vfs_ceph_new: refactor error-case in cephmount_mount_fs

Align code-style of 'cephmount_mount_fs' with rest of the code: use
'goto' for bail-out upon error case (with proper cleanups). For the
common case of successful operation complete execution and return final
value. Added extra debug-logging for good-path case.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Shachar Sharon <ssharon@redhat.com>
Reviewed-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: John Mulligan <jmulligan@redhat.com>
(cherry picked from commit d9b872afeee4dee49de2e6eb86e4b59e07804363)

vfs_ceph_new: switch to ceph_readdir_r

Prefer a safe version of ceph_readdir, where the directory entry struct
is allocated by the caller. Use a dynamic-allocated 'struct dirent'
which is associated with a directory vfs_ceph_fh (optional), which is
allocated on-the-fly upon start of READDIR and released at the end or
CLOSEDIR (or unlikely readdir error).

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Shachar Sharon <ssharon@redhat.com>
Reviewed-by: John Mulligan <jmulligan@redhat.com>
Reviewed-by: Anoop C S <anoopcs@samba.org>
Autobuild-User(master): Anoop C S <anoopcs@samba.org>
Autobuild-Date(master): Fri Oct 25 10:29:44 UTC 2024 on atb-devel-224

(cherry picked from commit ce459ddbcd0f32252858a7640f6871057eb14645)

docs_xml/vfs_ceph_new: Add new proxy option

Update man page to describe new 'proxy' module option.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Anoop C S <anoopcs@samba.org>
Autobuild-Date(master): Thu Sep 12 16:51:14 UTC 2024 on atb-devel-224

(cherry picked from commit 12ad4832a74a6fba3fb8954a8630b900f5763f18)

wscript_build: Do not link vfs_ceph_new against libcephfs

vfs_ceph_new dynamically loads the appropriate libcephfs shared
libraries which means that we don't statically link against it.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 095ece43a2edbdeb498e8bef07bdf09f9836bb40)

vfs_ceph_new: Use function pointers for API calls

Replace direct function calls with pointers holding their equivalent
addresses.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 962a40a6ff51f33adc39e466f1479dfb35ac0926)

vfs_ceph_new: Pass module config to userperm helpers

userperm helpers will switch to function references instead of direct
invocation of APIs. This would mean the matching config structure is
passed to those helpers.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit d5926cf492b1cb81c90b27e0537ec3f1b1a10f80)

vfs_ceph_new: Hold a config reference in vfs_ceph_fh

This is required to perform the cleanup when fsp extension destructor is
called as part of VFS_REMOVE_FSP_EXTENSION where mount information and
function references are to be used in upcoming changes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 250af54250bed87928250b95db663ff13cc1f42a)

vfs_ceph_new: Call vfs_ceph_userperm_new with handle->conn

vfs_ceph_userperm_new() only need connection structure from handle
for fetching the current unix token. Therefore modify the signature
to accept just handle->conn.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 8c1d774c7e206bb413284d500c5aa950ac3dd3d5)

vfs_ceph_new: Populate function pointers with addresses

Use dlysm() for assigning addresses to already declared libcephfs
low-level API function pointers.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Pair-Programmed-With: Anoop C S <anoopcs@samba.org>
Signed-off-by: Guenther Deschner <gd@samba.org>
Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit e4fc1df4b4486f5eed095135f905bdbf10bb1793)

vfs_ceph_new: Add required function pointers to config

Declare necessary libcephfs low-level APIs as function pointers to be
assigned with corresponding loadable addresses.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Pair-Programmed-With: Anoop C S <anoopcs@samba.org>
Signed-off-by: Guenther Deschner <gd@samba.org>
Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 0d2ad13d8e5191f4b01584afa0bcdbf1114042b8)

vfs_ceph_new: Dynamically open library for 'proxy' mode

Use dlopen() to load either of the shared libraries(libcephfs.so or
libcephfs_proxy.so) based on the configuration for 'proxy' module
parameter. Further down the line we will define the required APIs
as function pointers within the config structure.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Pair-Programmed-With: Anoop C S <anoopcs@samba.org>
Signed-off-by: Guenther Deschner <gd@samba.org>
Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 47812a279118befbaeffdd6c81e3d49b071f04c5)

vfs_ceph_new: Introduce new parametric option 'proxy'

Provide early support for consuming yet to come libcephfs proxy[1] for
optimized resource utilization. For better control we make use of an
additional module specific option 'proxy' to specify the intent to load
proxy library. With the default value 'no' a regular cephfs connection
is established. There is also an 'auto' mode which can fall back to the
regular connection if proxy requirements are not met.

[1] https://github.com/ceph/ceph/pull/58376

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 90464bdcafda0f0f0e4d2b549fd1675c076ee188)

vfs_ceph_new: Add a new struct to hold ceph module config

Consolidate all required configuration related data under a dedicated
structure named vfs_ceph_config. As of now it includes the location of
configuration file, file system name, ceph client user id and mount
related information. This is expected to grow in future with more
details as and when required. Apart from that small cleanups are also
done to make code more robust.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Pair-Programmed-With: Anoop C S <anoopcs@samba.org>
Signed-off-by: Guenther Deschner <gd@samba.org>
Signed-off-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit f1d418181d12bd513204cf2b77c37990939e2041)

vfs_ceph_new: implement DFS hooks using libcephfs low-level APIs

Refactor the VFS hooks 'create_dfs_pathat_fn' and 'read_dfs_pathat_fn'
in 'vfs_ceph_new.c' to use libcephfs low-level APIs: instead of using
path-based operations (as done in legacy 'vfs_ceph.c') use inode based
operations to create and read msdfs symbolic-links values.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15703

Signed-off-by: Shachar Sharon <ssharon@redhat.com>
Reviewed-by: Anoop C S <anoopcs@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Wed Sep 11 19:09:41 UTC 2024 on atb-devel-224

(cherry picked from commit 0cedd74e47ab919528420761a5bd2acb198f084c)

s3-libnet: avoid using lp_dns_hostname() in join code

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15777

This codepath is also used for provisiong non-local machines into AD
during offlinejoin operations. When creating accounts for non-local
machines we certainly need to be able to use arbitrary hostname other
than lp_netbios_name() (which is used internally by lp_dns_hostname()).
This partly reverts 0e96092c1895ecb41d4064111566b4ada71fe457.

Guenther

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Jan 30 07:35:05 UTC 2025 on atb-devel-224

(cherry picked from commit 062dc07e9b9c8e260548d0bca4d02819bdc60326)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Wed Feb 12 10:37:56 UTC 2025 on atb-devel-224

selfest: add test for non-local offlinejoin provision

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15777

Guenther

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 6d4ad4d6824e81ef85dd924d550222dd6a322a15)

s3-libads: dump ADS_MODSLIST before attempting the LDAP modify

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15777

Guenther

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit f02a4002d5c3cfcd7f36b3bcf13310ffd155de90)

sharesec: Check if share exists in configuration

Load config from registry without share info and check if sharename
exists from configuration. This results into lesser delay for the same.

