spoolss: cache_key handle allocation failures early

On cache_key allocation failure, set_printer_hnd_name() currently
stumbles through the rest of the function, and includes extra logic to
handle a NULL key later on.
Handling the allocation failure early makes sense, and also allows for
the NULL key checks to be removed.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Wed Apr 15 18:55:50 CEST 2015 on sn-devel-104

torture: Free the temporary memory context

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Apr 15 11:20:22 CEST 2015 on sn-devel-104

torture: Correctly invalidate the memory ccache.

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

autobuild: Add options to set mail host and send e-mail with logs

This helps when running the script in a cloud instance as
it is cheapest to shut it down once run

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Wed Apr 15 08:41:37 CEST 2015 on sn-devel-104

autobuild: Do not consider IDL.pm and Expr.pm changes to make a build bad

This allows a different yapp to be installed on the build host without failing
the whole autobuild

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>

Improve output of check-clean-tree.sh script

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>

.gitignore: Ignore pidl/MYMETA.json

This is generated by newer versions of MakeMaker

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>

autobuild: Do not wait when running just one target

This avoids having to remember to specify AUTOBUILD_RANDOM_SLEEP_OVERRIDE manually

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>

Fix incorrect order of arguments in error string

Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Böhme <rb@sernet.de>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Apr 15 05:57:25 CEST 2015 on sn-devel-104

Add DCERPC flag to call unbind hooks without destroying the connection itself upon termination of a connection with outstanding pending calls.

Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Apr 14 20:39:34 CEST 2015 on sn-devel-104

spoolss: purge the printer name cache on name change

Currently the name cache is only cleared on printer deletion. This means
that if a printer undergoes a name change, the old name remains in the
cache and can be subsequently used incorrecly if another printer takes
the same name as the old.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=11210

Reported-by: Franz Pförtsch <franz.pfoertsch@brose.com>
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): Tue Apr 14 05:37:50 CEST 2015 on sn-devel-104

s3: libsmbclient: After getting attribute server, ensure main srv pointer is still valid.

Bug #11186: Crash seen in libsmbclient due to free of server structure during SMBC_getxattr() call

Bug: https://bugzilla.samba.org/show_bug.cgi?id=11186

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Tue Apr 14 02:58:43 CEST 2015 on sn-devel-104

witness: add WITNESS_UNSPECIFIED_VERSION to IDL.

This version is seen in witness_GetInterfaceList replies from Windows Server
2012 R2.

Guenther

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Mon Apr 13 15:44:42 CEST 2015 on sn-devel-104

s4-torture: skip witness_AsyncNotify test for now.

We need to use real async calls in this test and add some clusapi commands so we
can remotely modify the cluster to actually trigger notifications.

Guenther

Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

s4-torture: add witness torture rpc testsuite.

Guenther

Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

srvsvc: add cluster specific share types to IDL.

Guenther

Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

s4-torture: add witness torture ndr testsuite.

Guenther

Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

librpc: further fixes for witness.idl.

Signed-off-by: Gregor Beck <gbeck@sernet.de>
Reviewed-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

s4:torture/winbind: add torture:winbindd_domain_without_prefix option

We should not assume that names in the domain
specified by 'torture:winbindd_netbios_domain' have no DOMAIN\ prefix.

On an AD DC we prefix all principals.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=11183

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Apr  9 19:35:38 CEST 2015 on sn-devel-104

s4:torture/local: add more torture_assert() checks

We need to make sure we return when torture_assert_passwd_equal()
or torture_assert_group_equal() fails.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=11183

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest/Samba4: use 'testallowed account' instead of 'test allowed'

local.nss test might print lines starting with 'test allowed:...'
and that confused the subunit parser.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest/knownfail: remove unused ^samba4.winbind.struct.show_sequence\(ad_dc\) line

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

rpcclient: Fix the timeout command

https://bugzilla.samba.org/show_bug.cgi?id=11199

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Apr  9 16:57:15 CEST 2015 on sn-devel-104

fss: Fix CID 1293354 Wrong operator used

Not used right now, but prevent future confusion.

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 Apr  8 19:36:36 CEST 2015 on sn-devel-104

waf: Fix systemd detection

https://bugzilla.samba.org/show_bug.cgi?id=11200

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Apr  8 15:06:26 CEST 2015 on sn-devel-104

ctdb-tests: Use ctdb_node_list_to_map() in tool stubs

Drop copy of old ctdb_control_nodemap().

