lib: Fix lstat check in directory_create_or_exist

The lstat check in directory_create_or_exist did not verify whether an
existing object is actually a directory. Also move the check to only
apply when mkdir returns EEXIST; this fixes CID 241930 Time of check
time of use.

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

ctdb-common: Clean up comments in TCP packet parsing

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Thu Aug 30 07:50:04 CEST 2018 on sn-devel-144

ctdb-common: Check the version field in IPv6 packets

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

ctdb-common: Improve TCP packet size and offset calculations

The IPv4 check for short packets was strange.  It appeared to ensure
that the capture included everything up to and including the window
size.  The checksum field immediately follows the window size field,
so just ensure that the packet is large enough to contain everything
up to the start of the checksum.

Add a similar check for IPv6 packets.

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

ctdb-tests: Extend TCP packet test to also do packet extraction

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

ctdb-common: Factor out TCP packet parsing code

This can be tested separately.

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

ctdb-common: Clean up types/declarations in TCP socket reading

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

ctdb-common: Fix error handling when parsing TCP packets

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

ctdb-common: Fix a bug in non-Linux (PCAP) TCP packet capturing

Captured packets include a link-layer header, which is considered in
the Linux code but not the PCAP code.  Also, the actual captured
length is in caplen, not len.

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

ctdb-common: Don't modify a const argument

The current code might be slightly more efficient but
intentionally (although temporarily) modifying a const argument just
seems wrong.

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

ctdb-common: Avoid magic numbers when building TCP packets

Most packet sizes and offsets are multiples of 32-bit words.  The IPv6
payload length is in octets.  The IPv6 version is the top 4 bits of
the relevant field.

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

ctdb-tests: Add tests for TCP packet marshalling

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

ctdb-common: Factor out TCP packet marshalling code

This can be tested separately.

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

ctdb-common: Avoid single line multi-assignment

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

ctdb-common: Set version more obviously in IPv6 NA packet

Version is the top 4 bits of this field.

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

ctdb-common: Clarify offset and packet length calculations

Calculate each offset from the beginning of the buffer and explicitly
use the sizes of structures.

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

ctdb-common: Use struct ether_arp to avoid manual offset calculations

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

ctdb-common: Be more careful with packet sizes

Ethernet packets must be at least 64 bytes.

For ARP the packet size was limited to 64 bytes.  This is probably OK
but the code might as well be a little more general.

For IPv6 NA there was no guarantee that the packet is at least 64
bytes.

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

ctdb-tests: Add tests for ARP and IPv6 NA marshalling

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

ctdb-common: Separate ARP and IPv6 NA marshalling code

This can be tested separately.

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

ctdb-common: Fix error handling when sending ARPs

There are numerous places in the code where errno can be lost causing
the wrong error to be printed by a caller.  Change ctdb_sys_send_arp()
to always return a useful errno on error instead of returning -1 and
sometimes having errno set correctly.

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

ctdb-common: Factor out common ARP code

Finding the interface and the MAC address are obvious.  Might as well
set up the common parts of the destination address structure.

Continue to open the socket and find the MAC address first.  This
might seem odd because marshalling and other subsequent steps may
fail.  However, in the future this code might be optimised to open a
single socket to send ARPs for a list of addresses on each interface,
so don't change the logic.

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

ctdb-common: Initialise structures when declared

Instead of using ZERO_STRUCT().

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

ctdb-tests: Add basic test to sanity check types in socket marshalling

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

ctdb-common: Restore dropped copyright attributions

Commit fa94a49dbbec4a65c368a533a534f952a9f147a7 accidentally dropped
some copyright attributions.  The original version of system_socket.c
was based on system_linux.c but many parts have been taking from
system_freebsd.c, which had these additional copyright attributions.

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

ctdb-common: Fix CID 1414745 - Out-of-bounds access

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

ctdb-daemon: Do not retry connection to eventd

Confirmation is now received from eventd that it is accepting
connections, so this is no longer needed.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13592

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

ctdb-daemon: Wait for eventd to be ready before connecting

The current method of retrying the connection to eventd means that
messages get logged for each failure.