In case of view we load config with all shares to ensure we get all
shares for diplay purpose.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15780

Signed-off-by: Vinit Agnihotri <vagnihot@redhat.com>
Reviewed-by: John Mulligan <jmulligan@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Fri Jan 10 10:45:30 UTC 2025 on atb-devel-224

(cherry picked from commit 1410803713440caf29a40aec30516489d1944665)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Wed Feb  5 10:43:01 UTC 2025 on atb-devel-224

sharesec: Add function to check existence of share from config

Add function to detect if a share name exists in the registry or config file.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15780

Signed-off-by: Vinit Agnihotri <vagnihot@redhat.com>
Reviewed-by: John Mulligan <jmulligan@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 78eb293e1cdd3635de0bcf46ffb9d842f27bcc9f)

param: Add API to load registry without share info

As number of shares increases loading entire registry configuration along with
share information becomes very costly operation.
Since we may not require share information all time, we can optimise
this by using API just loading configuration without any share info.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15780

Signed-off-by: Vinit Agnihotri <vagnihot@redhat.com>
Reviewed-by: John Mulligan <jmulligan@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 2927dba0434b960e4c381329bdc9fe474fb930ce)

sharesec: Fix warning frame not freed in order

This change should fix following warning:
Freed frame ../../source3/utils/sharesec.c:515, expected ../../source3/utils/sharesec.c:637

Frame was not getting freed in case of servicename is NULL.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15780

Signed-off-by: Vinit Agnihotri <vagnihot@redhat.com>
Reviewed-by: John Mulligan <jmulligan@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 04531e1b1d25d114c470922547bee769b07e5e60)

s3-sharesec: Add Test to verify command option "--view-all"

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15780

Signed-off-by: Guenther Deschner <gd@samba.org>
Signed-off-by: Vinit Agnihotri <vagnihot@redhat.com>
Reviewed-by: John Mulligan <jmulligan@samba.org>
(cherry picked from commit 0a12254ea8b3414deebc3e6329025052c650356e)

s4:rpc_server/lsa: let LookupSids* behave like Windows 2022/2025

The important part is the INVALID_SID should not
cause an early exit of the loop.

We need to return the intact names array with the
correct count. And only return INVALID_SID
if we would otherwise return NONE_MAPPED.

For SOME_NOT_MAPPED we need to ignore invalid sids
and just pretend they are not mapped.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 218a0f067c894cbf61cde6183a269c0474d64ddc)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Mon Feb  3 20:01:38 UTC 2025 on atb-devel-224

libcli/security: let dom_sid_lookup_predefined_sid() behave like Windows 2008R2

Windows 2008R2 (172.31.9.133) returns the following:

 #> rpcclient 172.31.9.133 -Uadministrator%A1b2C3d4 -c 'lookupsids S-1-22-1 S-1-22-1-0;lookupsids S-1-22;lookupsids S-1-3-0 S-1-3-99;lookupsids S-1-3'
 S-1-22-1 *unknown*\*unknown* (8)
 S-1-22-1-0 *unknown*\*unknown* (8)
 result was NT_STATUS_INVALID_SID
 S-1-3-0 \CREATOR OWNER (5)
 S-1-3-99 *unknown*\*unknown* (8)
 result was NT_STATUS_INVALID_SID

While the current Samba (172.31.9.163) returns the following:

 #> rpcclient 172.31.9.163 -Uadministrator%A1b2C3d4 -c 'lookupsids S-1-22-1 S-1-22-1-0;lookupsids S-1-22;lookupsids S-1-3-0 S-1-3-99;lookupsids S-1-3'
 result was NT_STATUS_INVALID_SID
 result was NT_STATUS_INVALID_SID
 S-1-3-0 \CREATOR OWNER (5)
 S-1-3-99 *unknown*\*unknown* (8)
 S-1-3 *unknown*\*unknown* (8)

With this change also return the same as Windows 2008R2:

 #> rpcclient 172.31.9.163 -Uadministrator%A1b2C3d4 -c 'lookupsids S-1-22-1 S-1-22-1-0;lookupsids S-1-22;lookupsids S-1-3-0 S-1-3-99;lookupsids S-1-3'
 S-1-22-1 *unknown*\*unknown* (8)
 S-1-22-1-0 *unknown*\*unknown* (8)
 result was NT_STATUS_INVALID_SID
 S-1-3-0 \CREATOR OWNER (5)
 S-1-3-99 *unknown*\*unknown* (8)
 result was NT_STATUS_INVALID_SID

This is a minimal fix in order to avoid crashes in the Windows Explorer.
The real fix needs more work and additional tests, as the behavior seems
to be different in newer Windows releases.

The following patch will let us behave like Windows 2022/2025...

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 9f369c62317d74615834f99a088caababef685fc)

python:tests/dcerpc/lsa: add tests for invalid LookupSids2 combinations

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit abba3495d23524142a4cf55d208dce041adee96b)

s4:pyrpc: allow connections with raise_result_exceptions=False

This is needed in order to do useful tests with
specific error codes and still checking all other
out parameters.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 6b1ff9a38fcddbe72b00e28960414526a42bde14)

pidl:Python: prepare code to avoid NTSTATUS/WERROR exceptions

They are returned as additional result.

It means callers can look at all out params,
even if the status is an error.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 7ecaf1a779370ef3ecf189e51a5e668329fa24c7)

pidl:Python: handle NTSTATUS/WERROR exceptions first

If we raise an exception we should not leak temporary
python objects.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 67c35d6b4ef6b7dbe9db3c52547b25580cd4756c)

pidl:Python: separate logic to calculate the signature string

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 13d9231800ea969675f3207cd2c863e433104b4d)

pidl:Python: check PyTuple_New() return value

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit f7282c0dffbf30d72051a81d46d831344a9bbcf9)

pidl:Python: initialize pointers and add 'result' at the end

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 9dfb0ed8d29bd4a9146cf38bd63c4bb55b5faf73)

pidl:Python: introduce $is_raisable_return helper variable

No change in the generated code.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 8cdf7af43a365b0545562033f6c51150f2fbb3a4)

pidl:Python: generate nicer code for PyNdrRpcMethodDef arrays

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14213

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit fa6d0fd1b00e4836215b4d80f1a9f527db82e01a)

s3:auth: let check_sam_security() add NETLOGON_NTLMV2_ENABLED

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15783

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 10f38aff7c23dc20bc98cf0e02a430f8b0d7b1ad)

s4:auth/ntlm: let authsam_check_password_internals() add NETLOGON_NTLMV2_ENABLED

Windows returns NETLOGON_NTLMV2_ENABLED in all
netr_LogonSamLogon* response messages.
Even if NTLMv1 was actually used and also
for password authentication.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15783

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 1414004ee953975c50e0ee374684ff8e01246946)

auth: Cleanup exit code paths in kerberos_decode_pac().

