Noel Power [Thu, 3 May 2018 18:30:06 +0000 (19:30 +0100)]
s4/selftest: enable samba.tests.samba_tool.user for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 14 Aug 2018 10:43:44 +0000 (11:43 +0100)]
python/samba/tests: Adjust teardown so second run of tests succeeds
A second run of this test (e.g. with --extra-python) or even
an manual re-run and the cache is still there.
use of '--cache-ldb-initialize' will fail and so with the test
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 14 Aug 2018 10:25:35 +0000 (11:25 +0100)]
python/samba/tests: PY2/PY3 required changes for samba.tests.samba_tool.user
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 1 May 2018 11:51:41 +0000 (12:51 +0100)]
s4/selftest: enable samba.tests.dcerpc.raw_protocol for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 1 May 2018 11:24:47 +0000 (12:24 +0100)]
python/samba/test/dcerpc: Py3 port of samba.tests.dcerpc.raw_protocol
Port code to allow this test run with either py2 or py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 30 Apr 2018 17:31:36 +0000 (18:31 +0100)]
python/samba/tests: ensure byte content (not strings)
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 30 Apr 2018 15:35:59 +0000 (16:35 +0100)]
s4/selftest: enable samba.tests.dcerpc.dnsserver(ad_dc) for python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 30 Apr 2018 15:20:05 +0000 (16:20 +0100)]
s4/selftest: enable samba.tests.dcerpc.dnsserver for python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 28 Aug 2018 11:05:14 +0000 (12:05 +0100)]
selftest: Update known fail with py3 variant of samba.tests.dcerpc.dnsserver
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 30 Apr 2018 15:06:12 +0000 (16:06 +0100)]
s4/selftest: enable samba.tests.samba_tool.dnscmd for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 30 Apr 2018 14:47:05 +0000 (15:47 +0100)]
s4/selftest: Enable samba.tests.samba_tool.sites for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 27 Apr 2018 14:49:40 +0000 (15:49 +0100)]
s4/selftest: enable samba.tests.samba_tool.help for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 13 Aug 2018 16:25:37 +0000 (17:25 +0100)]
python/samba/tests: Py2/Py2 enable samba.tests.samba_tool.help to run
Ensure bytes output is converted to text type for PY3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 27 Apr 2018 14:13:49 +0000 (15:13 +0100)]
s4/selftest: Enable samba.tests.samba_tool.provision_password_check for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 27 Apr 2018 14:11:54 +0000 (15:11 +0100)]
python/samba/netcmd: Fix password usage for py2/py3 compatability
getpass returns str (e.g. bytes) in python2 and str (unicode) in
py3. Adapt code to so we don't do illegal things (like try and decode)
a string in python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 27 Apr 2018 12:05:15 +0000 (13:05 +0100)]
s4/selftest: enable samba.tests.samba_tool.ntacl for Py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 27 Apr 2018 12:02:50 +0000 (13:02 +0100)]
python/samba/provision: remove use of str() func for binary data
Python 2 code works with str(policy["nTSecurityDescriptor"]) however
this cannot work with Python 3. One could argue even the str method
doesn't make sense at all (returning a string) for data.
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Thu, 26 Apr 2018 17:22:21 +0000 (18:22 +0100)]
python/samba/netcmd: changes for samab.tests.samba_tool.computer
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Thu, 26 Apr 2018 14:47:19 +0000 (15:47 +0100)]
s4/selftest: enable samba.tests.samba_tool.ou for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 24 Apr 2018 18:08:27 +0000 (19:08 +0100)]
s4/selftest: enable samba.tests.group for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Sat, 11 Aug 2018 13:25:26 +0000 (14:25 +0100)]
python/sambe/tests/samba_tool: PY3 port for tests.samba_tool.group
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 24 Apr 2018 15:55:49 +0000 (16:55 +0100)]
s4/selftest: enable samba.tests.samba_tool.user_check_password_script py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 23 Apr 2018 17:22:08 +0000 (18:22 +0100)]
s4/selftest: enable samba.tests.samba_tool.user_wdigest for python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Sat, 11 Aug 2018 13:02:25 +0000 (14:02 +0100)]
python/samba: PY3 port for samba.tests.samba_tool.user_wdigest test
In addition to the attributes that caused some issues specifically
with the test some other ldb.bytes objects (those seen to be
used as strings) have been adjusted (with str()) to ensure they should
work correct in PY3.
