torture3: Fix bug 10687

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): Thu Jul  3 19:07:15 CEST 2014 on sn-devel-104

ctdb-tests: Fix racy test for debugging hung scripts

Debugging can still be running when a monitor event times out and
scriptstatus output changes.

When debugging a hung script to a log file, write to a temporary file
and move the temporary file over the log file when done.  The test
then waits for the log file to appear.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Thu Jul  3 08:19:23 CEST 2014 on sn-devel-104

ctdb-scripts: Always print footer when debugging hung script

There shouldn't be an early exit for the "init" event.  Just make the
"ctdb scriptstatus" call conditional.

While here, move the comment about only running a single instance to
be near locking code.  The comment is more useful there.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>

s3-libnet: Improve error message.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Jul  2 16:54:10 CEST 2014 on sn-devel-104

s3: smbd: Locking, fix off-by one calculation in brl_pending_overlap().

Consider:

lock = start=110,size=10
pend_lock = 100, size=10

Do not overlap. However,

(lock->start <= pend_lock->start + pend_lock->size)
     110             100                10

is true, so it returns true (overlap).

lock->start <= pend_lock->start + pend_lock->size

should be:

lock->start < pend_lock->start + pend_lock->size

BUG: https://bugzilla.samba.org/show_bug.cgi?id=10685

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Wed Jul  2 10:18:17 CEST 2014 on sn-devel-104

make: Add gdbtestenv target.

This starts the sever components in a xterm with gdb.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s3:winbindd - fix bad bugfix for bug #10280 - winbind panic if AD server is down.

Previous bug fix reversed the sense of the test for out of memory.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=10280

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

smbd: Fix some typos

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): Wed Jul  2 02:25:47 CEST 2014 on sn-devel-104

smbd: Remove an unused variable

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

smbd: Clarify smb2 lock checks

When reading the code it was not immediately clear to me how one of the
conditions in [MS-SMB2] 3.3.5.14.2 was satisfied. A separate loop to me
is clearer and given that we don't expect thousands of locks in a single
call also not significantly less efficient.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: Add a check to verify MS-SMB2 3.3.5.14.2

If we have more than one lock and there is any blocking lock, we need
to fail with NT_STATUS_INVALID_PARAMETER. At a quick glance I did not
find this tested, so add it.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libcli: Remove an unused variable

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 Jul  1 00:43:18 CEST 2014 on sn-devel-104

smbd: validate_lock_entries does not need mem_ctx anymore

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

smbd: Simplify validate_lock_entries

Signed-off-by: Volker Lendecke <vl@samba.org>

smbd: Factor out brl_delete_lock_struct

memmove calculations are never nice, and this is going to be used in
validate_lock_entries soon :-)

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

smbd: Simplify validate_lock_entries

... saves >100 bytes object code :-)

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture3: Add some brlock entries in cleanup2

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

smbstatus: Fix an uninitialized variable

We only print valid share mode entries, stale ones don't count. In
traverse, let the callback decide about staleness.

https://bugzilla.samba.org/show_bug.cgi?id=10680

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libcli: Make cli_smb2_close_fnum async

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

lib: Align nt_time_to_unix_timespec with unix_timespec_to_nt_time

Both take and return values now

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

lib: Align unix_timespec_to_nt_time with nt_time_to_unix_timespec

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libcli: Make smb2cli_create return blobs

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

vfs_snapper: use a talloc hierarchy for arrays

Currently the config and snapshot array entries are freed individually.
A talloc hierarchy can be used to avoid this behaviour.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon Jun 30 21:41:13 CEST 2014 on sn-devel-104

vfs_snapper: don't redefine GMT_FORMAT macro

Use existing header definition.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

doc: add vfs_snapper manpage

Document usage of the snapper VFS module, detailing permissions required
to list and access snapshots.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

vfs: add previous file version support for vfs_snapper

Windows attempts to access previous file versions by first issuing an
FSCTL_SRV_ENUMERATE_SNAPSHOTS request, and then using the corresponding
snapshot timestamps in subsequent path based requests.