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): Tue Apr  7 10:20:41 CEST 2015 on sn-devel-104

ctdb-common: Move ctdb_node_list_to_map() to utilities

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

ctdb-daemon: Factor out new function ctdb_node_list_to_map()

Change ctdb_control_getnodemap() to use this.

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

ctdb-tools: Drop the recovery from "reloadnodes"

A recovery is not required: when deleting a node it should already be
disconnected and when adding a node it will also be disconnected.  The
new sanity checks in "reloadnodes" ensure that these assumptions are
met.

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

ctdb-daemon: Don't delay reloading the nodes file

Presumably this was done to minimise the chance of a recovery
occurring while the nodemaps are inconsistent across nodes.

Another potential theory is that the forced recovery in the
ctdb.c:control_reload_nodes_file() stops another recovery occurring
for ReRecoveryTimeout seconds, so this delay causes the reloads to
occur during that period.

This is no longer necessary because recoveries are now explicitly
disabled while node files are reloaded.

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

ctdb-recoverd: Avoid nodemap-related checks when recoveries are disabled

The potential resulting recovery won't run anyway.  Also recoveries
may have been disabled by "reloadnodes" and if the nodemaps are
inconsistent between nodes then avoid triggering an unnecessary
recovery.

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

ctdb-tool: Update "reloadnodes" to disable recoveries

If a recovery occurs when some nodes have reloaded and others haven't
then the nodemaps with be inconsistent so bad things will happen.

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

ctdb-recoverd: New message ID CTDB_SRVID_DISABLE_RECOVERIES

Also add test stub support.

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

ctdb-recoverd: Simplify disable_ip_check_handler() using ctdb_op_disable()

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

ctdb-recoverd: Add slightly more abstraction for disabling takeover runs

Factor out new function srvid_disable_and_reply(), which can be
re-used.

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

ctdb-recoverd: Reimplement ReRecoveryTimeout using ctdb_op_disable()

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

ctdb-recoverd: Use a goto for do_recovery() failures

This will allow extra things to be done on failure.

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

ctdb-recoverd: Reimplement disabling takeover runs using ctdb_op_disable()

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

ctdb-recoverd: Add a new abstraction ctdb_op_disable()

This can be used to disable and re-enable an operation, and do all the
relevant sanity checking.

Most of this is from existing functions
disable_takeover_runs_handler(), clear_takeover_runs_disable() and
reenable_takeover_runs().

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

ctdb-daemon: Pass on consistent flag information to recovery daemon

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

ctdb-tests: Add "ctdb reloadnodes" test for "node remains deleted"

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

ctdb-tools: Fix spurious messages about deleted nodes being disconnected

The code was too "clever".  The 4 different cases should be separate.
The "node remains deleted" case doesn't need the IP address comparison
(always 0.0.0.0) or the disconnected check.

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

rpc_server: Coverity fix for CID 1273079

leaked_storage: Variable pk going out of scope leaks the storage it points to.

On failure get_pk_from_raw_keypair_params function should free up
the private key (pk) it allocates internally.

Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
Reviewed-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Ira Cooper <ira@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Thu Apr  2 19:38:22 CEST 2015 on sn-devel-104

build:wafsamba: fix a typo

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Ira Cooper <ira@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Thu Apr  2 16:39:01 CEST 2015 on sn-devel-104

vfs_ceph: add empty ACL callbacks

If a vfs module has no ACL callbacks, smbd will use its default ACL
callbacks. These default ACL callbacks operate on local filesytem,
it's clearly wrong for ceph case.

libcephfs does not support ACL yet, so this patch adds ACL callbacks
that do nothing.

Signed-off-by: Yan, Zheng <zyan@redhat.com>
Reviewed-by: Ira Cooper <ira@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

vfs_ceph: use 'file descriptor' version xattr functions when possible

libcephfs version 0.94 adds 'file descriptor' version xattr functions.
This patch makes corresponding samba VFS callbacks use these new
functions.

Signed-off-by: Yan, Zheng <zyan@redhat.com>
Reviewed-by: Ira Cooper <ira@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb: check for talloc_asprintf() failure

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): Wed Apr  1 15:36:03 CEST 2015 on sn-devel-104

ctdb: Coverity fix for CID 1291643

CID 1291643: Resource leak: leaked_handle: Handle
variable lock_fd going out of scope leaks the handle.