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Sat, 11 Aug 2018 12:54:51 +0000 (13:54 +0100)]
python/samba: Add some helper functions to encode/decode strings
Sometimes in PY3 we have variables whose content can be 'bytes' or
'str' and we can't be sure which. Generally this is because the
code variable to be used can be initialised (or reassigned) a value
from different api(s) or functions depending on complex conditions or
logic. Or another common case is in PY2 the variable is 'type <str>'
and in PY3 it is 'class <str>' but the function to use e.g. b64encode
requires 'bytes' in PY3. In such cases in PY3 it would be nice to avoid
excessive testing in the client code, in PY2 we would like to ensure the
code runs unchanged. In otherwords we wish to avoid in PY2 unnecessary
decode() calls which convert otherwise 'str' types to 'unicode'.
Mixing 'str' and 'unicode' can cause strange and unexpected problems,
sometimes we can't avoid this but it we should avoid if we can.
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 10 Aug 2018 20:23:35 +0000 (21:23 +0100)]
s4/selftest: enable samba.tests.samba_tool.processes for PY3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 23 Apr 2018 11:21:53 +0000 (12:21 +0100)]
s3/selftest: Enable samba.tests.samba_tool.fsmo for python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 10 Aug 2018 19:29:46 +0000 (20:29 +0100)]
python/samba/tests: Port fsmo test to python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 20 Apr 2018 16:24:44 +0000 (17:24 +0100)]
enable samba_tool.visualize_drs samba_tool.visualize_drs for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 10 Aug 2018 17:07:11 +0000 (18:07 +0100)]
python/samba: changes needed for samba.tests.samba_tool.visualize PY2/PY3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 17 Apr 2018 15:48:56 +0000 (16:48 +0100)]
s4/selftest: Enable samba.tests.samba_tool.join for py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 17 Apr 2018 15:48:03 +0000 (16:48 +0100)]
python/samba: changes to make samba.tests.samba_tool.join run under py3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 10 Aug 2018 18:54:46 +0000 (19:54 +0100)]
python/compat: use cStringIO instead of StringIO in compat.py
Using cStringIO for py2 seems to incur alot less problems and less changes
to the py2/py3 code then using StringIO.StringIO
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Fri, 4 May 2018 10:49:23 +0000 (11:49 +0100)]
python/samba/tests: Use io.StringIO for py3 and StringIO.StrinIO for py2
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 17 Apr 2018 12:53:57 +0000 (13:53 +0100)]
s4/selftest/tests: enable samba.tests.samba_tool.timecmd for python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 16 Apr 2018 17:58:01 +0000 (18:58 +0100)]
s4/selftest: enable samba.tests.dcerpc.unix for python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 17 Apr 2018 10:56:40 +0000 (11:56 +0100)]
python/samba/tests: port samba.tests.unix for py3
L for long literal is no longer valid in Python3, also
long keyword has been removed. int and long have been unified
in python3. A duplicated test has been removed (for long and
int) only the int test remains now.
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Tue, 10 Apr 2018 15:03:37 +0000 (16:03 +0100)]
s4/selftest: enable samba.tests.dsdb_lock for python3
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Wed, 8 Aug 2018 16:03:48 +0000 (17:03 +0100)]
python/samba: port changes to allow samba.tests.dsdb_lock to work with PY3/PY2
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Wed, 8 Aug 2018 15:56:33 +0000 (16:56 +0100)]
python/samba: open files in binary mode as appropriate
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Wed, 8 Aug 2018 15:52:35 +0000 (16:52 +0100)]
python/samba: port open to io.open for PY2/PY3 compat
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Noel Power [Mon, 30 Apr 2018 14:46:04 +0000 (15:46 +0100)]
python/samba: Fix py2/3 relative module import issue
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Andrew Bartlett [Fri, 31 Aug 2018 08:26:19 +0000 (20:26 +1200)]
selftest: Remove excption for NSS_WRAPPER_HOSTS and RESOLV_WRAPPER_HOSTS
These must be set correctly for each command in provision also.
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>
Andrew Bartlett [Sat, 1 Sep 2018 11:12:44 +0000 (23:12 +1200)]
selftest: Set RESOLV_WRAPPER_CONF/RESOLV_WRAPPER_HOSTS when running dcpromo
Otherwise this relies on the order that tests run to cause the environment variable
to be left behind.