Instead, pass a pipe file descriptor to eventd and wait for it to
write 0 to the pipe to indicate that it is ready to accept client
connections.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13592

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

ctdb-daemon: Open eventd pipe earlier

The pipe will soon be needed earlier, so initialise it earlier.
Ensure the file descriptors are closed on error.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13592

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

ctdb-daemon: Improve error handling consistency

Other errors free argv, so do it here too.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13592

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

ctdb-event: Add support to eventd for the startup notification FD

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13592

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

ctdb-common: Add support for sock daemon to notify of successful startup

The daemon writes 0 into the specified file descriptor when it is up
and listening.  This can be used to avoid loops in clients that
attempt to connect until they succeed.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13592

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

pidl/tests: fix ndr_push_init_ctx() usage

The code in libndr got rid of the second argument quite some time ago
(2010): f9ca9e46ad24036bf00cb361a6cef4b2e7e98d7d

However, pidl tests did include a code that passed two arguments.
Recently GCC started to fail to compile such code.

Signed-off-by: Alexander Bokovoy <ab@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Thu Aug 30 04:48:00 CEST 2018 on sn-devel-144

s4:torture: Add a test for listing shares

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 Aug 29 21:28:41 CEST 2018 on sn-devel-144

selftest: Schedule libsmbclient tests against ad_dc

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

s4:torture: Improve torture_libsmbclient_init_context()

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

autobuild: Avoid subshell for tail -f invocation

This should mean one less process in the process tree, and less places to hold
FDs open.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13591

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Wed Aug 29 08:20:55 CEST 2018 on sn-devel-144

autobuild: use close_fds=True to avoid *.stderr and *.stdout inheriting into every process

This closes fds other than 0, 1, 2.

This ensures only the correct *.stderr and *.stdout is attached, via
the stdout/stderr parameter to Popen(), but not every other FD
currently open in python at the time Popen is called.

For the tail invocation and other calls to Popen(), because fds 0, 1,
2 are still attached, these function as before.

Per https://docs.python.org/2.6/library/subprocess.html:

"If close_fds is true, all file descriptors except 0, 1 and
2 will be closed before the child process is executed. (Unix only)."

And regarding the passed in parameters:

"stdin, stdout and stderr specify the executed programs’ standard
input,
standard output and standard error file handles, respectively.  "
...

"With None (the default), no redirection will occur;
the child’s file handles will be inherited from the parent. "

(The unwanted inherited files would be on a random high FD, where the
program wouldn't know what to do with them, but counting towards the
process FD limit).

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13591

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

travis-ci: Add py3 jobs to match new jobs in autobuild.py

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

autobuild: Move backup/restore environments into to samba-ad-dc-2 autobuild job

This tries to to split up the tasks more evenly and may help with the python3
work by isolating them from the long samba job.

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

autobuild: Move ad_dc_no_nss into to samba-ad-dc-2 autobuild job

This tries to to split up the tasks more evenly and may help with the python3 tests
against this environment if started from a more isolated job.

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

autobuild: Reduce duplication of task list in autobuild

The defaulttasks or builddirs are often updated out of sync, which causes confusion until
it is resolved.

We simply choose "." as the builddir for the tasks that
are not in the default set.

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

gitlab-ci: Run the new python3 autobuild tasks

These additional tasks should be less complex than the full build and help get us to
a pure python3 build eventually

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

autobuild: Implement a split python2 and python2 build pattern

The default tasks will run the tests without --extra-python specified and
the new -py3 tasks will run the python3 tests only.

This will reduce the complexity of the build combinations

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

examples: Add winexe re-implemented on current Samba libs

winexe from https://sourceforge.net/projects/winexe/ is a project
based on Samba libraries from 2012. According to the winexe git
repository the last Samba commit winexe was updated to is 47bbf9886f0c
from November 6, 2012. As winexe uses unpublished Samba internal
libraries, it broke over time.

