replmd: replmd_process_link_attribute() returns type of change made

In order to share work across related link attribute updates, we need
replmd_process_link_attribute() to let the caller know what actually
changed.

This patch adds an extra return type that'll be used in the next patch.
What we're interested in is: the update was ignored (i.e. it's old news),
a new link attribute was added (because this affects the overall
msg/element memory), and an existing link attribute was modified (due to
how links are actually stored, this includes deleting the link, as in
reality it simply involves setting the existing link to 'inactive').

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

source4 samr: Tidy DBG_WARNING calls

Move the calls to GUID_buf_string and dom_sid_str_buf into the
coresponding DBG_WARNING call, instead of using an intermediate variable.
While this violates the coding guidelines, doing this makes the code less
cluttred and means the functions are only called if the debug message is
printed.

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): Wed Nov 21 01:50:11 CET 2018 on sn-devel-144

s4-samr: Use GUID_buf_string() in dcesrv_samr_EnumDomainUsers()

This avoids memory allocation.

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

s4-samr: Use dom_sid_split_rid() to get the RID in dcesrv_samr_EnumDomainUsers

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

source4 samr: cache samr_EnumDomainUsers results

Add a cache of GUID's that matched the last samr_EnunDomainUsers made on a
domain handle.  The cache is cleared if resume_handle is zero, and when the
final results are returned to the caller.

The existing code repeated the database query for each chunk requested.

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

tests samr: Extra tests for samr_EnumDomainUserss

Add extra tests to test the content returned by samr_EnumDomainUsers,
and tests for the result caching added in the following commit.

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

s4-samr: Use GUID_buf_string() in dcesrv_samr_EnumDomainGroups()

This avoids memory allocation

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

s4-samr: Use dom_sid_split_rid() to get the RID in dcesrv_samr_EnumDomainGroups

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

source4 samr: cache samr_EnumDomainGroups results

Add a cache of GUID's that matched the last samr_EnunDomainGroups made on a
domain handle.  The cache is cleared if resume_handle is zero, and when the
final results are returned to the caller.

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

test samr: Extra tests for samr_EnumDomainGroups

Add extra tests to test the content returned by samr_EnumDomainGroups,
and tests for the result caching added in the following commit.

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

s4-samr: Use dom_sid_split_rid() to get the RID in dcesrv_samr_QueryDisplayInfo

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

source4 samr: cache samr_QueryDisplayInfo results

Add a cache of GUID's that matched the last samr_QueryDisplayInfo made on a
domain handle.  The cache is cleared if the requested start index is
zero, or if the level does not match that in the cache.

The cache is maintained in the guid_caches array of the dcesrv_handle.

Note: that currently this cache exists for the lifetime of the RPC
      handle.

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

tests samr: remove PEP8 warnings

Remove PEP8 warnings from the samr tests.

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

tests samr: Extra tests for samr_QueryDisplayInfo

Add extra tests to test the content returned by samr_QueryDisplayInfo,
which is not tested for the ADDC.  Also adds tests for the result
caching added in the following commit.

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

ldb_controls: Add some talloc error checking for controls

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13686

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

sync_passwords: Remove dirsync cookie logging for continuous operation

Under normal operation, users shouldn't see giant cookies in their logs.
We still log the initial cookie retrieved from the cache database, which
should still be helpful for identifying corrupt cookies.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13686

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

dirsync: Allow arbitrary length cookies

The length of the cookie is proportional to the number of DCs ever in
the domain (as it stores the uptodateness vector which has stale
invocationID).

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13686

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

replmd: Cache recycle-bin state to avoid DB lookup

By caching the recycle-bin state we can save ~6% of the join time.

Checking whether the recycle-bin is enabled involves an underlying DSDB
search. We do this ~4 times for each link we replicate (twice for the
link source and target). By caching the recycle-bin's state over the
duration of the replication, we can save 1000s of unnecessary DB
searches.

