pygensec: Use pytalloc_steal() in gensec_start_{client,server}()

This is better than casting to get to the pytalloc_Object structure directly

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pyparam: Use pytalloc_BaseObject_PyType_Ready()

This changes pyparam to use talloc.BaseObject() just like the PIDL output

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

py_passdb: Use pytalloc_BaseObject_PyType_Ready()

This changes py_passdb to use talloc.BaseObject() just like the PIDL output

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pycredentials: Use pytalloc_BaseObject_PyType_Ready()

This changes pycredentials to use talloc.BaseObject() just like the PIDL output

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pytalloc: Add pytalloc_BaseObject_PyType_Ready() wrapper

This avoids the need for the caller to set tp_base and tp_basicsize and
so removes those as possible errors.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pyparam: Do not use pytalloc_Object directly

This type should not be used directly, it should have been made private
to pytalloc.  This then allows removal of the (PyCFunction) cast

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pypassdb: Do not use pytalloc_Object directly

This type should not be used directly, it should have been made private
to pytalloc.  This then allows removal of the (PyCFunction) cast

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pycredentials: Remove PyCredentialCacheContainerObject

We can call pytalloc_reference() and avoid having this in the header file

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pycredentials: Do not use pytalloc_Object directly

This type should not be used directly, it should have been made private
to pytalloc

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

selftest: dbcheck should not be marked flapping

The primary cause of the flapping was due to the objectclass
sort routine being non-deterministic.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

dbcheck: Avoid spurious warnings in dbcheck due to objectclass sorting

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11433
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

dbcheck: Fix incorrect/duplicate attrid in replPropertMetaData

If custom schema is used in a replicated DC environment, these are created as soon as
an attribute is modified on more than one DC.  We have to remove these.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11443
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

selftest: Update release-4-1-0rc3 with more test records

CN=ops_run_anything2,OU=SUDOers,DC=release-4-1-0rc3,DC=samba,DC=corp

This will be modified during the dbcheck to show that new
versions of Samba will reset the attid correctly

CN=ops_run_anything3,OU=SUDOers,DC=release-4-1-0rc3,DC=samba,DC=corp

This will not be modified, and shows how a 4.1 DC without
replication would record custom schema objects.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

rpc_server/drsuapi: Block replication of incorrect/duplicate attrid in replPropertMetaData

If custom schema is used in a replicated DC environment, these are created as soon as
an attribute is modified on more than one DC.  We have to prevent replication
as otherwise we will corrupt the client replica state.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11443
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

repl_meta_data: Correctly use msDS-IntId for custom schema, not the prefixMap value

We must, when dealing with custom schema, respect the msDC-IntId value recorded
in the schema.  If we do not, then we will create multiple replPropertyMetaData
records for the one attribute.  This may cause confusion during replication.

This fixes the issue by always calling dsdb_attribute_get_attid() to obtain
the correct local (32 bit integer) attribute ID

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11443

pidl: Use a tmp_ctx helper variable

This is so we free the ndr_push_struct_blob() return value after
we make it into a string

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pidl: Use the $mem_ctx helper variable

This is already set to pytalloc_get_mem_ctx(py_obj)

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pidl: Fix our python reference handling

The new talloc.BaseObject allow us to hold a talloc context per
python object (there may be many referring to the same C object)
and the talloc context that the actual object pointer is under.

Another advantage is that talloc.BaseObject(), has less of
an ABI surface.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pyrpc: Clarify failure mode after pytalloc_reference_ex() improvements

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

talloc: Bump version number

The removal of the macros and replacement with proper functions
is a API, but not ABI break.  Only code that incorrectly
used the structure either in function signatures or
to access the members directly will need to be modified
before being built against this version of talloc.

Andrew Bartlett

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pytalloc: Add new BaseObject

This new object not only avoids the ABI issues of talloc.Object
it stores one more pointer, being the start of the array, and
so can be used to fix the PIDL bindings/talloc refcount issue.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pytalloc: add a _pytalloc_get_type() helper function and generate PyExc_TypeError on mismatch

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

talloc: add _pytalloc_get_ptr/_pytalloc_get_mem_ctx helper functions

This allows us to check which type is involved, and dereference
that type correctly

Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

talloc: Improve testsuite by avoiding path issues

We now rely on waf to tell us where the helper binary is.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pyrpc: Add warning about abuse of py_return_ndr_struct()

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pydsdb: Fix returning of ldb.MessageElement.

