 on sn-devel-104
This changes the sizeof(buf) from sizeof(void *), 8 on 64-bit machines, to sizeof("test") (eg 5).
Found by AddressSanitizer
Andrew Bartlett
Change-Id: I01f18b35c041f3b16be9f6da8ae5d1917d7e24d9
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Kamen Mazdrashki <kamenim@samba.org>
MS-CIFS requires a one byte pad to guarantee 16 bit alignment of the
data:
Pad (1 byte): This field is optional. When using the NT LAN Manager
dialect, this field can be used to align the Data field to a 16-bit
boundary relative to the start of the SMB Header. If Unicode strings are
being used, this field MUST be present. When used, this field MUST be
one padding byte long.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Needed as there was a proposal to re-architect
our multi-lock to dispense with lock order precedence,
which isn't how Windows does it (unfortunately,
as the new code would have been cleaner :-).
Tested against the Win2k12 SMB1 implementation.
This test is designed to show that
lock precedence on the server is based
on the order received, not on the ability
to grant. For example:
A blocked lock request containing 2 locks
will be satified before a subsequent blocked
lock request over one of the same regions,
even if that region is then unlocked. E.g.
(a) lock 100->109, 120->129 (granted)
(b) lock 100->109, 120-129 (blocks)
(c) lock 100->109 (blocks)
(d) unlock 100->109
lock (c) will not be granted as lock (b)
will take precedence.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Tue Jul 8 10:16:59 CEST 2014 on sn-devel-104
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Wed Jun 25 13:37:24 CEST 2014 on sn-devel-104
Adds a new test to raw.open.
Opens a file with SHARE_NONE, writes 1 byte at offset 1023,
attempts a second open with r/w access+truncate disposition,
then checks that open fails with SHARING_VIOLATION, and
the file is not truncated (is still size 1024). Correctly
detects the bug and fixed smbd for me.
https://bugzilla.samba.org/show_bug.cgi?id=10671
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
NTTIME has a nanosecond resolution. We should be tolerate if the system
is busy writing and reading the value. To reproduce this problem just
run the test under valgrind.
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Checks against a file with attribute READONLY, and
a security descriptor denying WRITE_DATA access.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: David Disseldorp <ddiss@suse.de>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon Nov 4 23:10:10 CET 2013 on sn-devel-104
smbd broke to none twice. Make sure this won't happen again :-)
This used to happen before the MSG_SMB_BREAK_RESPONSE merge. In
process_oplock_break_message we did not call remove_oplock, which would
have prevented this.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Wed Oct 23 14:06:13 CEST 2013 on sn-devel-104
The level we have to break to depends on the create disposition of the
second opener. If it's overwriting, break to none. If it's not, break
to level2.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
This is what Windows does in this case, we don't survive that. We break
to LEVEL2 here. Fixes and more precise test to follow.
We don't survive this anymore. Re-enable later.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
We can directly use smb_raw_open() to open a handle to a named pipe.
This avoids the need for the layer violation functions
dcerpc_smb_tree() and dcerpc_smb_fnum().
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
With FAKE_LEVEL_II_OPLOCKS around we did not grant LEVEL2 after
a NO_OPLOCK file got written to. Windows does grant LEVEL2 in this
case. With the have_level2_oplocks in brlocks.tdb we can now grant LEVEL2
in this case as well.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
If delete_on_close is set, there is no oplock break. Check that.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This flag existed to break an exclusive or batch oplock in just one
instead of two steps down to "no oplock" when we did an allocation or file
size change. Running raw.oplock against W2k12 differs in this respect
from W2k3: W2k12 takes two steps (via level2) to break to none. This
removes the special flag that we only had for compatibility with systems
older than W2k12...
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Sep 6 00:47:07 CEST 2013 on sn-devel-104
Add torture tests to probe the set of invalid
Windows EA names.
Bug 9992 - Windows error 0x800700FE when copying files with xattr names containing ":"
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Fri Jul 19 11:50:25 CEST 2013 on sn-devel-104
As we compare string make sure we have the null terminator.
Found by valgrind.
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Mar 20 21:53:20 CET 2013 on sn-devel-104
Bad name in SMB1 openX can cause a crash in iconv inside glibc.
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon Oct 1 23:29:25 CEST 2012 on sn-devel-104
Error was assigned to a variable that was not returned.
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Fri Sep 14 14:05:20 CEST 2012 on sn-devel-104
The extension of this test is to create an extended attribute, so we
can confirm that the easize field on a stream actually refers to the
parent file.
This has been run against Windows 7.
Andrew Bartlett