With 5K users this makes the join time ~5 secs faster.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Tim Beale <timbeale@samba.org>
Autobuild-Date(master): Tue Nov 20 08:40:16 CET 2018 on sn-devel-144

replmd: Split some code out into create_la_entry() helper function

replmd_store_linked_attributes() has gotten in szie and complexity. This
refactors some code out into a separate function to make things a bit
more manageable.

This patch should not alter functionality.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

replmd: Minor change to replmd_verify_link_target() args

We were passing in the entire src_msg, but all we really need is the
source object's DN (and even then, it's only used in error messages).

Change it so we only pass in what the function actually needs. This
makes it a bit easier to see what src_msg is actually used for.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

replmd: Skip redundant source object link checks

We receive the links grouped together by source object. We can save
ourselves some work by not looking up the source object for every single
link (if it's still the same object we're dealing with).

We've already made this change to replmd_process_linked_attribute().
This patch makes the same change to replmd_store_linked_attributes().
(We verify that we know about each link source/target as we receive each
replication chunk. replmd_process_linked_attribute() kicks in later as
the transaction completes).

Note some care is needed to hold onto the tmp_ctx/src_msg across
multiple passes of the for loop.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

replmd: Split up replmd_verify_linked_attribute() into src/target checks

Refactor replmd_verify_linked_attribute() so we split out the link
attribute source/target checks. This patch should not alter
functionality.

The source object check has been moved out to where
replmd_verify_linked_attribute() was called.

replmd_verify_linked_attribute() has been renamed, as it's now only
checking the link target.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

waf: Load the C compiler correctly

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Tue Nov 20 04:47:24 CET 2018 on sn-devel-144

librpc/tables.pl: remove unused $opt_output option

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

wafsamba: simplify SAMBA_PIDL_TABLES() rule

The builddir is not bin/default/ instead of just bin/,
so we don't need to strip 'default/' anymore.

And the '--output ${TGT}' part is not really implemented.

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

wafsamba: remove unused Build.BuildContext.pre_build overload

This is not needed and also fixed the interaction between
vim and ':make'

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

wafsamba: remove the need of BuildContext.bdir

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

wafsamba: remove hardcoded '..' and '/default/' from SAMBA_PIDL()

This makes it possible to remove some move waf 1.8 compat code.

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

wafsamba: add a fix for broken python threading if just one job is forced

This fixes random failures during (at least) configure on AIX.

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

wafsamba: fix CHECK_MAKEFLAGS() with waf 2.0.8

Changing Options.options.jobs in the build() hook
is too late in waf 2.

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

s4:heimdal_build: make use of libreplace getprogname() replacement

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

dsdb group_audit tests: fix use of strncmp

Replace the uses of:
   strncmp(expected, value, strlen(expected))
With:
    strcmp(expected, value)

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

netcmd: Flush replUpToDateVector when restoring offline backup

The replUpToDateVector could be incorrect after an offline backup was
restored. This means replication propagation dampening doesn't work
properly. In the worst case, a singleton DC would have no
replUpToDateVector at all, and so *all* objects created on that DC get
replicated every time a new DRS connection is established between 2 DCs.
This becomes a real problem if you used that singleton DC to create 100K
objects...

This patch flushes the replUpToDateVector when an offline backup gets
restored. We need to do this before we add in the new DC and remove the
old DCs.

Note that this is only a problem for offline backups. The online/rename
backups are received over DRS, and as part of the replication they
receive the latest replUpToDateVector from the DC being backed up.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

netcmd: Small backup refactor to avoid compatiblity problems

It will be easy to forget that the backupType marker doesn't exist on
v4.9. However, this seems like a dumb reason not to support v4.9
backup-files. Add a wrapper function to avoid potential problems
cropping up in future.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

netcmd: Add backupType marker to backed-up DB

We are starting to hit restore cases that are only applicable to a
particular type of backup. We already had a marker to differentiate
renames, but differentiating offline backups would also be useful.

Note that this raises a slight compatibility issue for backups created
on v4.9, as the marker won't exist. However, it's only offline backups
we will use this marker for (at the moment), and this option doesn't
exist on v4.9, so there's no problem.