Fix: on failure case release handle variable lock_fd

Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
Reviewed-by: Michael Adam <obnox@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

s3-passdb: Fix 'force user' with winbind default domain

If we set 'winbind use default domain' and specify 'force user = user'
without a domain name we fail to log in. In this case we need to try a
lookup with the domain name.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11185

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Mar 31 21:17:23 CEST 2015 on sn-devel-104

selftest: run the FSRVP test suite against s3fs

With FSRVP server support now present along with suitable mock-up test
infrastructure, run the FSRVP test suite against s3fs.

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

selftest: add snapshot share configuration

Define a share that uses both vfs_shell_snap and fake_snap.pl to create,
delete and expose fake snapshots in response to FSRVP requests.
Additionally test snapshot enumeration and access via the shadow_copy2
module.

Allow for simple testing of FSRVP message sequence timeouts, by
specifying an artificially small interval.

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

doc: add vfs_shell_snap manpage

Document usage of the shell_snap VFS module, explaining when and how
each of the shell script commands are called.

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

vfs: add vfs_shell_snap module

The shell_snap VFS module plumbs into the snapshot (aka shadow-copy)
management paths used by Samba's File Server Remote VSS Protocol (FSRVP)
server.
The following shell callouts may be configured in smb.conf:

shell_snap: check path command
- Called when an FSRVP client wishes to check whether a given
  share supports snapshot create/delete requests.
- The command is called with a single <share path> argument.
- The command must return 0 if <share path> is capable of being
  snapshotted.

shell_snap: create command
- Called when an FSRVP client wishes to create a snapshot.
- The command is called with a single <share path> argument.
- The command must return 0 status if the snapshot was
  successfully taken.
- The command must output the path of the newly created snapshot
  to stdout.

shell_snap: delete command
- Called when an FSRVP client wishes to delete a snapshot.
- The command is called with <base share path> and
  <snapshot share path> arguments.
- The command must return 0 status if the snapshot was
  successfully removed.

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

doc: explain vfs_btrfs remote snapshot configuration

This extends the vfs_btrfs man page to also cover FSRVP remote snapshot
behaviour and configuration.

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

doc: explain vfs_snapper remote snapshot configuration

This extends the vfs_snapper man page to also cover FSRVP remote
snapshot behaviour and configuration.
The permissions section is also extended to describe specific Samba and
Snapper requirements for remote snapshot creation and deletion using
DiskShadow.exe.

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

doc: "prune stale" and "sequence timeout" fssd parameters

This change adds smb.conf documentation for the "fss: prune stale" and
"fss: sequence timeout" parameters accepted by Samba's FSRVP server.

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

fsrvp: prune shadow copies if associated path doesn't exist

This patch implements some simple FSRVP server housekeeping. On startup
the server scans the cached entries, any entries where the underlying
system paths associated with shadow copies no longer exist are removed
from the cache and from the registry.

This behaviour is disabled by default, but can be enabled via the new
"fss: prune stale" smb.conf parameter.

Signed-off-by: Noel Power <noel.power@suse.com>
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

fsrvp: add remote snapshot RPC server

The Samba fss_agent RPC server is an implementation of the File Server
Remote VSS (Volume Shadow Copy Service) Protocol, or FSRVP for short.

FSRVP is new with Windows Server 2012, and allows authenticated clients
to remotely request the creation, exposure and deletion of share
snapshots.

The fss_agent RPC server processes requests on the FssAgentRpc named
pipe, and dispatches relevant snapshot creation and deletion requests
through to the VFS.
The registry smb.conf back-end is used to expose snapshot shares, with
configuration parameters and share ACLs cloned from the base share.

There are three FSRVP client implementations that I'm aware of:
- Samba rpcclient includes fss_X commands.
- Windows Server 2012 includes diskshadow.exe.
- System Center 2012.

FSRVP operations are only processed for users with:
- Built-in Administrators group membership, or
- Built-in Backup Operators group membership, or
- Backup Operator privileges, or
- Security token matches the initial process UID

MS-FSRVP specifies that server state should be stored persistently
during operation and retrieved on startup. Use the existing fss_srv.tdb
FSRVP state storage back-end to satisfy this requirement.