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>
Andrew Bartlett [Fri, 31 Aug 2018 08:22:54 +0000 (20:22 +1200)]
selftest: Set NSS_WRAPPER_HOSTS when creating the trusts
Otherwise this relies on the order that tests run to cause the environment variable
to be left behind.
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>
Ralph Boehme [Thu, 30 Aug 2018 13:57:33 +0000 (15:57 +0200)]
s3:smbd: add a comment stating that file_close_user() is redundant for SMB2
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Sep 1 01:26:35 CEST 2018 on sn-devel-144
Ralph Boehme [Wed, 29 Aug 2018 15:19:29 +0000 (17:19 +0200)]
s3:smbd: let session logoff close files and tcons before deleting the session
This avoids a race in durable handle reconnects if the reconnect comes
in while the old session is still in the tear-down phase.
The new session is supposed to rendezvous with and wait for destruction
of the old session, which is internally implemented with
dbwrap_watch_send() on the old session record.
If the old session deletes the session record before calling
file_close_user() which marks all file handles as disconnected, the
durable handle reconnect in the new session will fail as the records are
not yet marked as disconnected which is a prerequisite.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Ralph Boehme [Thu, 30 Aug 2018 13:50:02 +0000 (15:50 +0200)]
s3:smbd: reorder tcon global record deletion and closing files of a tcon
As such, this doesn't change overall behaviour, but in case we ever add
semantics acting on tcon record changes via an API like
dbwrap_watch_send(), this will make a difference as it enforces
ordering.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Ralph Boehme [Thu, 30 Aug 2018 17:15:19 +0000 (19:15 +0200)]
selftest: add a durable handle test with delayed disconnect
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Ralph Boehme [Fri, 31 Aug 2018 06:28:46 +0000 (08:28 +0200)]
s4:selftest: reformat smb2_s3only list
No change besides reformatting the list to one entry per line.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Ralph Boehme [Thu, 30 Aug 2018 15:27:08 +0000 (17:27 +0200)]
vfs_delay_inject: adding delay to VFS calls
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Volker Lendecke [Fri, 31 Aug 2018 04:25:30 +0000 (06:25 +0200)]
lib: Fix the build on FreeBSD
FreeBSD needs to explicitly #include <unistd.h> for geteuid() and close()
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Fri Aug 31 18:42:31 CEST 2018 on sn-devel-144
Christof Schmitt [Wed, 29 Aug 2018 19:07:42 +0000 (12:07 -0700)]
torture: Add test for directory_create_or_exist
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Christof Schmitt <cs@samba.org>
Autobuild-Date(master): Thu Aug 30 21:19:31 CEST 2018 on sn-devel-144
Christof Schmitt [Wed, 29 Aug 2018 19:04:29 +0000 (12:04 -0700)]
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>
Martin Schwenke [Fri, 17 Aug 2018 11:33:39 +0000 (21:33 +1000)]
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
Martin Schwenke [Tue, 28 Aug 2018 06:10:21 +0000 (16:10 +1000)]
ctdb-common: Check the version field in IPv6 packets
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Fri, 17 Aug 2018 11:26:04 +0000 (21:26 +1000)]
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>
Martin Schwenke [Fri, 17 Aug 2018 06:02:50 +0000 (16:02 +1000)]
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>
Martin Schwenke [Fri, 17 Aug 2018 05:00:31 +0000 (15:00 +1000)]
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>
Martin Schwenke [Fri, 17 Aug 2018 04:35:07 +0000 (14:35 +1000)]
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>
Martin Schwenke [Fri, 17 Aug 2018 04:14:12 +0000 (14:14 +1000)]
ctdb-common: Fix error handling when parsing TCP packets
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Fri, 17 Aug 2018 05:36:41 +0000 (15:36 +1000)]
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>
Martin Schwenke [Fri, 17 Aug 2018 03:46:50 +0000 (13:46 +1000)]
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>
Martin Schwenke [Fri, 17 Aug 2018 02:30:19 +0000 (12:30 +1000)]
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>
Martin Schwenke [Thu, 16 Aug 2018 06:47:21 +0000 (16:47 +1000)]
ctdb-tests: Add tests for TCP packet marshalling
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Thu, 16 Aug 2018 01:10:40 +0000 (11:10 +1000)]