This is a port of the winexe functionality to more modern Samba
versions. It still uses internal APIs, but it being part of the tree
means that it is much easier to keep up to date.

The Windows service files were taken literally from the original
winexe from the sourceforge git. Andrzej Hajda chose GPLv3 only and
not GPLv3+. As GPL evolves very slowly, this should not be a practical
problem for quite some time.

To build it under Linux, you need mingw binaries on your build
system. Under Debian stretch, the package names are gcc-mingw-w64 and
friends.

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 Aug 28 02:03:07 CEST 2018 on sn-devel-144

libsmb: Expose protocol-agnostic cli_writeall_send/recv

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

libsmb: Rename cli_writeall_send/recv to cli_smb1_writeall_send/recv

Preparing a protocol agnostic writeall

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

libsmb: Add protocol-agnostic cli_read

So far only cli_pull could be called directly without looking at the
protocol. We did not have a simple read that did the right thing
depending on the protocol

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

s3: VFS: vfs_full_audit: Ensure smb_fname_str_do_log() only returns absolute pathnames.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13565

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Mon Aug 27 20:23:55 CEST 2018 on sn-devel-144

s3: VFS: vfs_full_audit: Add $cwd arg to smb_fname_str_do_log().

Not yet used.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13565

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

travis-ci: Add python3-crypto to package list

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Mon Aug 27 12:00:11 CEST 2018 on sn-devel-144

autobuild: Fix -py3 support to look in tasks (the table with the tests) not tasknames (the requested list)

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

autobuild: Allow automatic handling of autobuild tasks ending in -py3

This will allow splitting up of the python2 and python3 tests without
duplication of this already complex file.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Sat Aug 25 11:53:15 CEST 2018 on sn-devel-144

selftest: Move samba.tests.lsa_string to none environment, it does not contact a server

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

selftest: Move samba.tests.dcerpc.rpcecho to chgdcpass:local

Before this patch, no python3-compatible tests ran against any of fl2000dc, vampire_2000_dc or chgdcpass
and so an autobuild of the samba-ad-dc-2-py3 environment would fail with no tests to run.

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

selftest: Ensure all python3 tests are tagged (named) python3

This will help when splitting them into a distinct build.

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

selftest: Ensure winbindd is talking to the DC (itself) at startup

This might reduce issues with the first winbind-using test failing

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

smbd: Remove unused KOPLOCK flags

This effectively reverts commit 17eba16b. It looks like these flags have
been introduced as part of the onefs support which has been removed
again. As there is no other use for the flags, remove them.

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Christof Schmitt <cs@samba.org>
Autobuild-Date(master): Sat Aug 25 03:23:05 CEST 2018 on sn-devel-144

Fix uClibc build on 64bit platforms by including stdint.h

Fixes an error detected by buildroot autobuilders:
http://autobuild.buildroot.net/results/573/573e2268e205e10d1352fa81122d8f225fdb4575/build-end.log

/home/rclinux/rc-buildroot-test/scripts/instance-1/output/host/mips64el-buildroot-linux-uclibc/sysroot/usr/include/stdint.h:122:27:
error: conflicting types for 'uintptr_t'
 typedef unsigned long int uintptr_t;
                           ^
In file included from ../lib/ldb/tests/ldb_msg.c:17:0:
../third_party/cmocka/cmocka.h:126:28: note: previous declaration of 'uintptr_t' was here
       typedef unsigned int uintptr_t;

The define __WORDSIZE is missing when cmocka.h decides how to
define uintptr_t, this patch includes stdint.h when needed.

