libcli/security: sddl conditional ACE: write -0 when asked

Credit to OSS-Fuzz.

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=65122

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

libcli/security: rearrange conditional ACE sddl_write_int

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=65122

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

libcli/security: tests for signed zeros in sddl condtional ACEs

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=65122

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

librpc: Do not allow u16string to be encoded in a big‐endian context

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

librpc: Change type of ‘u16string’ from ‘const uint16_t *’ to ‘const unsigned char *’

A u16string is supposed to contain UTF‐16 code units, but
ndr_pull_u16string() and ndr_push_u16string() fail to correctly ensure
this on big‐endian systems. Code that relies on the u16string array
containing correct values will then fail.

Fix ndr_pull_u16string() and ndr_push_u16string() to work on big‐endian
systems, ensuring that other code can use these strings without having
to worry about first encoding them to little‐endian.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

librpc: Add missing spaces to error messages

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3/lib: add log_panic_action()

Can be used to log a nice stack backtrace with full debug symbols by setting
"panic action" to something like

    panic action = cd /home/slow/git/samba/master && /home/slow/git/samba/master/selftest/gdb_backtrace %d

This is similar to log_stack_trace(), but that doesn't come with debug symbols.

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3/lib: factor out call_panic_action() from smb_panic_s3()

No change in behaviour. Best viewed with git show -w.

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

tests/krb5: Add Python implementation and tests for Group Key Distribution Service

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Thu Dec 21 21:19:30 UTC 2023 on atb-devel-224

python: Add NT Time utility functions

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

pyglue: Export some GKDI constants

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

pyglue: Export some more HRESULT constants

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

pyglue: Fix code spelling

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:crypto: Add GKDI module with some constants

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

gkdi.idl: Verify magic numbers of pulled GKDI structures

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4:libcli: Fix conversion from HRESULT and WERROR to Python objects

The inner values of HRESULT and WERROR are 32‐bit unsigned integers,
which might not be representable in type ‘int’. We must then use the ‘k’
format specifier, which corresponds to ‘unsigned long’, a type
guaranteed to be at least 32 bits in size.

Commit c81aff362fe99a65385c6f8337ffcb47c9456829 fixed
PyErr_FromNTSTATUS(), but it did not attempt to fix the other cases.

PyErr_FromHRESULT() might return a tuple like this:
(-2147024809, 'One or more arguments are invalid.')

which, after this commit, will become this:
(2147942487, 'One or more arguments are invalid.')

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4:libcli: Remove trailing whitespace

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python:tests: Catch strings passed to utf16_encoded_len() with embedded nulls

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

selftest: Rename ‘samba.unittests.test_gnutls_sp800_108’ to something more consistent with existing tests

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

gkdi.idl: Add ‘additional_info’ field to KeyEnvelope structure

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

gkdi.idl: Comment on domain and forest name fields

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:utils: Do not pass invalid file descriptor to close() (CID 1550131)

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

tests/krb5: Remove redundant definitions

These items are already defined elsewhere.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

ldb: Fix code spelling

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

buildtools: Use correct variable in error message

When this error message was copied from CHECK_FUNCS_IN(), the variable
name was not changed. This results in messages like this:

“Mandatory library 'lib' not found for functions '<class 'list'>'”

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:cmdline: Fix code spelling

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

pyglue: Remove unnecessary uses of discard_const_p()

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python:tests: Don’t needlessly create single‐element tuple

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:util: Use portable integer constants

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:param: Remove unnecessary use of discard_const_p()

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python:tests: Rename parameter to be consistent with overridden method

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python:tests: Raise exception of more specific type NotImplementedError

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

python:tests: Use ‘False’ in boolean expression rather than ‘None’

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

librpc:ndr: Don’t unnecessarily parenthesize macro arguments

If we’re just passing a parameter to another macro which we know
correctly parenthesizes its arguments, then we don’t need to
parenthesize the parameter ourselves.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:util: Remove redundant casts in PUSH_*() macros

The PUSH_*() macros already cast their arguments to the expected type,
so we don’t need to cast the arguments *again* prior to invoking the
macros.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:util: Don’t unnecessarily parenthesize macro arguments

If we’re just passing a parameter to another macro which we know
correctly parenthesizes its arguments, then we don’t need to
parenthesize the parameter ourselves.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:util: Cast macro parameter ‘val’ to expected type

These macros are now consistent with PUSH_BE_U8() and with the
PUSH_LE_*() macros.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:util: Parenthesize macro parameters

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s4:auth: Clarify comment about requiring FAST armor

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:lib: Define TIME_FIXUP_CONSTANT_INT using INT64_C() macro

This is more portable than using preprocessor conditionals.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

lib:util: Define TIME_FIXUP_CONSTANT_INT using INT64_C() macro

This is more portable than using preprocessor conditionals.

Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:ctdbd_conn: fix ctdbd_public_ip_foreach() for ipv6 addresses

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15534

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Dec 21 11:09:30 UTC 2023 on atb-devel-224

WHATSNEW: Add entry for "samba-tool user get-kerberos-ticket"

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Autobuild-User(master): Douglas Bagnall <dbagnall@samba.org>
Autobuild-Date(master): Thu Dec 21 03:04:12 UTC 2023 on atb-devel-224

selftest: Add tests for "samba-tool user get-kerberos-ticket"

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

python/netcmd: Improve documentation for "samba-tool user getpassword"

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

python/netcmd: Add "samba-tool user get-kerberos-ticket" to get a ticket for a gMSA

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

WHATSNEW: Add entry for "samba-tool user getpassword" changes

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

python: tests: blackbox test for GMSA

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

samba-tool: document that -H can be used with gMSA accounts

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

samba-tool: fix some grammar in getpassword docstrings

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

samba-tool: Make samba-tool user getpassword support a ';previous=1' option

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

samba-tool user getpassword: Prepare to support a ;previous=1 option, change behaviour for ;rounds=

This will return the previous password, but the pattern is to include
the option in the returned attribute name, so we need to use
vatter["raw_attr"], not 'a'.

This changes the behaviour for the ;rounds= option used when we hold
the plaintext password (possibly under GPG encryption).

This is now consistant with other parameters in the LDAP attribute,
and is now included in the returned attribute name.

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

selftest: Modify expected output of 'samba-tool user getpassword' to be more consistant

This is consistant with ;format= support for time attributes and
other users of this parameter style elsewhere in LDAP.

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

selftest: fix failing user setpassword test

A side effect of being able to generate at read time unicodePwd for a gMSA is that we can also generate the unicodePwd from a virtualSambaGPG password.

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

samba-tool: Add support for getting the generated unicodePwd for a gMSA account

This pre-hashed value may be more practical to use than the random "UTF-16"
password.  In particular it is easy to compare with the DB values.

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

netcmd: user: samba-tool support to allow non-windows use of GMSA accounts (show password)

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

samba-tool user getpassword: Use UTF16_MUNGED charcnv to map "UTF16" to UTF8

This copes with random invalid UTF-16 as seen with gMSA accounts.

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

samba-tool: Prepare to allow samba-tool user getpasswords to operate against a remote server

While passwords are not normally available for read, Group Managed Service Account
passwords are, as this is how they are distributed.

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

netcmd: models: add object sid field to User model

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

netcmd: getpassword: print OK message on stderr

This makes it easier to machine parse the output in tests

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

selftest: Avoid assertTrue() and assertFalse() where a better test exists

This allows the unittest framework to show the strings that the value was
not found in.

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

selftest: require named parameters for callers of connect_samdb() and connect_samdb_ex()

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

selftest: add get_env_credentials()

This is like get_credentials but works for tests that are based
on environment variable for usernames and passwords.

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

selftest: make get_loadparm a classmethod

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

selftest: make _get_attribute use parse_ldif

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

selftest: make _get_attribute a method on base class

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

selftest: pep8: fix incorrect number of blank lines

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

selftest: remove unused imports from virtualCryptSHA tests

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

selftest: function _get_attribute() was in two places

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

Revert "rpc_server:srvsvc - retrieve share ACL via root context"

This reverts commit 80c0b416892bfacc0d919fe032461748d7962f05.

With the previous patch it is no longer required. We open
share_info.tdb as root when starting up rpcd_classic and keep it open.

Commit 80c0b416892bfacc0d919fe032461748d7962f05 only fixed the
problem in one place, but we had it in a lot more places...

Bug: https://bugzilla.samba.org/show_bug.cgi?id=15265
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Wed Dec 20 11:20:51 UTC 2023 on atb-devel-224

rpcd_classic: Open share_info.tdb as root

srvsvc needs it, but for example NetShareGetInfo() runs as a
user. Opening share_info.tdb at that point is too late.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=15265
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

smbd: Modernize a few DEBUG statements

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Tue Dec 19 17:07:04 UTC 2023 on atb-devel-224

smbd: Remove an unused function prototype

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

smbd: Slightly simplify smbd_smb2_query_directory_send()

Avoid an if-statement when just assigning the result of the
if-condition to a boolean.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

smbd: Give source3/smbd/dir.c its own header file

The prototypes were spread across 3 different files.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

smbd: Remove unused srv_put_dos_date2()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

smbd: Use srv_put_dos_date2_ts() in reply_printqueue()

srv_put_dos_date2_ts() uses convert_timespec_to_time_t() on the passed
timespec, which for ts_nsec==0 returns just tv_sec.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

smbd: Introduce srv_put_dos_date2_ts()

All but one uses of srv_put_dos_date2() converted from struct
timespec. Put that into a new routine.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

profile: Fix rusage reporting

getrusage already gives accumulated values, so add them to the tdb
record after smbprofile_stats_accumulate. Maybe we should not zero out
our internal copy after writing the tdb, this would save a call to
smbprofile_stats_accumulate() once a second.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

profile: Fix a small memleak

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

smbd: Remove "conn" from struct dptr_struct

We have it available behind "dir_hnd".