Removing the markers has been refactored out into a separate function to
handle the optional presence of the new marker.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

tests: Add assertion that replUpToDateVector is present after backup

We noticed that offline backups were missing a replUpToDateVector for
the original DC, if the backup was taken on a singleton DC. This patch
adds an assertion to the existing test-cases to highlight the problem.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

WHATSNEW: Added samba-tool 'backup offline' and 'groups stats'

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

smbtorture: Close unused pipe fds in kernel_oplocks8

This fixes a hang of the testcase when hitting an error in the child (e.g.
localdir does not exist)

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Nov 16 21:52:13 CET 2018 on sn-devel-144

dsdb: Slightly simplify samdb_check_password

Avoid an "else" where we have the early return

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

lib: Fix CID 1441264 Error handling issues (CHECKED_RETURN)

This is not worth keeping a CID around :-)

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

torture: Fix the 32-bit build

Unfortunately there's no off_t printf specifier as there's one for
size_t. So we have to use intmax_t.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Nov 15 19:45:24 CET 2018 on sn-devel-144

s3: VFS: fake_acls: Fix Coverity CID 1435850 Uninitialized pointer read.

map_acl_perms_to_permset() can return an error, check it.

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

ctdb-tests: Make the debug hung script test cope with unreadable stacks

Ideally this would just involve using "test -r".  However, operating
system security features may mean that kernel stacks are not readable
even though they appear to be.

Instead, try reading that stack of a process on the test node.  If
that succeeds then so should reading the stack of the "stuck" sleep
process in the test.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13684

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Tim Beale <timbeale@catalyst.net.nz>
Autobuild-User(master): Tim Beale <timbeale@samba.org>
Autobuild-Date(master): Thu Nov 15 08:15:32 CET 2018 on sn-devel-144

third_party: Update socket_wrapper to version 1.2.1

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): Thu Nov 15 04:50:31 CET 2018 on sn-devel-144

s4:torture: Use 65520 for maxopenfiles

The socket_wrapper limit is 65535 open sockets.

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

s3: VFS: Remove logically dead code. Coverity CID: 1419117

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Böhme <slow@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Nov 14 21:35:45 CET 2018 on sn-devel-144

s3:passdb: Use discard_const_p() in py_passdb

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Uri Simchoni <uri@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Nov 14 11:46:06 CET 2018 on sn-devel-144

s4:dsdb: Use const char in py_dsdb_garbage_collect_tombstones()

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Uri Simchoni <uri@samba.org>

s4:librpc: Use discard_const_p for ndr/py_misc

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Uri Simchoni <uri@samba.org>

libldb: Fix const char in pyldb

../lib/ldb/pyldb.c: In function ‘PyDict_AsMessage’:
../lib/ldb/pyldb.c:90:22: error: initialization discards ‘const’ qualifier from pointer target type [-Werror=discarded-qualifiers]
 #define PyStr_AsUTF8 PyUnicode_AsUTF8
                      ^~~~~~~~~~~~~~~~
../lib/ldb/pyldb.c:1359:19: note: in expansion of macro ‘PyStr_AsUTF8’
   char *key_str = PyStr_AsUTF8(key);
                   ^~~~~~~~~~~~
../lib/ldb/pyldb.c: In function ‘py_ldb_msg_getitem_helper’:
../lib/ldb/pyldb.c:3336:7: error: assignment discards ‘const’ qualifier from pointer target type [-Werror=discarded-qualifiers]
  name = PyStr_AsUTF8(py_name);
       ^
../lib/ldb/pyldb.c: In function ‘py_ldb_msg_setitem’:
../lib/ldb/pyldb.c:3502:12: error: assignment discards ‘const’ qualifier from pointer target type [-Werror=discarded-qualifiers]
  attr_name = PyStr_AsUTF8(name);
            ^

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Uri Simchoni <uri@samba.org>

CI: Autobuild: Remove build_samba_ad_dc_2_py3 CI job

We now run a purepython3 ad-dc-2 test job, later when the whole
build is running under python3 we will resurrect build_samba_ad_dc_2_py3
but as (build_samba_ad_dc_py2) for python2

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Wed Nov 14 08:20:55 CET 2018 on sn-devel-144

CI: Add new CI job for new purepy3-ad-dc-2 autobuild task

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

script: Add new (temporary) pure python3 ad-dc-2 test

Ideally we want all the tests to run under python3 by default (no
special task for this) and then convert the existing '-py3' tasks
to run the python tests with python3.
However at the moment the convertion process is not ready to do this,
for a while we need to run separate autobuild tasks for this.