Patch sent upstream:
https://lists.samba.org/archive/samba-technical/2018-January/125306.html

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Fri Aug 24 17:22:10 CEST 2018 on sn-devel-144

ctdb-tests: Check that no IPs are assigned when failover is disabled

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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): Fri Aug 24 14:13:12 CEST 2018 on sn-devel-144

ctdb-tests: Add an extra conf loading test case

This shows that config file loading continues in spite of unknown keys
if ignore_unknown is true.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-doc: Switch tunable DisableIPFailover to a config option

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-config: Switch tunable DisableIPFailover to a config option

Use the "failover:disabled" option instead.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-config: Integrate failover options into conf-tool

Update and add tests accordingly.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-failover: Add failover configuration options

Only a "disabled" option for now.  Not documented because it isn't
used yet.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-tests: Drop DisableIPFailover simple test

This is about to become a config file option that can't be dynamically
changed at run-time, so drop this test for now.  This test will be added
once the tunable becomes a config file option.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-daemon: Pass DisableIPFailover tunable via environment variable

Preparation for obsoleting this tunable.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-common: Allow boolean configuration values to have yes/no values

This make the new configuration style more consistent with the old one.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-doc: Switch tunable TDBMutexEnabled to a config option

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-config: Switch tunable TDBMutexEnabled to a config option

Use the "database:tdb mutexes" option instead.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-doc: Add support for migrating tunables to ctdb.conf options

This will become common, so will be useful to have support for.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-doc: Change option "no realtime" option to "realtime scheduling"

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-config: Change option "no realtime" option to "realtime scheduling"

Negative options can be confusing, so switch to a positive option.

This was supposed to be done months ago but was forgotten.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-doc: Handle boolean options in config migration more carefully

Values for ctdb.conf options are now returned by
get_ctdb_conf_option().  The main goal is to allow old boolean options
to be replaced by new logically negated options.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-doc: Make config migration script notice removed CTDB_BASE option

This should never have been a user-level option, but some people used
it.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13589

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

ctdb-common: Fix aliasing issue in IPv6 checksum

Since commit 9c51b278b1700cd5f3e2addc19b7c711cc2ea10b the compiler has
been able to inline the affected call to uint16_checksum().  Given
that the data (phdr) is being accessed by an incompatible
pointer (data) there is an aliasing problem when the call is inlined.
This results in incorrect behaviour with -O2/-O3 when compiling with
at least GCC 6, 7, and 8.

Fix this by making the types compatible.

Also fixes CID 1437604 (Reliance on integer endianness).  This is a
false positive because the uint16_checksum doesn't depend on the order
of the input uint16_t items.

https://bugzilla.samba.org/show_bug.cgi?id=13588

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

PEP8: line up a couple of lists

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): Fri Aug 24 10:58:16 CEST 2018 on sn-devel-144

PEP8: improve formatting around various operators

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

PEP8: improve formatting around bit-wise OR ("|")

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

dcerpc py tests: improve argument formatting

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

dcerpc/dns_server tests: use record malformed as probably intended

Both ways the record is malformed, but it is more likely we meant
AAAAAAAAAA... 1
than
A 1A 1A 1A 1A ...

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

PEP8: add spaces after operators

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

PEP8: better formatting of (CONST1|CONST2)

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

ldb tests: remove unused code from match_rules.py

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

PEP8: improve spacing around colons

These dropped out of Joe's patches during rebase and review.

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

PEP8: more space before equals

This dropped out during rebase

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

PEP8: add pycodestyle config in setup.cfg

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

PEP8: fix W391: blank line at end of file

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

PEP8: fix W293: blank line contains whitespace

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

PEP8: fix W291: trailing whitespace

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

PEP8: fix E713: test for membership should be 'not in'

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

PEP8: fix E712: comparison to False should be 'if cond is False:' or 'if not cond:'

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

PEP8: fix E711: comparison to None should be 'if cond is not None:'

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

PEP8: fix E703: statement ends with a semicolon

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

PEP8: fix E701: multiple statements on one line (colon)

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

PEP8: fix E502: the backslash is redundant between brackets

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

PEP8: fix E401: multiple imports on one line

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

PEP8: fix E306: expected 1 blank line before a nested definition, found 0

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

PEP8: fix E305: expected 2 blank lines after class or function definition, found 1

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

PEP8: fix E303: too many blank lines (2)

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