libsocket: Add "mem_ctx" to socket_create()

Every caller did a talloc_steal() after socket_create(). Just pass in the
correct memory context.

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

replace: Only warn if rpc/rpc.h haven't been found

This is also used by talloc, tevent, etc. Those libs don't need or use
rpc.h

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Tue Feb 27 03:23:37 CET 2018 on sn-devel-144

winbindd: Remove "DUMP_EVENT_LIST" message

This was no longer implemented, remove it completely

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

vfs_glusterfs: Fix the wrong pointer being sent in glfs_fsync_async

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13297

Pair-Programmed-With: Anoop C S <anoopcs@redhat.com>
Signed-off-by: Poornima G <pgurusid@redhat.com>
Reviewed-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Mon Feb 26 20:17:50 CET 2018 on sn-devel-144

ldb_debug tests: Fix binary data in debug log

Tests to ensure:
    When duplicate objects are added, the GUID was printed in the debug log
    are passed through the escape function.
    And that duplicate DN's do not generate debug log entries.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13185

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Mon Feb 26 07:29:49 CET 2018 on sn-devel-144

ldb_debug: Fix binary data in debug log

When duplicate objects were added, the GUID was printed in the debug log
The GUID was not escaped and therefore displayed as binary content.

This patch splits out the duplicate DN creation error and the duplicate
GIUD error.  Duplicate DN's are a normal event and don't require debug
logging.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13185

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

Minor cleanup to libnet_join_member

Prevent code duplication by consolidating cleanup task
at the end of the function.

Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Reviewed-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Feb 24 23:19:05 CET 2018 on sn-devel-144

tests/blackbox/smbcontrol: ping tests names ping target upon failure

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jeremy Allison <jra@samba.org>

text/python/blackbox: allow optional message in ProcessError

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jeremy Allison <jra@samba.org>

ldb tests: fix null test on incorrect variable

Fix up tests that were  performing a null check on the wrong variable
after a call to ldb_msg_new

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Sat Feb 24 15:50:35 CET 2018 on sn-devel-144

s4:lib:com: Fix function declartions

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:selftest: Run the smbtorture3 TLDAP test

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

repl_md: avoid returning LDB_SUCCESS on failure

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

tests: Authenticate again so make sure we are not locked out

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

selftest: Fail testsuite on unknown environment

The status quo was that tests in typo'd/unknown environments would be skipped.
This was a side effect of the mechanism by which samba3 tests were run, which
has now been changed.

For an example of this being a problem in the past, see commit 523bd03f (which
fixes an issue I noticed when reading subunit logs, and which did not fail the
test).

Signed-off-by: Jamie McClymont <jamiemcclymont@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>

s3: smbd: Fix possible directory fd leak if the underlying OS doesn't support fdopendir()

HPUX has this problem.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13270

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): Fri Feb 23 22:56:35 CET 2018 on sn-devel-144

winbind: don't try to do an authenticated SMB connection as AD DC

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13278

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Feb 23 17:58:23 CET 2018 on sn-devel-144

winbind: set_dc_type_and_flags() is not needed on a DC

On a DC we load the trusts in the parent in add_trusted_domains_dc()
from our local configuration. There's no need to find out the trust details
via network calls.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13278

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

winbind: make sure we don't contact trusted domains via LDAP as AD DC

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13278

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

winbind: make sure we don't contact trusted domains via SAMR as AD DC

This is not needed for the normal operation of an AD DC.

Administrators should just use other tools instead of
wbinfo to list and query users and groups.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13278

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

winbind: let cm_connect_netlogon_transport() only work against direct trust as AD DC

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13278

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

winbind: force the usage of schannel in cm_connect_lsa() as AD DC

This makes sure we only talk to direct trusts.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13278

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3:smb_macros.h: add IS_AD_DC as addition to IS_DC

In the long run we should remove this again (as well as IS_DC).

But for now this makes some code changes in winbindd easier to
follow.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13278

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

dsdb/encrypted_secrets: remove dependency to libnettle and use our own aes_gcm_128_*()

We already rely on gnutls in order to implement SSL/TLS, so using that
to speed up crypto like aes gcm 128 is fine, but as we already have
code for that algorithm, we should use that instead of adding a new
dependency to libnettle.

Some (I guess newer versions) of gnutls use nettle internally, so
we may end up using that code, but we should not have a direct dependency.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13276

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

vfs_fruit: use off_t, not size_t for TM size calculations

size_t is only a 32-bit integer on 32-bit platforms. We must use off_t
for file sizes.

https://bugzilla.samba.org/show_bug.cgi?id=13296

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

winbind: Use one queue for all domain children

If we have multiple domain children, it's important
that the first idle child takes over the next waiting request.