This change sees vfs_snapper provide snapshot enumeration data via the
get_shadow_copy_data VFS hook. Path based operations are also
intercepted, with @GMT-$time components converted to snapper .snapshot/#
paths.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

vfs: add vfs_snapper module

Provides an interface for accessing snapshots exposed by Snapper. The
module communicates with snapperd on the local machine using the D-Bus
interface.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

smbd: Make messaging_send_to_children static

This function is only used in server.c

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Jun 30 17:20:00 CEST 2014 on sn-devel-104

smbd: Use common function for FORCE_DIS and KILL_CLIENT_IP in parent

The parent smbd process only forwards the message to the child
processes. Use a common function instead of two separate ones that do
the same thing.

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

s3: SMB2 : Fix leak of blocking lock records in the database.

Based on a fix from Hemanth Thummala <hemanth.thummala@gmail.com>
Bug #10673 - Increasing response times for byte range unlock requests.

The previous refactoring makes it obvious we need to call
remove_pending_lock() in all places where we are returning
from the SMB2 blocking lock call.

https://bugzilla.samba.org/show_bug.cgi?id=10673

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ira Cooper <ira@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Jun 30 14:59:16 CEST 2014 on sn-devel-104

s3: smb2: Simplify logic in reprocess_blocked_smb2_lock().

SMB2 blocking locks can only have one lock per request, so
there can never be any other locks to wait for.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ira Cooper <ira@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

s3: smb2: Remove unused code from remove_pending_lock().

SMB2 blocking locks can only have one lock per request, so
there can never be any previous locks to remove.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ira Cooper <ira@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

ctdb-packaging: Update configure.rpm with minimum library versions

This makes configure command consistent with spec file.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Mon Jun 30 11:23:18 CEST 2014 on sn-devel-104

ctdb-packaging: Minimum required tevent library is 0.9.16

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

s3-shadow-copy2: Protect against already converted names

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Fri Jun 27 09:03:53 CEST 2014 on sn-devel-104

s3-shadow-copy2: Add more debugs

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>

s3-shadow-copy2: Fix incorrect case submounts

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>

s3-shadow-copy2: Remove TODO and fix comments

The patch "s3-shadow-copy2: Fix dir/@GMT-2012.10.15-13.48.43 form of
paths" takes care of a case marked as TODO, remove it and adjust the
comments accordingly.

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

s3-shadow-copy2: Fix dir/@GMT-2012.10.15-13.48.43 form of paths

The previous clause in shadow_copy2_strip_snapshot would only handle @GMT-
at the end of a pathname if it was the *only* pathname component. XP
seems to send @GMT- at the end under certain circumstances even with a
path prefix.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>

s3-shadow-copy2: Add extreme debug output to shadow_copy2_strip_snapshot

This is sooo hairy to debug when things go wrong :-(

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>

dbwrap: Print wait times with full precision

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>

smb2: Fix smb2_lease_state

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Thu Jun 26 17:30:00 CEST 2014 on sn-devel-104

tdb: defragment the freelist in tdb_allocate_from_freelist()

While we are traversing the freelist anyways, merge a record
with the left if it is also a free list record.

That partially makes up for the fragmentation introduced by
the lack of merging with right records in tdb_free().

Note there is a potential slight downside:
If the left record we merge the current record into was earlier
in the chain and has hence already been met in traverse,
then we can not use the enlarged record even if it might be
a new best fit.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Thu Jun 26 12:16:03 CEST 2014 on sn-devel-104

tdb: add "freelist_size" sub-command to tdbtool

With the new code, this has the side effect of
merging adjacent records in the freelist if the
database is not read-only.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: use tdb_freelist_merge_adjacent in tdb_freelist_size()

So that we automatically defragment the free list when freelist_size is called
(unless the database is read only).