One more memory leak missed and now fixed. tmp_ctx
must be freed once the pac data is talloc_move'd.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15782

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Jennifer Sutton <jennifersutton@catalyst.net.nz>
Reviewed-by: Christian Ambach <ambi@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Fri Jan 17 12:01:47 UTC 2025 on atb-devel-224

(cherry picked from commit f9eb0b248da0689c82656f3e482161c45749afb6)

auth: Add missing talloc_free() in error code path.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15782

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Thu Jan 16 14:32:39 UTC 2025 on atb-devel-224

(cherry picked from commit c514ce8dcadcbbf0d86f3038d2be0f9253a76b75)

lib:replace: Don't use deprecated readline CPPFunction cast

HAVE_RL_COMPLETION_FUNC_T was unused and not checking for the right
function.

libcli/smbreadline/smbreadline.c: In function ‘smb_readline’:
libcli/smbreadline/smbreadline.c:139:17: warning: ‘CPPFunction’ is deprecated [-Wdeprecated-declarations]
  139 |                 rl_attempted_completion_function = RL_COMPLETION_CAST completion_fn;
      |                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
libcli/smbreadline/smbreadline.c:139:50: error: assignment to ‘char ** (*)(const char *, int,  int)’ from incompatible pointer type ‘char ** (*)(void)’ [-Wincompatible-pointer-types]
  139 |                 rl_attempted_completion_function = RL_COMPLETION_CAST completion_fn;
      |                                                  ^

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15788

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Tue Jan 21 19:38:37 UTC 2025 on atb-devel-224

(cherry picked from commit 9aa5c43315d83c19514251a11c4fba5a137f2821)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Thu Jan 23 15:15:06 UTC 2025 on atb-devel-224

lib:replace: Remove trailing spaces from readline.h

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15788

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
(cherry picked from commit dbff53f77680f0ccd022f2d864ba6ae7585a34b7)

lib:util: Fix stack-use-after-return in crypt_as_best_we_can()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15784

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Pavel Filipenský <pfilipensky@samba.org>
Autobuild-User(master): Douglas Bagnall <dbagnall@samba.org>
Autobuild-Date(master): Fri Jan 17 23:21:13 UTC 2025 on atb-devel-224

(cherry picked from commit 6cd9849b58ec653cbffc602e3c96996a082faf53)

util:datablob: data_blob_pad checks its alignment assumption

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Fri Dec 20 07:59:51 UTC 2024 on atb-devel-224

(cherry picked from commit 8b84282008dc372d67ba01c8fe256ef756c3dcfb)

pytest: password_hash uses internal _glue.crypt

This will remove an external dependency.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 552053b6445611ecef6ac4c11c55ebf92f03571d)

samba-tool user: hashlib.sha1 is always present

We maybe thought we were checking that sha1 was in hashlib, but we were
only checking that hashlib is in the Python library (`hashlib.sha1()`
would not raise ImportError).

The documentation says hashlib always contains sha1 -- if that
changes, it is better we know by failing noisily with the import error
at the top of the file.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 4af4dd8135e8edbe2a16cfdfc7ded8c145c82e98)

samba-tool user: use _glue.crypt, not crypt.crypt

Because we know we have _glue.crypt, and we know it raises exceptions
rather than returning None, we can simplify the checks.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 405187d2ef4920a9a284649c9c3287f5844d5180)

pytest: test that _glue.crypt works

The test vectors were generated via Python 3.10 crypt module, which
directly wraps crypt(3), which in this case is from glibc 2.39-0ubuntu8.3.

We mainly test the sha256 and sha512 vectors, which seems to be all we
use, and which are said to be widely supported.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 5636d30c0959fd4a211ee7b8d1b267dcdbf0b963)

pyglue: add crypt() function

This wraps talloc_crypt_blob() from lib/util/util_crypt.c which in
turn wraps the system crypt[_r[n]].

We want this because the Python standard library crypt module is going
away. That one also wrapped the system crypt or crypt_r, so there
should be no change.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(backported from commit 88e3c82d88a68cf972f8189e1c3718698b49974a)

util: add a crypt strerror helper

This will be used by Python also.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 5f365e71c1fa8cdc533159283a5977164b5d39f2)

dsdb:password_hash: use talloc_crypt_blob()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit c7597380b479208e33a403211cec9b3c7bd3f034)

dsdb:password_hash: move hash_blob allocation up

This will make the next patch simpler.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 1edb12f79593d0b2aac36d5acdaaae6f495772f6)

util: add a crypt wrapper, derived from dsdb:password_hash

This is going to be used by the dsdb password_hash module, and exposed
to Python via pyglue.

We're doing this because Python 3.13 has dropped crypt from the Python
standard library.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15756

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
(backported from commit 833455c7f9f71583d567e3a53e854567cd8c3b0b,
Signed-off-by added)

s4:dsdb: fix logic of dsdb_trust_routing_by_name()

We need to use the longest dnsname match as possible.

If we are the domain samba.example.com and have a trust
to example.com, a routing request for dc.samba.example.com
should return the tdo for samba.example.com instead
of example.com.

I reproduced the problem with the following diff:

> diff --git a/selftest/target/Samba.pm b/selftest/target/Samba.pm
> index 15d7692b5d64..6e9595b784c4 100644
> --- a/selftest/target/Samba.pm
> +++ b/selftest/target/Samba.pm
> @@ -564,7 +564,7 @@ sub realm_to_ip_mappings
>   'samba2000.example.com'           => 'dc5',
>   'samba2003.example.com'           => 'dc6',
>   'samba2008r2.example.com'         => 'dc7',
> - 'addom.samba.example.com'         => 'addc',
> + 'addom.samba2008r2.example.com'         => 'addc',
>   'addom2.samba.example.com'        => 'addcsmb1',
>   'sub.samba.example.com'           => 'localsubdc',
>   'chgdcpassword.samba.example.com' => 'chgdcpass',
> diff --git a/selftest/target/Samba4.pm b/selftest/target/Samba4.pm
> index 0e4cf50235c3..6bca0cfd0c89 100755
> --- a/selftest/target/Samba4.pm
> +++ b/selftest/target/Samba4.pm
> @@ -2631,7 +2631,7 @@ sub setup_fl2008r2dc
>       return undef;
>   }
>
> - $env = $self->setup_trust($env, $ad_dc_vars, "forest", "");
> + $env = $self->setup_trust($env, $ad_dc_vars, "forest", "--skip-validation");
>   if (!defined $env) {
>       return undef;
>   }
> @@ -2843,7 +2843,7 @@ sub _setup_ad_dc
>   $server = "addc";
>   }
>   if (!defined($dom)) {
> - $dom = "addom.samba.example.com";
> + $dom = "addom.samba2008r2.example.com";
>   }
>   my $env = $self->provision_ad_dc($path, $server, "ADDOMAIN",
>    $dom,

and running:
 make -j testenv SELFTEST_TESTENV="fl2008r2dc:local"