This object is not based on pytalloc_Object and so this causes
a segfault (later a failure) when the struct definitions diverge.
We must also not reuse the incoming ldb_message_element as a talloc
context and overwrite the values, instead we should create a new
object and return that.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

pyldb: Add warning about pyldb_MessageElement_AsMessageElement()

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>

libnet: make Kerberos domain join site-aware

When joining a domain using Kerberos authentication, create a
configuration file for the Kerberos libs to prefer on-site
domain controllers, without relying on the winbindd Kerberos
locator, which many not be operational at this stage.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11769

Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Mar  8 01:30:35 CET 2016 on sn-devel-144

dsgetdcname: fix flag check

Fix the check for zero requseted flags.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11769

Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

dsgetdcname: return an IP address on rediscovery

When dsgetdcname return its result based on discovery
process (instead of retrieving cached value), always
return the found server's IP address in dc_address field,
rather than its netbios name.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11769

Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

idmap_autorid: Protect against corrupt databases

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

idmap_autorid: Fix a use-after-free

Parsing the domain_range_index references data.dptr

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

s3: smbd: Change open_streams_for_delete() to take a struct smb_filename *.

Prepare for changing vfs_streaminfo to do the same.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Mon Mar  7 21:12:56 CET 2016 on sn-devel-144

s3: smbd: Change open_streams_for_delete() to take a struct smb_filename *.

Prepare for changing vfs_streaminfo to do the same.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Change open_streams_for_delete() to static.

Not used outside of open.c

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Change delete_all_streams() to take a const struct smb_filename *.

Prepare for changing the interface to vfs_streaminfo().

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: vfs_fruit. If we have an fsp, use it in preference to a pathname in vfs_streaminfo.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: Avoid a gencache_parse when setting a delete marker

We know that we want to put something into _notrans, no point in
doing another round trip into gencache.tdb.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: Avoid looking at fcntl'ed gencache.tdb

gencache_notrans.tdb is much cheaper to look at than gencache.tdb because it's
mutexed and thus avoids expensive fcntl locks. This patch aggressively uses the
shared _notrans tdb for both positive and negative entries. It's a replacement
for the memcache copy in every process that was removed a few patches ago.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: Simplify gencache_pull_timeout callers

gencache_pull_timeout used to point at the "/" right after the timeout.  None
of the callers was interested in the "/", they are interested in the payload.

Increment the endpointer in gencache_pull_timeout and rename it.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: Make gencache_pull_timeout look at uint8_t

At this point we're still looking at TDB_DATA. This patch moves the casts to a
more appropriate place.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: Remove memcache from gencache

The main reason for this was to avoid access to the fcntl-governed transaction
based gencache.tdb. A later patch will make this unnecessary by filling
gencache_notrans more aggressively.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: Fix a typo in gencache

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: Simplify gencache_del

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

lib: skip deleted entries in gencache_iterate

"net cache flush" can give nasty error messages like

Couldn't delete entry! key = IDMAP/UID2SID/12345

These happen when there's an already deleted entry in
gencache_notrans.tdb, indicated by a 0 timeout. This happens if two
gencache_del function calls have happened right after the other and a
gencache_stabilize has not wiped them.

In gencache_iterate, don't show these deleted entries

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

krb5_wrap: Do not use deprecated KRB5 functions

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): Mon Mar  7 17:57:39 CET 2016 on sn-devel-144

ctdb-packaging: Set --libexecdir in RPM spec file

If the files section uses %{_libexecdir} then CTDB must also be
configured to build and install with the same path, otherwise rpmbuild
can fail due to a mismatch.  "rpmbuild --showrc" indicates that the
default %configure command sets:

        --libexecdir=%{_libexecdir} \

A mismatch will occur on SUSE systems, where SLES 12 and OpenSUSE 12
set _libexecdir to %{_exec_prefix}/lib.

The failure was initially seen when testing on Debian where
_libexecdir is set to %{_prefix}/lib/x86_64-linux-gnu, though Debian
is an unlikely platform for building RPMs...

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): Mon Mar  7 14:43:15 CET 2016 on sn-devel-144

ctdb-locking: Use real-time only for actual record or DB locking

Use real-time priority only for obtaining record and database locks.
Do not open databases with real-time priority as it can cause thundering
herd on fcntl lock while opening tdb database.  Also relinquish real-time
priority after the lock is obtained.

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 Mar  7 11:29:00 CET 2016 on sn-devel-144

ctdb-takeover: Inform clients when dropping all IP addresses

CTDB releases all IPs in following cases: starting up, shutting down,
node gets banned, node does not come out of recovery for a long time.
Always inform samba when CTDB releases IP addresses.

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

ctdb-takeover: Do not kill smbd processes on releasing IP

CTDB already notifies Samba with RELEASE_IP message.  Samba can take
appropriate action based on that.