Before we had the problem that a request could get stuck in the
queue of a busy child, while later requests could get served fine by
other children.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13292

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Feb 23 09:04:23 CET 2018 on sn-devel-144

winbind: Maintain a binding handle per domain and always go via wb_domain_request_send()

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

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13292

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

winbind: make choose_domain_child() static

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13292

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: add locator_child_handle() and use it instead of child->binding_handle

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13292

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: add idmap_child_handle() and use it instead of child->binding_handle

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13292

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: improve wb_domain_request_send() to use wb_dsgetdcname_send() for a foreign domain

Commit ed3bc614cccec6167c64ac58d78344b6426cd019 got the logic wrong while
trying to implement the logic we had in init_child_connection(),
which was removed by commit d61f3626b79e0523beadff355453145aa7b0195c.

Instead of doing a WINBINDD_GETDCNAME request (which would caused an error
because the implementation was removed in commit
958fdaf5c3ba17969a5110e6b2b08babb9096d7e), we sent the callers request
and interpreted the result as WINBINDD_GETDCNAME response, which
led to an empty dcname variable. As result the domain child
opened a connection to the primary domain in order to lookup
a dc.

If we want to connect the primary domain from the parent via
a domain child of the primary domain.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13295

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: use state->{ev,request} in wb_domain_request_send()

This will reduce the diff for the following changes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13295

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: avoid using fstrcpy(dcname,...) in _dual_init_connection

domain->dcname was converted from fstring to char * by commit
14bae61ba36814ea5eca7c51cf1cc039e9e6803f.

Luckily this was only ever called with an empty string in
state->request->data.init_conn.dcname.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13294

Signed-off-by: Stefan Metzmacher <metze@samba.org>

winbind: cleanup winbindd_cli_state->pwent_state if winbindd_getpwent_recv() returns an error

A client may skip the explicit endpwent() if getgrent() fails.

This allows client_is_idle() return true in more cases.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13293

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: cleanup winbindd_cli_state->grent_state if winbindd_getgrent_recv() returns an error

A client may skip the explicit endgrent() if getgrent() fails.

This allows client_is_idle() return true in more cases.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13293

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: call lp_winbind_enum_{users,groups}() already in set{pw,gr}ent()

This way we don't keep winbindd_cli_state->{pw,gr}ent_state arround forever,
if the client forgets an explicit end{pw,gr}ent().

This allows client_is_idle() return true in more cases.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13293

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: protect a pending wb_child_request against a talloc_free()

If the (winbind) client gave up we call TALLOC_FREE(state->mem_ctx)
in remove_client(). This triggers a recursive talloc_free() for all
in flight requests.

In order to maintain the winbindd parent-child protocol, we need
to keep the orphaned wb_simple_trans request until the parent
got the response from the child.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13290

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

winbind: use tevent_queue_wait_send/recv in wb_child_request_*()

We need a way to keep the child->queue blocked without relying on
the current 'req' (wb_child_request_state).

The next commit will make use of this.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13290

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tevent: version 0.9.36

* improve documentation of tevent_queue_add_optimize_empty()
* add tevent_queue_entry_untrigger()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

tevent: add tevent_queue_entry_untrigger()

Pair-Programmed-With: Volker Lendecke <vl@samba.org>

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

tevent: improve documentation of tevent_queue_add_optimize_empty()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

s4:auth_sam: allow logons with an empty domain name

It turns out that an empty domain name maps to the local SAM.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13206

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Fri Feb 23 04:08:26 CET 2018 on sn-devel-144

tests/bind.py: Add a bind test with NTLMSSP with no domain

Confirmed to pass against Windows 2012 R2.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13206

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

tests/py_creds: Add a SamLogonEx test with an empty string domain

This test passes against 4.6, but failed against 4.7.5 and master.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13206

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

s3:cliconnect.c: remove useless ';'

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13206

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:libsmb: allow -U"\\administrator" to work

cli_credentials_get_principal() returns NULL in that case.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13206

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

build: fix libceph-common detection

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13277

Guenther

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Thu Feb 22 19:30:12 CET 2018 on sn-devel-144

util/rfc1738: update preamble

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Douglas Bagnall <dbagnall@samba.org>
Autobuild-Date(master): Thu Feb 22 06:13:49 CET 2018 on sn-devel-144

util/rfc1738: simplify and fix rfc1738_escape_part()

We now encode according to RFC 3986 (section 2.1 - 2.3).