Inside the testenv:
bin/smbclient //addc.addom.samba2008r2.example.com/netlogon \
  -U$TRUST_USERNAME@$TRUST_REALM%$TRUST_PASSWORD \
  --use-kerberos=required \
  -c 'ls'

It lets the KDC of ADDOM.SAMBA2008R2.EXAMPLE.COM to
generate a (referral) ticket for
krbtgt/SAMBA2008R2.EXAMPLE.COM@ADDOM.SAMBA2008R2.EXAMPLE.COM
instead of
cifs/addc.addom.samba2008r2.example.com@ADDOM.SAMBA2008R2.EXAMPLE.COM

As ADDOM.SAMBA2008R2.EXAMPLE.COM has a forest trust (without msDS-TrustForestTrustInfo)
to SAMBA2008R2.EXAMPLE.COM dsdb_trust_update_best_tln() overwrote the
best match of addom.samba2008r2.example.com with samba2008r2.example.com.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15778

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Autobuild-User(master): Douglas Bagnall <dbagnall@samba.org>
Autobuild-Date(master): Wed Jan  8 04:14:47 UTC 2025 on atb-devel-224

(cherry picked from commit 56947612452c32bd26f30ad6c9767830fe608d67)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Thu Jan  9 11:16:34 UTC 2025 on atb-devel-224

s4:scripting: fix gen_hresult.py

Commit 6877e4849e81ed8f7f355fa4069f13a134e999ce reversed the
return values of hresult_errstr() and hresult_errstr_const().

hresult_errstr() should return "HRES_SEC_E_WRONG_PRINCIPAL",
while hresult_errstr_const() should return
"The target principal name is incorrect.".

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15769

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
(cherry picked from commit 6e0e9c4efc86f5ec4566aa0220ed4a74ad5436c1)

VERSION: Bump version up to Samba 4.21.4...

and re-enable GIT_SNAPSHOT.

Signed-off-by: Jule Anger <janger@samba.org>

VERSION: Disable GIT_SNAPSHOT for the 4.21.3 release.

Signed-off-by: Jule Anger <janger@samba.org>

WHATSNEW: Add release notes for Samba 4.21.3.

Signed-off-by: Jule Anger <janger@samba.org>

docs:manpages: Update 'net ads keytab create'

BUG: https://bugzilla.samba.org/show_bug.cgi?id=6750

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Pavel Filipensky <pfilipensky@samba.org>
Autobuild-Date(master): Mon Dec 16 19:32:32 UTC 2024 on atb-devel-224

(cherry picked from commit 7b73c574d93668edd94f2eb18b58568d420487f4)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Tue Dec 31 15:31:52 UTC 2024 on atb-devel-224

pam_winbind: Fix Bug 15771

Bug: https://bugzilla.samba.org/show_bug.cgi?id=15771

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit 7e8bfe738a3d91b0724d57ed54acf40ebe026c14)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Thu Dec 19 10:13:37 UTC 2024 on atb-devel-224

s4:drs:test:getncchanges skips some tests with reserved_usn = 0

These tests are not affected by the reserved_usn change, so there is
no need to run them twice.

The test_repl_get_tgt_multivalued_links fails with or without
reserved_usn set to zero, but it fails differently in either case.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 7dac035896b368bf3a86acf58260eef39d195d19)

s4:drs:test:getncchanges: remove timeout failure

We don't need a timeout failure any more, since replication should
always work. Leaving the timeout in might sometimes cause a flapping
test if replication is being slow for some reason.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 44a478038b6ec78aaec832d9dbde7fa6b2cdd639)

s4:drsuapi:getncchanges: allow 0 reserved_usn reply

Azure AD will set reserved_usn to zero when we expect it to be
the number we gave them.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 7a623d8d5626b4e6c88ffb85e36f0934d89ed830)

s4:drsuapi:getncchanges: use DBG_ERR() macro

The next commit will indent this more, so it's a bit squished up.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 5ef27019033fd73decd111f9426e7f8982cbb806)

s4:drsuapi:getncchanges: fix whitespace

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 2e1ccb35239fc6fe129c943bb7305bd4612d72d7)

s4:drs:tests: repeat getncchanges test with zero reserved_usn

This emulates the behaviour of Azure AD.

As this is quite slow we will later reduce the test load in this case,
but for now we want to run all the getncchanges tests this way.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 67c7609ab755291de27c620120a1c71b557452e4)

s4:drs:tests: add hook for changing highwatermark

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 796e92a530004406dcb3fea33f54833c722480a0)

s4:drs:test:getncchanges: add a timeout failure

In the next commit we are going to add tests in which the client
modifies the highwatermark in a way that resets replication (on Samba
only). After that we'll fix it.

If we leave the test in an eternal loop, the commit history will not
be bisectable, so we are temporarily going to turn long waits into
failures.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15701

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Jennifer Sutton <josutton@catalyst.net.nz>
(cherry picked from commit 4b4a7c3fd465267c43d9586ab79ca8f84c0cad24)

selftest: Add test for vfs crossrename module

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15724

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
(cherry picked from commit 02d4f58a2f7ac2db60dd2e4d16a3cbf71b3f08a9)

docs:manpage: vfs_crossrename is not fully stackable VFS module

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15724

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
(cherry picked from commit 94c9a99c56db438c391a966c927ec2f862c373e7)

s3:vfs_crossrename: add back checking for errno ENOENT

strace gives a clue: samba try to remove 'file.txt' in the dst folder but
actually it is not existed yet, and got an errno = ENOENT,

renameat(32, "file.txt", 31, "file.txt") = -1 EXDEV (Invalid cross-device link)
unlinkat(31, "file.txt", 0)             = -1 ENOENT (No such file or directory)

Commit 5c18f074be92 ("s3: VFS: crossrename. Use real dirfsp for
SMB_VFS_RENAMEAT()") seems unintentionally removed errno ENOENT checking,
so add it back could address 1st issue.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15724

Signed-off-by: Jones Syue <jonessyue@qnap.com>
Reviewed-by: Ralph Boehme <slow@samba.org>
(cherry picked from commit 1a089a16c40e0b3bc5d4fcde559157cf137056c2)

s3:vfs_crossrename: crossrename_renameat() needs to return 0 if copy_reg() is successful

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15724

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
(cherry picked from commit 0a9adc85e77bc557bb8be12237fa31c4142dd3d5)

s3:vfs_crossrename: avoid locking panic in copy_reg()

Use low level backend functions that don't go through the FSA layer.
Done via calling transfer_file() as it was in version before 5c18f07

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15724

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
(cherry picked from commit 0a5da82f75a43838be3419cab10a50750fa500d7)

s4:rpc_server: make use of dcesrv_assoc_group_common_destructor()

Currently this should not be needed, but it's better to
call dcesrv_assoc_group_common_destructor() in all assoc_group
destructors.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15765

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 19657be71d7cec5ac58a5d6969dc1d6ae7c5b517)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Mon Dec 16 17:36:29 UTC 2024 on atb-devel-224

s3:rpc_server: make use of dcesrv_assoc_group_common_destructor()

We need to detach dcesrv_iface_state from dcesrv_assoc_group,
if dcesrv_assoc_group is free'ed first.