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

vfs_snapper: create/delete snapshot support

Extend vfs_snapper to support the new remote snapshot creation and
deletion hooks added for FSRVP.

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

vfs_btrfs: add snapshot create/delete calls

The "btrfs: manipulate snapshots" smb.conf parameter is disabled by
default, to encourage use of, and pass requests through to, the
vfs_snapper module.
When enabled, issue BTRFS_IOC_SNAP_CREATE_V2 and BTRFS_IOC_SNAP_DESTROY
ioctls accordingly. The ioctls are issued as root, so rely on permission
checks in the calling FSRVP server process.

Base share paths must exist as btrfs subvolumes in order to
be supported for snapshot operations.

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

replace: check for dirname() and basename()

These functions are provided by libgen.h, and conform to POSIX.1-2001.

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

vfs: add snapshot create/delete hooks

This change adds three new VFS hooks covering snapshot manipulation:
- snap_check_path
  Check whether a path supports snapshots.
- snap_create
  Request the creation of a snapshot of the provided path.
- snap_delete
  Request the deletion of a snapshot.

These VFS call-outs will be used in future by Samba's File Server Remote
VSS Protocol (FSRVP) server.

MS-FSVRP states:
  At any given time, Windows servers allow only one shadow copy set to
  be going through the creation process.
Therefore, only provide synchronous hooks for now, which can be
converted to asynchronous _send/_recv functions when the corresponding
DCE/RPC server infrastructure is in place.

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

torture: add local FSRVP server state tests

Test the storage and retrieval of FSRVP server state, with varying
shadow-copy set, shadow copy and share map hierarchies.

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

fsrvp: add server state storage back-end

MS-FSRVP specifies:
  the server MUST persist all state information into an implementation-
  specific configuration store.

This change adds a fss_srv TDB database to preserve FSRVP server state,
with the following keys used to track shadow copy state and hierarchy:
- sc_set/<shadow copy set GUID>
  A shadow copy set tracks a collection of zero or more shadow copies,
  as initiated by a StartShadowCopySet FSRVP client request.
- sc_set/<shadow copy set GUID>/sc/<shadow copy GUID>
  A shadow copy defines information about a snapshot base volume, the
  snapshot path, and a collection of share maps. It is initiated by an
  AddToShadowCopySet client request.
- sc_set/<shadow copy set GUID>/sc/<shadow copy GUID>/smap/<smap GUID>
  A share map tracks new shares that are created to expose shadow
  copies.

All structures are marshalled into on-disk format using the previously
added fsrvp_state IDL library.

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

librpc: add FSRVP server state idl

FSRVP server state must be retained persistently. This change adds IDL
definitions for the share map, shadow-copy and shadow-copy set types,
which will be used for marshalling and unmarshalling state alongside
database storage or retrieval.

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

waf: Remove 'linkflags.remove(x)' line added in error.

Fixes bug #11165 - Bug in configure scripts when system-mitkrb5 is used

https://bugzilla.samba.org/show_bug.cgi?id=11165

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Tue Mar 31 04:32:52 CEST 2015 on sn-devel-104

tdb: Do not build test binaries if it's not a standalone build

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Mar 31 01:56:02 CEST 2015 on sn-devel-104

s4-torture: add test to verify nbt_name with "." ending handling.

Windows uses a username of 'domain.example.com.' and we need to return it that
way in the NETLOGON_SAM_LOGON_RESPONSE_EX.

See e6e2ec0001fe3c010445e26cc0efddbc1f73416b for further details.

Guenther

Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Mon Mar 30 16:18:04 CEST 2015 on sn-devel-104

s4-torture: use torture_comment instead of printf in raw notify test.

Guenther

Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-torture: use tctx variable name in raw notify test consistently.

Guenther

Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4:torture:raw:notify: torture_assert on creation of secondary tcon

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

s4:torture:raw:notify: use torture_assert instead of printf in test_notify_tree

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

s4:torture:raw:notify: let NOTIFY_MASK_TEST use torture_assert macros

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

s4:torture:raw:notify: remove extra do-loop in NOTIFY_MASK_TEST macro.