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

util/rfc1738: remove unused versions of rfc1738_escape

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

util/rfc1738: simplify and fix rfc1738_unescape()

Improvements:

* NULL is returned when the string is incorrectly formed.

* Badly formed escapes like "% b" that were accepted by sscanf() are now
  rejected.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

util/tests: add rfc1738 cmocka tests

These don't pass yet.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

util/rfc1738_unescape(): return end pointer or NULL on error

At present we don't detect errors, but when we do we'll return NULL.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

shift read_hex_bytes() and parse_guid_string() into lib/util

read_hex_bytes() is going to be used in lib/util/rfc1738.c.

parse_guid_string() is shifted for two reasons: Firstly, it is called
very often in some operations, sometimes constituting a few percent of
the CPU load, and it makes several calls to read_hex_bytes(). We want
the compiler to be able to inline those calls if it thinks that is
wise. Secondly, there are other places that could do with fast GUID
parsing.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

selftest/tests.py: remove always-needed, never-set with_cmocka flag

We have cmocka in third_party, so we are never without it.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

unittests.lib_util_modules: test module probe with "skel", not "unix"

The unix module is not available as a module on some systems.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

selftest: subunithelper needs to follow the subunit spec more closely

In particular allow ]\n without \n]\n as used by cmocka

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4:rpc_server: fix call_id truncation in dcesrv_find_fragmented_call()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13289

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Wed Feb 21 19:02:56 CET 2018 on sn-devel-144

tests:dcerpc/raw_protocol: reproduce call_id truncation bug

We need to make sure the server handles call_id values > UINT16_MAX.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13289

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

heimdal_build: use closefrom from libreplace

this silences a lot of "... has been redefined" compiler warnings on
platforms that don't have closefrom

Signed-off-by: Bjoern Jacke <bjacke@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

s4:rpc_server/lsa: implement forwarding lsa_Lookup{Sids,Names}() requests to winbindd

This might not be perfect yet, but it's enough to allow names from trusted
forests/domain to be resolved, which is very important for samba based
domain members.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

winbindd: implement wb_irpc_lsa_{LookupNames4,LookupSids3}()

This will be used by the LSA Server on an AD DC to request remote views
from trusts.

In future we should implement wb_lookupnames_send/recv similar to
wb_lookupsids_send/recv, but for now using wb_lookupname_send/recv in a loop
works as a first step.

We also need to make use of req->in.level and req->in.client_revision
once we want to support more than one domain within our own forest.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: rewrite lookup sids/names code to honor the given lookup level

[MS-LSAT] 2.2.16 LSAP_LOOKUP_LEVEL defines the which views each level should
consult.

Up to now we support some wellknown sids, the builtin domain and our
account domain, but all levels query all views.

This commit implements 3 views (predefined, builtin, account domain)
+ a dummy winbind view (which will later be used to implement the
gc, forest and trust views)..

Depending on the level we select the required views.

This might not be perfect in all details, but it's enough
to pass all existing tests, which already revealed bugs
during the development of this patch.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

test_trust_ntlm.sh: add lookup name tests

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

libcli/security: add dom_sid_lookup_predefined_{sid,name}()

This basically implements [MS-LSAT] 3.1.1.1.1 Predefined Translation Database
and Corresponding View.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:dsdb: add dsdb_trust_domain_by_{sid,name}()

This gets the lsa_ForestTrustDomainInfo for the searched
domain as well as the lsa_TrustDomainInfoInfoEx for the
direct trust (which might be the same for external trust or
the forest root domain).

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: prepare dcesrv_lsa_LookupNames* for async processing

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: prepare dcesrv_lsa_LookupSids* for async processing

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: base dcesrv_lsa_LookupNames2() on dcesrv_lsa_LookupNames_common()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: base dcesrv_lsa_LookupNames() on dcesrv_lsa_LookupNames_common()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: rename 'state' variable to 'policy_state' in dcesrv_lsa_LookupNames2()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: rename 'state' variable to 'policy_state' in dcesrv_lsa_LookupSids2()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: rename 'state' variable to 'policy_state' in dcesrv_lsa_LookupSids_common()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: simplify [ref] pointer handling in dcesrv_lsa_LookupNames()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: simplify [ref] pointer handling in dcesrv_lsa_LookupSids()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: remove unused 'status' variable in dcesrv_lsa_LookupSids_common()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: make sure dcesrv_lsa_LookupNames2() gets prepared [ref] pointers

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: expect prepared [ref] pointers in dcesrv_lsa_LookupNames_common()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: make sure dcesrv_lsa_LookupSids_common() gets prepared [ref] pointers