Typically this doesn't happen, but it does when
rpc_worker_connection_terminated explicitly calls
talloc_unlink(conn, conn->assoc_group)
and dcesrv_iface_state_store_conn() is used.

But we better do it in all assoc_group destructors.

==381007==ERROR: AddressSanitizer: heap-use-after-free on address 0x50d000004f80 at pc 0x7f15fc12e0ac bp 0x7ffe43267780 sp 0x7ffe43267778
READ of size 8 at 0x50d000004f80 thread T0
    #0 0x7f15fc12e0ab in dcesrv_iface_state_destructor ../../librpc/rpc/dcesrv_handles.c:166
    #1 0x7f15fc0f7d76 in _tc_free_internal ../../lib/talloc/talloc.c:1158
    #2 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #3 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #4 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #5 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #6 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #7 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #8 0x7f15fc0f924c in _talloc_free_internal ../../lib/talloc/talloc.c:1248
    #9 0x7f15fc0f924c in _talloc_free ../../lib/talloc/talloc.c:1792
    #10 0x7f15fadac024 in ncacn_terminate_connection ../../source3/rpc_server/rpc_server.c:263
    #11 0x7f15fadac024 in dcesrv_transport_terminate_connection ../../source3/rpc_server/rpc_server.c:251
    #12 0x7f15fc11e5ef in dcesrv_terminate_connection ../../librpc/rpc/dcesrv_core.c:2968
    #13 0x7f15fc125446 in dcesrv_read_fragment_done ../../librpc/rpc/dcesrv_core.c:3196
    #14 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #15 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #16 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #17 0x7f15fb4f69a1 in _tevent_req_nterror ../../lib/util/tevent_ntstatus.c:46
    #18 0x7f15fabda2f4 in dcerpc_read_ncacn_packet_done ../../librpc/rpc/dcerpc_util.c:612
    #19 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #20 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #21 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #22 0x7f15fbff4228 in tstream_readv_pdu_readv_done ../../lib/tsocket/tsocket_helpers.c:313
    #23 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #24 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #25 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #26 0x7f15fbff1800 in tstream_readv_done ../../lib/tsocket/tsocket.c:593
    #27 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #28 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #29 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #30 0x7f15fadbc1a3 in tstream_npa_readv_msg_mode_handler ../../libcli/named_pipe_auth/npa_tstream.c:697
    #31 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #32 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #33 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #34 0x7f15fbff4228 in tstream_readv_pdu_readv_done ../../lib/tsocket/tsocket_helpers.c:313
    #35 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #36 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #37 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #38 0x7f15fbff1800 in tstream_readv_done ../../lib/tsocket/tsocket.c:593
    #39 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #40 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #41 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #42 0x7f15fbff9691 in tstream_bsd_readv_handler ../../lib/tsocket/tsocket_bsd.c:2080
    #43 0x7f15fbff6f85 in tstream_bsd_fde_handler ../../lib/tsocket/tsocket_bsd.c:1764
    #44 0x7f15fb7d9ac1 in tevent_common_invoke_fd_handler ../../lib/tevent/tevent_fd.c:174
    #45 0x7f15fb7ef185 in epoll_event_loop ../../lib/tevent/tevent_epoll.c:696
    #46 0x7f15fb7ef185 in epoll_event_loop_once ../../lib/tevent/tevent_epoll.c:926
    #47 0x7f15fb7e77b8 in std_event_loop_once ../../lib/tevent/tevent_standard.c:110
    #48 0x7f15fb7d7549 in _tevent_loop_once ../../lib/tevent/tevent.c:820
    #49 0x7f15fc936b7c in rpc_worker_main ../../source3/rpc_server/rpc_worker.c:1249
    #50 0x5632ae1e1ec3 in main ../../source3/rpc_server/rpcd_lsad.c:132
    #51 0x7f15f7c2a2ad in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
    #52 0x7f15f7c2a378 in __libc_start_main_impl ../csu/libc-start.c:360
    #53 0x5632ae162e64 in _start ../sysdeps/x86_64/start.S:115

0x50d000004f80 is located 112 bytes inside of 136-byte region [0x50d000004f10,0x50d000004f98)
freed by thread T0 here:
    #0 0x7f15fcefb418 in free ../../../../libsanitizer/asan/asan_malloc_linux.cpp:52
    #1 0x7f15fc0f857d in _tc_free_internal ../../lib/talloc/talloc.c:1222
    #2 0x7f15fc0f8d0f in _talloc_free_internal ../../lib/talloc/talloc.c:1248
    #3 0x7f15fc0f8d0f in talloc_unlink ../../lib/talloc/talloc.c:1473
    #4 0x7f15fc934580 in rpc_worker_connection_terminated ../../source3/rpc_server/rpc_worker.c:143
    #5 0x7f15fc9310bd in dcesrv_connection_destructor ../../source3/rpc_server/rpc_worker.c:175
    #6 0x7f15fc0f7d76 in _tc_free_internal ../../lib/talloc/talloc.c:1158
    #7 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #8 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #9 0x7f15fc0f924c in _talloc_free_internal ../../lib/talloc/talloc.c:1248
    #10 0x7f15fc0f924c in _talloc_free ../../lib/talloc/talloc.c:1792
    #11 0x7f15fadac024 in ncacn_terminate_connection ../../source3/rpc_server/rpc_server.c:263
    #12 0x7f15fadac024 in dcesrv_transport_terminate_connection ../../source3/rpc_server/rpc_server.c:251
    #13 0x7f15fc11e5ef in dcesrv_terminate_connection ../../librpc/rpc/dcesrv_core.c:2968
    #14 0x7f15fc125446 in dcesrv_read_fragment_done ../../librpc/rpc/dcesrv_core.c:3196
    #15 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #16 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #17 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #18 0x7f15fb4f69a1 in _tevent_req_nterror ../../lib/util/tevent_ntstatus.c:46
    #19 0x7f15fabda2f4 in dcerpc_read_ncacn_packet_done ../../librpc/rpc/dcerpc_util.c:612
    #20 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #21 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #22 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #23 0x7f15fbff4228 in tstream_readv_pdu_readv_done ../../lib/tsocket/tsocket_helpers.c:313
    #24 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #25 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #26 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #27 0x7f15fbff1800 in tstream_readv_done ../../lib/tsocket/tsocket.c:593
    #28 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #29 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #30 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #31 0x7f15fadbc1a3 in tstream_npa_readv_msg_mode_handler ../../libcli/named_pipe_auth/npa_tstream.c:697
    #32 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #33 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234