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

Remove posixAccount from samdb.py

Signed-off-by: Rowland Penny rpenny@samba.org
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Garming Sam <garming@samba.org>
Autobuild-Date(master): Mon Mar  7 08:13:42 CET 2016 on sn-devel-144

ctdb-packaging: Drop changelog section from RPM spec file

This is unmaintained and misleading.

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): Mon Mar  7 05:00:15 CET 2016 on sn-devel-144

g_lock: Remove some unneeded includes

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ira Cooper <ira@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Sun Mar  6 19:34:42 CET 2016 on sn-devel-144

lib: Avoid "includes.h" in dbwrap.c

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

lib: Introduce SERVER_ID_BUF_LENGTH

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

smbd: Clean up the logic inside vfs_chown_fsp() to prevent future security issues.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Sat Mar  5 12:53:11 CET 2016 on sn-devel-144

VFS: Modify lchown to take a const struct smb_filename * instead of const char *

Preparing to reduce use of lp_posix_pathnames().

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

VFS: Modify chown to take a const struct smb_filename * instead of const char *

Preparing to reduce use of lp_posix_pathnames().

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

VFS: vfs_netatalk. Fix wrong VFS call used inside atalk_lchown()

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

vfs_glusterfs: Fix use after free in AIO callback.

The wrapper->state pointer is not getting NULLed during free
allowing use of freed memory, causing a crash.

Thanks to Red Hat for discovering this issue.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11774

Signed-off-by: Ira Copper <ira@samba.org>
Reviewed-by: Poornima G <pgurusid@redhat.com>
Tested-by: Christopher Blum <cblum@redhat.com>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Mar  5 09:08:53 CET 2016 on sn-devel-144

source3: Honor the core soft limit of the OS.

We should honor the soft limits set by the operating system.

In any case, 16M doesn't make a useful coredump for modern
Samba.

Signed-off-by: Ira Cooper <ira@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Mar  5 00:39:48 CET 2016 on sn-devel-144

s3/rpc_server/srvsvc/srv_srvsvc_nt.c: change snum to signed int

snum is used to store the result of the find_service() call which
returns a signed int, with -1 being an error.

with an uint32_t the error check (< 0) is always false.

Signed-off-by: Aurelien Aptel <aaptel@suse.com>
Reviewed-by: Volker Lendecke <Volker.Lendecke@SerNet.DE>
Reviewed-by: Jeremy Allison <jra@samba.org>

s3,s4 torture/denytest: fix possible infinite loop

Signed-off-by: Aurelien Aptel <aaptel@suse.com>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Reviewed-by: David Disseldorf <ddis@suse.de>
Autobuild-User(master): Alexander Bokovoy <ab@samba.org>
Autobuild-Date(master): Fri Mar  4 21:23:45 CET 2016 on sn-devel-144

s4/torture/libnetapi/libnetapi_user.c: fix typo

fixed misplaced parenthesis and wrong sizeof().

Signed-off-by: Aurelien Aptel <aaptel@suse.com>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Reviewed-by: David Disseldorf <ddis@suse.de>

ctdb-common: For AF_PACKET socket types, protocol is in network order

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11770

From man page of packet(7):

                                             protocol is the  IEEE  802.3
   protocol  number  in  network  byte  order.  See the <linux/if_ether.h>
   include file for a list of allowed protocols.  When protocol is set  to
   htons(ETH_P_ALL),  then all protocols are received.

Protocol argument was changed from network order to host order wrongly
in commit 9f8395cb7d49b63a82f75bf504f5f83920102b29.

Specifying "protocol" field to socket(AF_PACKET, ...) call only affects
the packets that are recevied.  So use protocol = 0 when sending raw
packets.

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): Fri Mar  4 12:58:50 CET 2016 on sn-devel-144

vfs: Fix the vfs_gpfs build

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): Fri Mar  4 02:09:15 CET 2016 on sn-devel-144

heimdal asn1: avoid fclose(NULL) caused by missing braces

Thanks to GCC6 -Wmisleading-indentation.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-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 Mar  3 16:21:52 CET 2016 on sn-devel-144

regtree: avoid GCC indentation warning

This was not actually a bug, but GCC6 (sort of reasonably) thought it could be.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

smbd:smb2: move op variable into scope of use in smb2_create_send

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

smbd:smb2: implement create replay

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

smbXsrv.idl: add create_action to smbXsrv_open

Needed for create replay.

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

smbXsrv:open: add smb2srv_open_lookup_replay_cache()

A function to find an open from the replay cache,
based on the create_guid handed in.

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

smb2:create: create replay cache when request has a create_guid

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

smbXsrv:open: maintain a replay cache

This caches a map create_guid -> file_id, so that
a replayed create can find the already created
open again.

This is automatically deleted once the first use
of the file handle is happening (triggered by
the lookup for the file-id).