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:rpc_server/lsa: use LSA_LOOKUP_OPTION_SEARCH_ISOLATED_NAMES/LSA_CLIENT_REVISION_1 in compat code

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

rpcclient: add lookupsids_level command

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

rpcclient: fix variable initialisation and add parenthesis to if clauses

Just a few README.Coding fixes.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13286

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

provision: fix the 'dnsdomain' for the local sam of a domain member

A member has a local AD database, which should not use the 'dnsdomain'
as the one on domain controllers.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13285

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

traffic_packets.py: let Lookup{Sids,Names}() work against a sane server

In order to resolve predefined sids or names we need to use
level = LSA_LOOKUP_NAMES_ALL (1).

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13284

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

nsswitch: fix double free errors in nsstest.c

We need to zero out static pointers on free.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13283

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4:torture: zero initialize variables in test_LookupSidsReply()

This avoids crashes if the server returns unexpected results.  The test
should just report the failure in that case.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13282

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

winbindd: make use of talloc_zero_array() in wb_lookupsids*()

It just feels better for such a complex function.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13281

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3:cli_lsarpc: use talloc_zero_array() in dcerpc_lsa_lookup_names_generic()

It just feels better for such a complex function.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13281

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

s3:cli_lsarpc: use talloc_zero_array() in dcerpc_lsa_lookup_sids_generic()

It just feels better for such a complex function.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13281

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

winbindd: initialize type = SID_NAME_UNKNOWN in wb_lookupsids_single_done()

We check for !NT_STATUS_LOOKUP_ERR(), but wb_lookupsid_recv()
only initializes the results together with NT_STATUS_OK.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13280

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

winbindd: don't split the rid for SID_NAME_DOMAIN sids in wb_lookupsids

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13279

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

replace: remove some duplicate checks

Signed-off-by: Bjoern Jacke <bjacke@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

wscript: drop checks for setnetgrent/endnetgrent/getnetgrent

we don't use setnetgrent/endnetgrent/getnetgrent since security share passed
away.

Signed-off-by: Bjoern Jacke <bjacke@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

s3: remove dead already commented code

Signed-off-by: Bjoern Jacke <bjacke@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

selftest: change technique for running specific envs

Currently testsuites excluded with --exclude/--include-env are skipped when
encountered in the middle of a run, so they are included in progress reporting,
and the @todo list does not accurately show what will be done.

This change skips them earlier, preventing them from being added to @todo, as is
done with pattern-based including/excluding.

As well as making the progress indicator more accurate, this means that
selftest.pl can use @todo to determine when we are finished with an environment.

Signed-off-by: Jamie McClymont <jamiemcclymont@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Wed Feb 21 07:47:58 CET 2018 on sn-devel-144

selftest: move to declaratively specifying environments and their dependencies

This removes the tangle of code for starting up dependencies, and allows
selftest.pl to query dependencies (hence it can know when things can be shut
down early and how to order environments for optimal memory usage - that patch
not yet submitted).

It also removes the slightly hacky special-casing of the ad_members, and sets
$target->{vars} centrally (so each setup_ function does not need to).

Signed-off-by: Jamie McClymont <jamiemcclymont@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

selftest: remove prototypes from some subroutine templates

Rationale, as sent to samba-technical:

  > Why do you remove the explicit argument declarations for so many functions?
  > These declarations help catch usage errors, i.e. wrong number of calling
  > arguments, especially when you add arguments to functions.

  My understanding is that because these are object methods, the prototypes
  actually were not providing the checking benefits [see PERLSUB(1) (section
  "Prototypes")] -- this is evidenced by the fact that some of the prototypes I
  removed actually had the wrong number of arguments!

  Now that the subroutines are being dynamically called by function references,
  the checking also appears not to apply [see the same source].

  There was also a more concrete reason: the setup code will automatically set
  up multiple environment dependencies and pass each of their vars as an
  individual parameter. Accomplishing this was [seemingly] impossible with the
  prototypes in place.

  Additionally, there seems to be a consensus among perl devs that prototypes
  are generally harmful: see this post (by my colleague) for example:
  http://www.perlmonks.org/?node_id=406231

Signed-off-by: Jamie McClymont <jamiemcclymont@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:auth: Pass mem_ctx to init_system_session_info()

We have a stackframe we can use for the lifetime of the session.

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): Wed Feb 21 02:46:40 CET 2018 on sn-devel-144

s3:auth: Pass mem_ctx to init_guest_session_info()

Use a mem_ctx which gets freed if possible.

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

s3:auth: Pass a mem_ctx to make_new_session_info_guest()

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