previously allocated by thread T0 here:
    #0 0x7f15fcefc777 in malloc ../../../../libsanitizer/asan/asan_malloc_linux.cpp:69
    #1 0x7f15fc0fbc57 in __talloc_with_prefix ../../lib/talloc/talloc.c:783
    #2 0x7f15fc0fd8cf in __talloc ../../lib/talloc/talloc.c:825
    #3 0x7f15fc0fd8cf in _talloc_named_const ../../lib/talloc/talloc.c:982
    #4 0x7f15fc0fd8cf in _talloc_zero ../../lib/talloc/talloc.c:2421
    #5 0x7f15fc93156e in rpc_worker_assoc_group_new ../../source3/rpc_server/rpc_worker.c:681
    #6 0x7f15fc93156e in rpc_worker_assoc_group_find ../../source3/rpc_server/rpc_worker.c:730
    #7 0x7f15fc120a18 in dcesrv_bind ../../librpc/rpc/dcesrv_core.c:1158
    #8 0x7f15fc120a18 in dcesrv_process_ncacn_packet ../../librpc/rpc/dcesrv_core.c:2324
    #9 0x7f15fc120a18 in dcesrv_loop_next_packet ../../librpc/rpc/dcesrv_core.c:3222
    #10 0x7f15fc933722 in rpc_worker_new_client ../../source3/rpc_server/rpc_worker.c:489
    #11 0x7f15fc933722 in rpc_worker_new_client_filter ../../source3/rpc_server/rpc_worker.c:558
    #12 0x7f15fbef95ca in messaging_dispatch_waiters ../../source3/lib/messages.c:1343
    #13 0x7f15fbefb589 in messaging_dispatch_rec ../../source3/lib/messages.c:1371
    #14 0x7f15fbefb589 in messaging_recv_cb ../../source3/lib/messages.c:431
    #15 0x7f15faddba9e in msg_dgm_ref_recv ../../lib/messaging/messages_dgm_ref.c:144
    #16 0x7f15fadd6cc3 in messaging_dgm_recv ../../lib/messaging/messages_dgm.c:1426
    #17 0x7f15fadd7618 in messaging_dgm_read_handler ../../lib/messaging/messages_dgm.c:1316
    #18 0x7f15fb7d9ac1 in tevent_common_invoke_fd_handler ../../lib/tevent/tevent_fd.c:174
    #19 0x7f15fb7ef185 in epoll_event_loop ../../lib/tevent/tevent_epoll.c:696
    #20 0x7f15fb7ef185 in epoll_event_loop_once ../../lib/tevent/tevent_epoll.c:926
    #21 0x7f15fb7e77b8 in std_event_loop_once ../../lib/tevent/tevent_standard.c:110
    #22 0x7f15fb7d7549 in _tevent_loop_once ../../lib/tevent/tevent.c:820
    #23 0x7f15fc936b7c in rpc_worker_main ../../source3/rpc_server/rpc_worker.c:1249
    #24 0x5632ae1e1ec3 in main ../../source3/rpc_server/rpcd_lsad.c:132
    #25 0x7f15f7c2a2ad in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15765

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 627a7857844804a29c6612df5da4605c94edb3f9)

dcesrv_core: add dcesrv_assoc_group_common_destructor()

We need to detach dcesrv_iface_state from dcesrv_assoc_group,
if dcesrv_assoc_group is free'ed first.

==381007==ERROR: AddressSanitizer: heap-use-after-free on address 0x50d000004f80 at pc 0x7f15fc12e0ac bp 0x7ffe43267780 sp 0x7ffe43267778
READ of size 8 at 0x50d000004f80 thread T0
    #0 0x7f15fc12e0ab in dcesrv_iface_state_destructor ../../librpc/rpc/dcesrv_handles.c:166
    #1 0x7f15fc0f7d76 in _tc_free_internal ../../lib/talloc/talloc.c:1158
    #2 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #3 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #4 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #5 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #6 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #7 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #8 0x7f15fc0f924c in _talloc_free_internal ../../lib/talloc/talloc.c:1248
    #9 0x7f15fc0f924c in _talloc_free ../../lib/talloc/talloc.c:1792
    #10 0x7f15fadac024 in ncacn_terminate_connection ../../source3/rpc_server/rpc_server.c:263
    #11 0x7f15fadac024 in dcesrv_transport_terminate_connection ../../source3/rpc_server/rpc_server.c:251
    #12 0x7f15fc11e5ef in dcesrv_terminate_connection ../../librpc/rpc/dcesrv_core.c:2968
    #13 0x7f15fc125446 in dcesrv_read_fragment_done ../../librpc/rpc/dcesrv_core.c:3196
    #14 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #15 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #16 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #17 0x7f15fb4f69a1 in _tevent_req_nterror ../../lib/util/tevent_ntstatus.c:46
    #18 0x7f15fabda2f4 in dcerpc_read_ncacn_packet_done ../../librpc/rpc/dcerpc_util.c:612
    #19 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #20 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #21 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #22 0x7f15fbff4228 in tstream_readv_pdu_readv_done ../../lib/tsocket/tsocket_helpers.c:313
    #23 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #24 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #25 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #26 0x7f15fbff1800 in tstream_readv_done ../../lib/tsocket/tsocket.c:593
    #27 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #28 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #29 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #30 0x7f15fadbc1a3 in tstream_npa_readv_msg_mode_handler ../../libcli/named_pipe_auth/npa_tstream.c:697
    #31 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #32 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #33 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #34 0x7f15fbff4228 in tstream_readv_pdu_readv_done ../../lib/tsocket/tsocket_helpers.c:313
    #35 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #36 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #37 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #38 0x7f15fbff1800 in tstream_readv_done ../../lib/tsocket/tsocket.c:593
    #39 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #40 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #41 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #42 0x7f15fbff9691 in tstream_bsd_readv_handler ../../lib/tsocket/tsocket_bsd.c:2080
    #43 0x7f15fbff6f85 in tstream_bsd_fde_handler ../../lib/tsocket/tsocket_bsd.c:1764
    #44 0x7f15fb7d9ac1 in tevent_common_invoke_fd_handler ../../lib/tevent/tevent_fd.c:174
    #45 0x7f15fb7ef185 in epoll_event_loop ../../lib/tevent/tevent_epoll.c:696
    #46 0x7f15fb7ef185 in epoll_event_loop_once ../../lib/tevent/tevent_epoll.c:926
    #47 0x7f15fb7e77b8 in std_event_loop_once ../../lib/tevent/tevent_standard.c:110
    #48 0x7f15fb7d7549 in _tevent_loop_once ../../lib/tevent/tevent.c:820
    #49 0x7f15fc936b7c in rpc_worker_main ../../source3/rpc_server/rpc_worker.c:1249
    #50 0x5632ae1e1ec3 in main ../../source3/rpc_server/rpcd_lsad.c:132
    #51 0x7f15f7c2a2ad in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
    #52 0x7f15f7c2a378 in __libc_start_main_impl ../csu/libc-start.c:360
    #53 0x5632ae162e64 in _start ../sysdeps/x86_64/start.S:115