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

librpc:smbXsrv.idl: add flags to smbXsrv_open

These flags reflect the need for and state of the replay cache.

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

smbd:smb2: allow the REPLAY_OPERATION flag for SMB3+ requests

Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>

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

torture:smb2: add smb2.replay.replay-dhv2-lease3

create with a lease, and replay with lease
with a different lease key.

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

torture:smb2: add smb2.replay.replay-oplock-lease

create with an oplock, and replay with a lease.

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

torture:smb2: add smb2.replay.replay-dhv2-lease-oplock

Open with a lease and replay with an oplock.

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

torture:smb2: add smb2.replay.replay-dhv2-lease2

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

torture:smb2: add smb2.replay.replay-dhv2-lease1

This is a variant of the replay-dhv2-oplock1 test for leases
instead of for oplocks.

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

torture:smb2:replay: extend CHECK_CREATE_OUT() to know leases

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

torture:smb2: split rename2 into multiple tests and extend these

- replay-regular
- replay-dhv2-oplock1
- replay-dhv2-oplock2
- replay-dhv2-oplock3

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

torture:smb2: rename replay1 -> replay-commands

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

vfs: remove an outdate comment from the files_struct definition

Leases are implemented since a while...

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Thu Mar  3 12:16:47 CET 2016 on sn-devel-144

VFS: Modify chmod_acl to take a const struct smb_filename * instead of const char *

Preparing to reduce use of lp_posix_pathnames().

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

VFS: Modify chmod to take a const struct smb_filename * instead of const char *

Preparing to reduce use of lp_posix_pathnames().

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3:smbd: add negprot remote arch detection for OSX

Remote arch detection for OSX clients has been broken for some time, since
both Samba and OSX started supporting SMB2.  Fix it by adding modern OSX
client detection support to the negprot remote arch detection routine.

Signed-off-by: Justin Maggard <jmaggard10@gmail.com>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <rb@sernet.de>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Thu Mar  3 09:03:53 CET 2016 on sn-devel-144

s3:smbd: rework negprot remote arch detection

Negprot remote arch detection is very cryptic.  Rework it so it's easier
to understand, and therefore more extensible, following the protocol table
in inline comments.  This also allows us to remove some hacks.

Signed-off-by: Justin Maggard <jmaggard10@gmail.com>
Reviewed-by: Ralph Boehme <rb@sernet.de>
Reviewed-by: Jeremy Allison <jra@samba.org>

selftest: test access based share enum parameter

BUG: https://bugzilla.samba.org/show_bug.cgi?id=8093

Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Uri Simchoni <uri@samba.org>
Autobuild-Date(master): Wed Mar  2 23:51:56 CET 2016 on sn-devel-144

access based share enum: handle permission set in configuration files

change function is_enumeration_allowed to check  permissions set by
fields: valid users, invalid users, only user.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=8093

Signed-off-by: Alberto Maria Fiaschi <alberto.fiaschi@estar.toscana.it>
Reviewed-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

torture:smb2: skip replay4 if server does not support multi-channel

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Wed Mar  2 20:38:49 CET 2016 on sn-devel-144

smbd: fix crash in smbXsrv_client_global_remove()

Probably copy-n-paste error.
Uncovered by the multi-channel-related tests we're
currently writing to exercise this code more.

Pair-Programmed-With: Guenther Deschner <gd@samba.org>

Signed-off-by: Michael Adam <obnox@samba.org>
Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

vfs:zfs fix build after get_nt_acl_fn signature change.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Uri Simchoni <uri@samba.org>
Autobuild-User(master): Uri Simchoni <uri@samba.org>
Autobuild-Date(master): Wed Mar  2 10:50:58 CET 2016 on sn-devel-144

vfs:glusterfs: fix build after opendir signature change

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Mar  2 04:34:15 CET 2016 on sn-devel-144

vfs:ceph: fix build after opendir signature change.

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

vfs/time_audit: use duration we get from the async backend

Finally! The previous commits changed the VFS and the async backend to
pass the duration of an aync operation up the stack.

We now can use this value instead of doing our own sampling which avoids
the following problem:

1. SMB2 read request received, added to the async queue

2. SMB2 create_file request comes in, is processed and blocks for N
   seconds in open()

3. async read completes in the dispatcher thread, completion callback
   will be called when we enter the main tevent loop

4. open() completes after N seconds

5. main tevent event loop is entered, async results are processed

6. async read result is processed, time sampling will include the N
   seconds blocked in open()

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

vfs/glusterfs: measure libglusterfs aio function call duration

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

vfs/aio_linux: measure libaio aio function call duration

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

vfs/aio_fork: measure async request syscall duration

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>