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: add tdb_freelist_merge_adjacent()

This is intended to be called to reduce the fragmentation in the
freelist. This is to make up the deficiency of the freelist
to be not doubly linked. If the freelist were doubly linked,
we could easily avoid the creation of adjacent freelist entries.
But with the current singly linked list, it is only possible
to cheaply merge a new free record into a freelist entry on the left,
not on the right...

This can be called periodically, e.g. in the vacuuming process
of a ctdb cluster.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: add utility function check_merge_ptr_with_left_record()

Variant of check_merge_with_left_record() that reads the record
itself if necessary.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: simplify tdb_free() using check_merge_with_left_record()

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: add utility function check_merge_with_left_record()

Check whether the record left of a given freelist record is
also a freelist record, and if so, merge the two records.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: improve comments for tdb_free().

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: factor merge_with_left_record() out of tdb_free()

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: fix debug message in tdb_free()

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: reduce indentation in tdb_free() for merging left

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: increase readability of read_record_on_left()

by using early returns and better variable names,
and reducing indentation.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tdb: factor read_record_on_left() out of tdb_free()

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

torture4: Add a little test that truncate actually works :-)

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

smbd: Remove 2 indentation levels

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

s4: torture: Add regression test case for #10671 - Samba file corruption as a result of failed lock check.

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>

s3: smbd - Prevent file truncation on an open that fails with share mode violation.

Fix from Volker, really - just tidied up a little.
The S_ISFIFO check may not be strictly neccessary,
but doesn't hurt (might make the code a bit more complex
than it needs to be).

Fixes bug #10671 - Samba file corruption as a result of failed lock check.

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>

s3: smbd - fix processing of packets with invalid DOS charset conversions.

CVE-2014-3493

Bug 10654 - Segmentation fault in smbd_marshall_dir_entry()'s SMB_FIND_FILE_UNIX handler

https://bugzilla.samba.org/show_bug.cgi?id=10654

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Jun 25 03:47:55 CEST 2014 on sn-devel-104

s3: nmbd: Fix bug 10633 - nmbd denial of service

The Linux kernel has a bug in that it can give spurious
wakeups on a non-blocking UDP socket for a non-deliverable packet.

When nmbd was changed to use non-blocking sockets it
became vulnerable to a spurious wakeup from poll/epoll.

Fix sys_recvfile() to return on EWOULDBLOCK/EAGAIN.

CVE-2014-0244

https://bugzilla.samba.org/show_bug.cgi?id=10633

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

ctdb-build: Instead of default test_wrap, install fixed test_wrap

tests/scripts/test_wrap script is updated based on the configured paths.
This should be installed instead of the stock version in the source.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Tue Jun 24 09:37:39 CEST 2014 on sn-devel-104

ctdb-build: Fix sed expression to protect '\'

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb-build: Use correct path variables for ctdb_run_tests.sh

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb-build: Use CTDB_ETCDIR instead of ETCDIR/ctdb

This avoids hardcoding path components in source.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb-build: Use CTDB_RUNDIR instead of VARDIR/run/ctdb

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb-build: Add special target to get build version

This avoids the need to build a tarball to get version information.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb-build: Allow waf to build rpms without configure first

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

torture: Remove confusing reference to (1 sec == 0.25)

The write should never update the time, so the fraction of the write
time delay we use is not important.

Andrew Bartlett

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Jun 24 01:44:06 CEST 2014 on sn-devel-104

torture: tidy up whitespace for clarity

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

selftest: Run base.delaywrite against plugin_s4_dc only, and with kerberos for faster connections

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture-base.delaywrite: assert the for a delayed write, that it just takes longer than our configured delay

This removes the hardcoded TIMEDELAY_SECS that was then made variable
by the confusing "secs" variable

Andrew Bartlett

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: Only error if the write time was updated is less than the expected delay

The previous test was far, far too tight, it was in seconds 1/4 of the
fraction of the normal delay we had configured Samba to use so (1/4) *
(500 000 / 2000 000) = 1/16 (sec).  This margin appears to just be too
tight for our loaded test server.

Andrew Bartlett

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: Store writetimeupdatedelay in a double to allow easier comparison

In particular, this avoids a comparison with
double diff = timeval_elapsed() being promoted to an integer.