0x50d000004f80 is located 112 bytes inside of 136-byte region [0x50d000004f10,0x50d000004f98)
freed by thread T0 here:
    #0 0x7f15fcefb418 in free ../../../../libsanitizer/asan/asan_malloc_linux.cpp:52
    #1 0x7f15fc0f857d in _tc_free_internal ../../lib/talloc/talloc.c:1222
    #2 0x7f15fc0f8d0f in _talloc_free_internal ../../lib/talloc/talloc.c:1248
    #3 0x7f15fc0f8d0f in talloc_unlink ../../lib/talloc/talloc.c:1473
    #4 0x7f15fc934580 in rpc_worker_connection_terminated ../../source3/rpc_server/rpc_worker.c:143
    #5 0x7f15fc9310bd in dcesrv_connection_destructor ../../source3/rpc_server/rpc_worker.c:175
    #6 0x7f15fc0f7d76 in _tc_free_internal ../../lib/talloc/talloc.c:1158
    #7 0x7f15fc0f7acd in _tc_free_children_internal ../../lib/talloc/talloc.c:1669
    #8 0x7f15fc0f7acd in _tc_free_internal ../../lib/talloc/talloc.c:1184
    #9 0x7f15fc0f924c in _talloc_free_internal ../../lib/talloc/talloc.c:1248
    #10 0x7f15fc0f924c in _talloc_free ../../lib/talloc/talloc.c:1792
    #11 0x7f15fadac024 in ncacn_terminate_connection ../../source3/rpc_server/rpc_server.c:263
    #12 0x7f15fadac024 in dcesrv_transport_terminate_connection ../../source3/rpc_server/rpc_server.c:251
    #13 0x7f15fc11e5ef in dcesrv_terminate_connection ../../librpc/rpc/dcesrv_core.c:2968
    #14 0x7f15fc125446 in dcesrv_read_fragment_done ../../librpc/rpc/dcesrv_core.c:3196
    #15 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #16 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #17 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #18 0x7f15fb4f69a1 in _tevent_req_nterror ../../lib/util/tevent_ntstatus.c:46
    #19 0x7f15fabda2f4 in dcerpc_read_ncacn_packet_done ../../librpc/rpc/dcerpc_util.c:612
    #20 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #21 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #22 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #23 0x7f15fbff4228 in tstream_readv_pdu_readv_done ../../lib/tsocket/tsocket_helpers.c:313
    #24 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #25 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #26 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #27 0x7f15fbff1800 in tstream_readv_done ../../lib/tsocket/tsocket.c:593
    #28 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #29 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234
    #30 0x7f15fb7dcdb7 in _tevent_req_error ../../lib/tevent/tevent_req.c:252
    #31 0x7f15fadbc1a3 in tstream_npa_readv_msg_mode_handler ../../libcli/named_pipe_auth/npa_tstream.c:697
    #32 0x7f15fb7dcae5 in _tevent_req_notify_callback ../../lib/tevent/tevent_req.c:177
    #33 0x7f15fb7dcd1c in tevent_req_finish ../../lib/tevent/tevent_req.c:234

previously allocated by thread T0 here:
    #0 0x7f15fcefc777 in malloc ../../../../libsanitizer/asan/asan_malloc_linux.cpp:69
    #1 0x7f15fc0fbc57 in __talloc_with_prefix ../../lib/talloc/talloc.c:783
    #2 0x7f15fc0fd8cf in __talloc ../../lib/talloc/talloc.c:825
    #3 0x7f15fc0fd8cf in _talloc_named_const ../../lib/talloc/talloc.c:982
    #4 0x7f15fc0fd8cf in _talloc_zero ../../lib/talloc/talloc.c:2421
    #5 0x7f15fc93156e in rpc_worker_assoc_group_new ../../source3/rpc_server/rpc_worker.c:681
    #6 0x7f15fc93156e in rpc_worker_assoc_group_find ../../source3/rpc_server/rpc_worker.c:730
    #7 0x7f15fc120a18 in dcesrv_bind ../../librpc/rpc/dcesrv_core.c:1158
    #8 0x7f15fc120a18 in dcesrv_process_ncacn_packet ../../librpc/rpc/dcesrv_core.c:2324
    #9 0x7f15fc120a18 in dcesrv_loop_next_packet ../../librpc/rpc/dcesrv_core.c:3222
    #10 0x7f15fc933722 in rpc_worker_new_client ../../source3/rpc_server/rpc_worker.c:489
    #11 0x7f15fc933722 in rpc_worker_new_client_filter ../../source3/rpc_server/rpc_worker.c:558
    #12 0x7f15fbef95ca in messaging_dispatch_waiters ../../source3/lib/messages.c:1343
    #13 0x7f15fbefb589 in messaging_dispatch_rec ../../source3/lib/messages.c:1371
    #14 0x7f15fbefb589 in messaging_recv_cb ../../source3/lib/messages.c:431
    #15 0x7f15faddba9e in msg_dgm_ref_recv ../../lib/messaging/messages_dgm_ref.c:144
    #16 0x7f15fadd6cc3 in messaging_dgm_recv ../../lib/messaging/messages_dgm.c:1426
    #17 0x7f15fadd7618 in messaging_dgm_read_handler ../../lib/messaging/messages_dgm.c:1316
    #18 0x7f15fb7d9ac1 in tevent_common_invoke_fd_handler ../../lib/tevent/tevent_fd.c:174
    #19 0x7f15fb7ef185 in epoll_event_loop ../../lib/tevent/tevent_epoll.c:696
    #20 0x7f15fb7ef185 in epoll_event_loop_once ../../lib/tevent/tevent_epoll.c:926
    #21 0x7f15fb7e77b8 in std_event_loop_once ../../lib/tevent/tevent_standard.c:110
    #22 0x7f15fb7d7549 in _tevent_loop_once ../../lib/tevent/tevent.c:820
    #23 0x7f15fc936b7c in rpc_worker_main ../../source3/rpc_server/rpc_worker.c:1249
    #24 0x5632ae1e1ec3 in main ../../source3/rpc_server/rpcd_lsad.c:132
    #25 0x7f15f7c2a2ad in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15765

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 5b929860e269e2968a0ec3759a6125ae990b43c3)

smbd: fix breaking leases on rename

We must also break leases on other opens if the open of the rename doesn't have
a lease itself. The existing test test_lease_v2_rename() that was added
alongside the deferred rename server code didn't cover this case.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15697

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): Thu Oct 31 12:47:24 UTC 2024 on atb-devel-224