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4/scripting/bin: gensec_client.update needs bytes

PY3 test was failing as param passed to update was str rather than bytes

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python/samba/tests: Ensure samba_dnsupdate called with correct python version

We need to examine the contents of PYTHON env variable which should defined the
python version to be used when running tests.

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4/scripting/bin: PY3 convert servicePrincipalName attr to string

res[0]["servicePrincipalName"] is an instance of ldb.bytes in PY3
If we wish to get the string value we need to call the custom
str function which attempts to decode the bytes to 'utf8'

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed by: Noel Power <npower@samba.org>

python/samba/tests: PY3 Fix str/bytes issue for json.loads

Python 3.4 seems to need a string

    parsed = json.loads (out_jsobj)
  File "/usr/lib/python3.4/json/__init__.py", line 312, in loads
    s.__class__.__name__))
TypeError: the JSON object must be str, not 'bytes'

however Python 3.5 seems to be happy to consume bytes (or string)

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python/samba: PY3 fix can't compare string with int

Testing a string against an int value is illegal and
is not necessary in this case, this patch removes the
problematic test.

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4/dsdb/tests: PY3 port of samba4.ldap.acl.python

convert various attribute results to str so assert function
as expected for tests.
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4/dsdb/tests/python: PY3 port samba4.ldap.secdesc

User str/bytes as needed for various asserts

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4/dsdb/tests/python: PY3 port samba4.ldap_schema.python

+ Misc attributes needed to be converted to strings from bytes to
ensure various asserts work as expected.
+ Fix ndr_unpack call which needs bytes not str

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python/samba/netcmd: PY3 port samba4.blackbox.ldapcmp_restore

Convert attribute from bytes to str
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python/samba/netcmd: PY3 port samba.tests.samba_tool.edit

Need to write bytes to file

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3/torture: PY3 port samba.ntlm_auth

Use bytes in test rather than str

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python/samba/tests: PY3 port samba.tests.blackbox.netads_json

convert output from check_output to text
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python/samba/tests: Port samba.tests.blackbox.samba_dnsupdate for PY2/PY3

Make sure either the output of tests and/or the item we are searching match
in type. Output of cmd in python3 is bytes, depending on the was the test is
written it may be easier just to convert all output or just a single string
that is used in the test

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

testprogs/blackbox: PY3 bulk change for python scripts use correct python

Change all instance where python scripts are called so that the
correct python version as specified by $PYTHON is used

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python/samba/provision: PY3 port samba4.dlz_bind9.version

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

ctdb: Fix an out of bound array access

Found by cppcheck.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13680

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

lib:util Always initialize start and space

Found by cppcheck.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13680

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

libgpo: Make sure status is intialized

Found by cppcheck.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13680

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

librpc:ndr: Initialize inblob

Found by cppcheck.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13680

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

s3:rpcclient: Initialize domain_name

This could be passed uninitialized to dcerpc_netr_DsRGetDCName()

Found by cppcheck.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13680

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

s3: lib: nmbname: Ensure we limit the NetBIOS name correctly. CID: 1433607

Firstly, make the exit condition from the loop explicit (we must
never write into byte n, where n >= sizeof(name->name).

Secondly ensure exiting from the loop that n==MAX_NETBIOSNAME_LEN,
as this is the sign of a correct NetBIOS name encoding (RFC1002)
in order to properly read the NetBIOS name type (which is always
encoded in byte 16 == name->name[15]).