Andrew Bartlett

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: convert raw.mux to use torture_comment() macros instead of printf()

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Kamen Mazdrashki <kamenim@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: convert raw.mux to use torture_assert() macros

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: convert CHECK_STATUS into torture_assert_ntstatus_equal in raw.mux tests

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Kamen Mazdrashki <kamenim@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

selftest: Only run samba4.rpc.samr.passwords.{lockout,badpwdcount} once

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: Use torture_assert() macros in rpc.samr test

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Kamen Mazdrashki <kamenim@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

s3: Fix fsctl_validate_neg_info to pass MS compliance suite.

It turns out that all the client and server need to agree on is what
protocol should have been negotiated.  If they disagree, they should
disconnect.  The contents of the list of protocols used during
negotiate and during FSCTL_VALIDATE_NEGOTIATE_INFO do not need to match.

Signed-off-by: Ira Cooper <ira@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Mon Jun 23 14:28:25 CEST 2014 on sn-devel-104

s3: Refactor smbd_smb2_request_process_negprot

Breakout smb2_protocol_dialect_match to support future work in
fsctl_validate_neg_info.

Signed-off-by: Ira Cooper <ira@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

torture3: Reproducer for bug 10593

We panic if we get an oplock break response for a cancelled create request

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Sat Jun 21 23:05:47 CEST 2014 on sn-devel-104

smbd: Fix bug 10593

Bug 10593 is a panic that happens if we get an oplock break reply via
dbwrap_watch for which we can't find the SMB request anymore. This
error condition can legally happen when a client cancels the create
request before the oplock break response comes in. This patch drops the
dbwrap_watch_send request waiting for the oplock break when the request
is cancelled. Yet another talloc hierarchy problem, but if done right,
talloc hierarchies can make rundown of state easy :-)

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

smbd: First watch, then defer

We exit if any of these if-statement fails, so a simple swap should not
make a difference.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

smbd: Store "struct deferred_open_record" instead of anonymous data on pml

The main point is to get a talloc parent that will go away when the
request is cancelled

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Make cli_ntcreate cancellable

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Make cli_ntcreate1 cancellable

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Make cli_smb2_create_fnum cancellable

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Make smb2cli_create cancellable

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Enable oplocks for smb2 cli_ntcreate

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Align cli_ntcreate with other sync wrappers

... saves 5 lines :-)

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: remove smb2 switch from cli_ntcreate

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Replace async cli_ntcreate by cli_create

Done by rename cli_ntcreate_send/recv to cli_ntcreate1_send/recv and
cli_create_send/recv to cli_ntcreate_send/recv

Possibly cli_create might be the better name, but I am sooo used to
cli_ntcreate() that I don't really want to rename this ;-)

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Add cli_create_send/recv

Async wrapper around smb1 and smb2 create

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

libsmb: Make cli_smb2_create_fnum async

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

torture: add FSRVP share snapshot ACL test

The new test_fsrvp_share_sd test sets a unique ACL on the base share,
and then confirms that snapshot shares carry the same ACL.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Sat Jun 21 15:10:12 CEST 2014 on sn-devel-104

torture: add FSRVP message sequence timeout test

Sleep at various points in the FSRVP snapshot creation state machine,
and confirm that the state timeout is reflected in subsequent server
responses.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

torture: add timeout sleeps to fsrvp create helper

Attempt to trip message sequence timeouts at various points in the FSRVP
shadow-copy creation state machine.

The default timeout-injection sleep durations correspond to those
documented in MS-FSRVP (+500ms). They can also be manually set using the
"fss: sequence timeout" parameter.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

torture: validate FSCTL_SRV_ENUMERATE_SNAPSHOTS response

Check the NumberOfSnapShotsReturned and SnapShotArraySize fields in the
FSCTL_SRV_ENUMERATE_SNAPSHOTS response match expected values.

This is a regression test for bso#10549.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

nss_wrapper: Fix some "discarding const" warnings

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>