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

s4:torture:raw:notify: use torture_assert instead of printf in failure case

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

s4:torture:raw:notify: remove superfluous conditional goto

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

s4:torture:raw:notify: treat torture_open_connection calls with torture_assert

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

s4:torture:raw:notify: use torture_assert with torture_setup_dir

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

s4:torture:raw:notify: add a few comments to torture_assert calls

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

s4:torture:raw:notify: improve the CHECK_WSTR() macro

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

s4:torture:raw:notify: make check_rename_reply() properly use torture_result

Only change currently: the CHECK_WSTR calls report the line
number of this function now instead of the handed in
line of the callers. This could be fixed by turning this
function into a macro...

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

s4:torture:raw:notify: remove CHECK_WSTR2.

The original CHECK_WSTR() macro was not setting torture failure,
leading to errors instead of propoer failures.

The original CHECK_WSTR2() macro was exactly like the CHECK_WSTR
macro but using propoer torture_result() calls.

This patch removes the original CHECK_WSTR(), renames CHECK_WSTR2
to CHECK_WSTR and adapts the callers, hence removing the source
of many potential missing torture_assert messages.

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

s4:torture:raw:notify: remove CHECK_VAL.

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>

s4:torture:raw:notify: remove CHECK_STATUS.

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>

torture: add torture_assert_not_null[_goto]

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

torture: add torture_assert_int_not_equal_goto

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

s3:trusts_util: generate completely random passwords in trust_pw_change()

Instead of having every 2nd byte as '\0' in the utf16 password,
because the utf8 form is based on an ascii subset, we convert
the random buffer from CH_UTF16MUNGED to CH_UTF8.

This way we have a random but valid utf8 string,
which is almost like what Windows is doing.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s3:trusts_util: pass new_trust_version to netlogon_creds_cli_ServerPasswordSet() in trust_pw_change()

We should maintain current and previous passwords on both sides of the trust,
which mean we need to pass our view of the new version to the remote DC.

This avoid problems with replication delays and make sure the kvno
for cross-realm tickets is in sync.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s3:trusts_util: make use of pdb_get_trust_credentials() and pdb_get_trusted_domain() in trust_pw_change()

Using pdb_get_trust_credentials() works for all kind of trusts
and gives us much more details regarding the credentials.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s3:trusts_util: add support for SEC_CHAN_DNS_DOMAIN in trust_pw_change()

SEC_CHAN_DNS_DOMAIN trusts use longer passwords, Windows uses 240 UTF16 bytes.

Some trustAttribute flags may also make impact on the length on Windows,
but we could be better if we know that the remote domain is an AD domain.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s3:rpc_server/lsa: we need to normalize the trustAuth* blobs before storing them

The number of current and previous elements need to match and we have to
fill TRUST_AUTH_TYPE_NONE if needed.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s4:rpc_server/lsa: we need to normalize the trustAuth* blobs before storing them

The number of current and previous elements need to match and we have to
fill TRUST_AUTH_TYPE_NONE if needed.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s4:rpc_server/lsa: notify winbindd about new trusted domains

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s3:winbindd: add MSG_WINBIND_NEW_TRUSTED_DOMAIN that takes a lsa_TrustDomainInfoInfoEx

When a new trusted domain is added in the LSA server, we need to immediately
have the domain within winbindd. This notification is done via a
MSG_WINBIND_NEW_TRUSTED_DOMAIN message.

In future we might want just a "rescan direct trusts" message,
but that requires a lot of redesign within winbindd.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

lsa.idl: mark lsa_TrustDomainInfoInfoEx as public

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s4:selftest: run dbcheck against the ad_dc environment too

This is the environment that is configured like real world configurations.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s4:rpc_server/lsa: implement the policy security descriptor

We now check the requested access mask in OpenPolicy*()
and return NT_STATUS_ACCESS_DENIED if the request is not granted.

E.g. validating a domain trust via the Windows gui requires this
in order prompt the user for the credentials. Otherwise
we fail any other call with ACCESS_DENIED later and the
gui just displays a strange error message.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

s4:rpc_server/lsa: normalize the access_mask for lsa account objects

We still grant all access in the access_mask, but we don't check the
mask at all yet...

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

libcli/security: add security_descriptor_for_client() helper function

This prepares a possibly stripped security descriptor for a client.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

libcli/security: support "IS" in SDDL for SID_NT_IUSR

TODO: we should import the whole lists from [MS-DTYP].

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>