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 Nov 13 20:54:56 CET 2018 on sn-devel-144

libcli/smb: don't overwrite status code

The original commit c5cd22b5bbce724dcd68fe94320382b3f772cabf from bug
9175 never worked, as the preceeding signing check overwrote the status
variable.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=9175

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Tue Nov 13 17:28:45 CET 2018 on sn-devel-144

s4:torture/smb2/session: test smbXcli_session_set_disconnect_expired() works

This adds a simple test that verifies that after having set
smbXcli_session_set_disconnect_expired() a session gets disconnected
when it expires.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=9175

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

s3:smb2_sesssetup: check session_info security level before it gets talloc_move'd

We talloc_move() session_info to session->global->auth_session_info
which sets session_info to NULL.

This means security_session_user_level(NULL, NULL) will always return
SECURITY_ANONYMOUS so we never sign the session setup response.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Tue Nov 13 14:22:46 CET 2018 on sn-devel-144

s4:torture/smb2/session: session reauth response must be signed

This test checks that a session setup reauth is signed even when neither
client nor server require signing.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

s4:torture/smb2/session: add force_signing to test_session_expire1i

Existing callers pass true, so no change in behaviour. The next commit
adds an additional test that passes force_signing=false.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

s4:torture/smb2/session: require a signed session setup reauth response

All existing tests using this function require signing, so currently
this passes. A subsequent commit adds a test where neither client nor
server require signing and that's where this trap will explode.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

s4:torture/smb2/session: invalidate credential cache

Invalidate credential cache before connecting to the server, otherwise
we will reuse the credentials from the credential cache populated by the
preceeding tests.

Also invalidate it at the end, otherwise subsequent tests might run into
problems if the credentials expire while authenticating.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

libcli/smb: use require_signed_response in smb2cli_conn_dispatch_incoming()

This can be used by the upper layers to force checking a response is
signed. It will be used to implement verification of session setup
reauth responses in a torture test. That comes next.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

libcli/smb: defer singing check a little bit

This allows adding an additional condition to the if check where the
condition state may be modified in the "if (opcode ==
SMB2_OP_SESSSETUP)" case directly above.

No change in behaviour.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

libcli/smb: maintain require_signed_response in smbXcli_req_state

Not used for now, that comes next.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

libcli/smb: add smb2cli_session_require_signed_response()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

s3:selftest: also run smb2.session torture testsuite against ad_member

The next commit adds a subtest to the smb2.session testsuite that
requires Kerberos (ad_dc would work), but where neither SMB2 server or
client must require signing (ad_dc, being an AD DC, requires signing).

The ad_member environment supports Kerberos with the SMB2 server not
mandating signing, that'll do.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

s3:selftest: split "raw.session" and "smb2.session"

The next commit is going to add a testsuite to "smb2.session".

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13661

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

replace: Add check for variable program_invocation_short_name

It appears that wafsamba's configure() defines _GNU_SOURCE
unconditionally, so checking _GNU_SOURCE isn't enough to know if this
variable is available.

For example, it isn't available on AIX with the xlc compiler:

  [ 6/10] Compiling lib/replace/replace.c
  ...
  "../../lib/replace/replace.c", line 991.16: 1506-045 (S) Undeclared identifier program_invocation_short_name.

Instead, add a configure check for program_invocation_short_name and
use it.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Tue Nov 13 11:11:11 CET 2018 on sn-devel-144

nsswitch: Handle possible NULL return value of getprogname()

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

lib:replace: Do not leak the file pointer in rep_getprogname()

And return NULL on error.

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

smbd: Use wire_perms_to_unix in unix_perms_from_wire

Same code, less lines

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Nov 13 04:10:54 CET 2018 on sn-devel-144

winbindd: Fix crash when taking profiles

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13629
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Björn Baumbach <bbaumbach@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Nov 12 20:20:52 CET 2018 on sn-devel-144

pam_winbind: Set the request type as client name

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Mon Nov 12 13:44:09 CET 2018 on sn-devel-144

nss_wins: Set client process name

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

nss_winbind: Set the client process name

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

krb5_plugin: Set the wbclient process name

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

wbclient: Add wbcSetClientProcessName()

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

s3:winbind: Log client process name for PAM auth

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

s3:winbind: Log client process name in winbind_misc

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

s3:winbind: Log client process name

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