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

lib: Remove duplicate prototypes

They also exist in lib/util/time.h

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

lib: Fix some whitespace

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>

ctdb-server: Drop unnecessary copy of destination address

Modernise debug while touching the code.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Dec 15 12:09:21 UTC 2023 on atb-devel-224

ctdb-daemon: Use ctdb_connection_to_buf() to simplify

The one case that is no longer handled specially is when the
destination address is IPv4 loopback.  This may previously have been
used to avoid flooding the logs when testing.  However, that seems
unnecessary - if testing with 127.0.0.1 then make it a public address.

Modernise debug while touching the code.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

smbd: Remove callback for release_ip when "state" is free'ed

If a client connects to a non-public address first followed by a connect
to public address with the same client_guid and a connection to
the non-public address gets disconnected first, we hit by a use-after-free
talloc_get_type_abort() called from release_ip() as
"xconn" is already gone, taking smbd_release_ip_state with it.

We need to decide between calling ctdbd_unregister_ips() by default, as
it means the tcp connection is really gone and ctdb needs to remove the
'tickle' information.  But when a connection was passed to a different
smbd process, we need to use ctdbd_passed_ips() as the tcp connection is
still alive and the 'tickle' information should not be removed within
ctdb.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

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

Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

s3:selftest: add samba3.blackbox.smbXsrv_client_ctdb_registered_ips

This demonstrates the crash that happens if a client connects to a
non-public address first followed by a connect
to public address with the same client_guid and a connection to
the non-public address gets disconnected first, we hit by a
use-after-free talloc_get_type_abort() called from release_ip() as
"xconn" is already gone, taking smbd_release_ip_state with it.

Note that we also need to mark some subtests as flapping
as there's a 2nd problem that happens in the interaction
between smbd processes and ctdb when passing a multichannel
connection to an existing process, it means we sometimes
loose the 'tickle' information within ctdb to that tcp connection.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

selftest: export/use CTDB related envvars in order to run the ctdb command

This makes it easier to test things...

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: add ctdbd_passed_ips()

This is similar to ctdbd_unregister_ips(), but with the
difference that ctdb keeps the 'tickle' information for
the tcp connection alive, because another smbd process
took care of that tcp connection in a multichannel scenario.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: add ctdbd_unregister_ips()

This reverts the effect of ctdbd_register_ips().
We'll use this in order to disconnect individual
multichannel connections.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: Add deregister_from_ctdbd()

This is to remove a callback during rundown of smbds.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

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

Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: let register_with_ctdbd() call CTDB_CONTROL_REGISTER_SRVID just once

We do the dispatching to multiple handlers in ctdbd_msg_call_back()
and we don't need more than one message from ctdb.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: don't use uninitialized memory in ctdbd_register_ips()

We dump the structure into the socket, so we need to zero the content
including possible padding.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb: add/implement CTDB_CONTROL_TCP_CLIENT_PASSED

With multichannel a tcp connection is registered first with
a temporary smbd process, that calls CTDB_CONTROL_TCP_CLIENT
first and then passes the tcp connection to the longterm smbd
that already handles all connections belonging to the specific
client_guid. That smbd process calls CTDB_CONTROL_TCP_CLIENT
again, but the 'tickle' information is already there.
When the temporary smbd process exists/disconnects from ctdb
or calls CTDB_CONTROL_TCP_CLIENT_DISCONNECTED, the 'tickle'
information is removed, while the longterm smbd process
still serves the tcp connection.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb: add/implement CTDB_CONTROL_TCP_CLIENT_DISCONNECTED

With multichannel a ctdb connection from smbd may hold multiple
tcp connections, which can be disconnected before the smbd
process terminates the whole ctdb connection, so we a
way to remove undo 'CTDB_CONTROL_TCP_CLIENT' again.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb: add ctdb_connection_same() helper

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb: make use of ctdb_canonicalize_ip_inplace() in ctdb_control_tcp_client()

We could also remove the src_addr and dest_addr helper variables
completely, but that would be too much for this commit.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb: add ctdb_canonicalize_ip_inplace() helper

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb: remove unused ctdb->client_ip_list and print debug on ctdb_tcp_list instead

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

VERSION: move COPYRIGHT_STARTUP_MESSAGE as SAMBA_COPYRIGHT_STRING into version.h

We also prodive a samba_copyright_string() helper similar to
samba_version_string().

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15377

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Björn Jacke <bjacke@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Dec 15 10:44:42 UTC 2023 on atb-devel-224