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>
Martin Schwenke [Tue, 14 Aug 2018 03:49:25 +0000 (13:49 +1000)]
ctdb-common: Avoid single line multi-assignment
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Wed, 29 Aug 2018 05:48:21 +0000 (15:48 +1000)]
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>
Martin Schwenke [Fri, 10 Aug 2018 08:38:23 +0000 (18:38 +1000)]
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>
Martin Schwenke [Tue, 14 Aug 2018 02:36:25 +0000 (12:36 +1000)]
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>
Martin Schwenke [Tue, 14 Aug 2018 01:25:02 +0000 (11:25 +1000)]
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>
Martin Schwenke [Thu, 16 Aug 2018 05:15:55 +0000 (15:15 +1000)]
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>
Martin Schwenke [Mon, 13 Aug 2018 11:39:12 +0000 (21:39 +1000)]
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>
Martin Schwenke [Fri, 10 Aug 2018 07:23:56 +0000 (17:23 +1000)]
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>
Martin Schwenke [Fri, 10 Aug 2018 07:04:32 +0000 (17:04 +1000)]
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>
Martin Schwenke [Fri, 10 Aug 2018 06:14:40 +0000 (16:14 +1000)]
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>
Martin Schwenke [Sat, 11 Aug 2018 08:32:17 +0000 (18:32 +1000)]
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>
Martin Schwenke [Fri, 17 Aug 2018 03:58:18 +0000 (13:58 +1000)]
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>
Martin Schwenke [Tue, 14 Aug 2018 03:18:01 +0000 (13:18 +1000)]
ctdb-common: Fix CID
1414745 - Out-of-bounds access
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Mon, 27 Aug 2018 04:53:37 +0000 (14:53 +1000)]
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>
Martin Schwenke [Mon, 27 Aug 2018 04:47:38 +0000 (14:47 +1000)]
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>
Martin Schwenke [Mon, 27 Aug 2018 04:44:24 +0000 (14:44 +1000)]
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>
Martin Schwenke [Mon, 27 Aug 2018 05:28:47 +0000 (15:28 +1000)]
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>
Martin Schwenke [Fri, 24 Aug 2018 04:52:29 +0000 (14:52 +1000)]
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>
Martin Schwenke [Fri, 24 Aug 2018 04:44:12 +0000 (14:44 +1000)]
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>
Alexander Bokovoy [Thu, 5 Jul 2018 13:39:52 +0000 (16:39 +0300)]
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
Andreas Schneider [Mon, 27 Aug 2018 13:59:59 +0000 (15:59 +0200)]
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
Andreas Schneider [Thu, 23 Aug 2018 12:12:43 +0000 (14:12 +0200)]
selftest: Schedule libsmbclient tests against ad_dc
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Andreas Schneider [Thu, 23 Aug 2018 12:00:29 +0000 (14:00 +0200)]
s4:torture: Improve torture_libsmbclient_init_context()
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Andrew Bartlett [Mon, 27 Aug 2018 09:13:29 +0000 (21:13 +1200)]
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
Andrew Bartlett [Mon, 27 Aug 2018 09:00:58 +0000 (21:00 +1200)]
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>
Andrew Bartlett [Mon, 27 Aug 2018 01:34:35 +0000 (13:34 +1200)]
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>
Andrew Bartlett [Sat, 25 Aug 2018 07:40:12 +0000 (09:40 +0200)]
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>
Andrew Bartlett [Sat, 25 Aug 2018 07:03:07 +0000 (09:03 +0200)]
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>
Andrew Bartlett [Thu, 16 Aug 2018 01:54:14 +0000 (13:54 +1200)]
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>
Andrew Bartlett [Thu, 16 Aug 2018 01:53:20 +0000 (13:53 +1200)]
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>
Andrew Bartlett [Thu, 16 Aug 2018 01:50:31 +0000 (13:50 +1200)]
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>
Volker Lendecke [Tue, 10 Apr 2018 15:18:18 +0000 (17:18 +0200)]
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
Volker Lendecke [Wed, 4 Apr 2018 14:32:01 +0000 (16:32 +0200)]
libsmb: Expose protocol-agnostic cli_writeall_send/recv
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Volker Lendecke [Wed, 4 Apr 2018 14:19:52 +0000 (16:19 +0200)]
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>
git.samba.org / samba.git / log