(cherry picked from commit efbbe8d6f80ceb6107f20486623eee949409c0ff)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Mon Dec  9 16:21:19 UTC 2024 on atb-devel-224

smbd: force sync rename with lease break

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15697

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit bc2d87981967bc65155ba09eb5b3e3f913bec50e)

smbd: return correct error for compound related requests that went async

For a compound related request chain of eg CREATE+NOTIFY+GETINFO, the NOTIFY
will typically go async. When this is noted in smbd_smb2_request_pending_queue()
the pending async tevent_req is cancelled which means we return
NT_STATUS_CANCELLED to the client while Windows returns
NT_STATUS_INTERNAL_ERROR.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15697

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit a5635791cfdb10f64bf2bf7c72c58f7591249a0d)

smbtorture: test rename with other opens on the file

Windows allows this. Samba also already implements this correctly.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15697

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 3890ac2fafc5e17919fa39542440a05ef72a3fa5)

smbtorture: add a bunch of tests for async rename and async interim responses

All tests pass against Windows 2022, we have some bugs.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15697

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 42e739ab62cb573d72215737027cf3c7f1fcd212)

smbtorture: rename CHECK_VALUE() to CHECK_VAL() in smb2/compound.c

Prepares for using macros from lease_break_handler.h which makes use of
CHECK_VAL() while relying on a definition of CHECK_VAL() in the .c file.

While at it, add a goto done which is always a good thing to get clear failures
from tests.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15697

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 4932b433ff2f1c4e603073624a5d22140acfb2ed)

vfs_btrfs: Also call vfs_offload_token_ctx_init() in btrfs_offload_write_send()

If a client for whatever reason calls FSCTL_SRV_COPYCHUNK[_WRITE] without
FSCTL_SRV_REQUEST_RESUME_KEY, we call btrfs_offload_write_send
before btrfs_offload_read_send.

This is similar to 462b74da79c51f9ba6dbd24e603aa904485d5123

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15758

Signed-off-by: Andreas Schneider <asn@samba.org>
Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Tue Dec  3 18:04:26 UTC 2024 on atb-devel-224

ctdb-common: Map ENOENT for a missing event script to ENOEXEC

This handles the case where an event script is disabled by unlinking,
while an event is being run, after the script list has been created.
Without this change the script will fail.  With this change the script
will be marked as DISABLED.  See the comment added by this commit for
more details.

Add a testcase to simulate the race, using an event script to disable
subsequent ones.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15755
RN: Avoid event failure race when disabling an event script

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Thu Nov 21 01:42:09 UTC 2024 on atb-devel-224

(cherry picked from commit fee31b6cb2b8f7dd111bdd9d2ff5479c31cbca37)

VERSION: Bump version up to Samba 4.21.3...

and re-enable GIT_SNAPSHOT.

Signed-off-by: Jule Anger <janger@samba.org>

VERSION: Disable GIT_SNAPSHOT for the 4.21.2 release.

Signed-off-by: Jule Anger <janger@samba.org>

WHATSNEW: Add release notes for Samba 4.21.2.

Signed-off-by: Jule Anger <janger@samba.org>

ctdb-scripts: Track connections for all ports for public IPs

Currently TCP ports like NFS lock manager are not tracked.  It is
easier to track all connections than to add a configuration system to
try to track specified ports, so do that.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit 590a86dbe4adf45ac8d15497934e25ea98148034)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Mon Nov 25 14:09:18 UTC 2024 on atb-devel-224

ctdb-scripts: Get connections after tickle list

Running ss to get current connections before running ctdb gettickles
means the ss output might be out of date when the 2 lists are
compared.  Some tickles might have been added after ss was run by some
other means (e.g. SMB tickles, added internally) and they would be
deleted according to the stale ss output.

This isn't currently a problem because update_tickles() is currently
only called with port 2049, so all tickles are managed by this code.
That will change in a subsequent commit.

Changing the order means the reverse problem can occur, where
update_tickles() attempts to delete an already deleted tickle.  That
may happen occasionally but is harmless because it doesn't result in
missing information.  It (currently) just causes a message to be
logged at DEBUG level.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit c3695722b6316b624aa6c44cad4f44279303d1b1)

ctdb-scripts: Move connection tracking to 10.interface

This should really be done for all connections to public IP addresses.
Leave the port number there for now - this is just the first step.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit 9683bb3ac2bbdf0e83c3be3681f9d1c8ee7cc327)

ctdb-server: Drop a log message to DEBUG level

This is harmless, so it doesn't generally need to be logged.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit f4a8f84328c5e692ce63bec05bb71fcb469a3e9c)

ctdb-server: Clean up connection tracking functions

Apply README.Coding, modernise logging, pre-render connection as a
string for logging, switch terminology from "tickle" to "connection",
tidy up comments.

No changes in functionality.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit 3c19c8df778070705485b3c993e695ca1636bfa7)

ctdb-scripts: Use ss -H option to simplify

This option has been available since ~2018 and has been implemented in
the stub since then.  I guess we didn't use it because CentOS 7?

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit 0505d06b12a04a5c5e813fb3f4799278f9e5b7eb)

ctdb-scripts: Remove superseded compatibility code

Since commit 224e99804efef960ef4ce2ff2f4f6dced1e74146, square brackets
have been parsed by daemon and tool code, so drop the compatibility
code from here.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit 32e4f786601712e57992ce4c8f46e5d38620a5dd)

ctdb-scripts: update_tickles() should use the public IPs cache

This avoids duplicating logic.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit b3e2c69ad92c0d20bb10146d2dd6d0d475455298)

ctdb-scripts: Don't list connections when not hosting IPs

With an empty IP filter, all incoming connections to port 2049 will be
listed, not just those to public IP addresses.  This causes error
messages like the following to be logged:

  ctdb-eventd[...]: 60.nfs: Failed to add 1 tickles

since the connection being added seems to be for a random NFS mount
that doesn't use a public IP addresses.

This has been a problem for a long time (probably since commit
04fe9e20749985c71fef1bce7f6e4c439fe11c81 in 2015).  It isn't currently
a huge deal because it only affects NFS connections.  However, this
code will soon be used to track connections to public IP addresses on
all ports.  This would result in a constant stream of log messages,
since there will always be some active connections.

The theory behind the fix is that if a node hosts no public IPs then
it should have no relevant connections and has no business changing
the list of registered tickles.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15320
RN: Update CTDB to track all TCP connections to public IP addresses

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jerry Heyman <jheyman@ddn.com>
(cherry picked from commit 1a4a6c46f1cdabfea67c264d6576a597a70c3007)

smbd: avoid a panic in close_directory()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15754

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Mon Nov 18 12:43:28 UTC 2024 on atb-devel-224

(backported from commit f86d1a35917cdcc07b2fa741c6a369aaad24abaf)
[slow@samba.org: context changes from Directory Leases]

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Wed Nov 20 09:55:08 UTC 2024 on atb-devel-224

examples:winexe: Initialize Trustee.ptstrName at the right time

This is a regression of a39cb60 examples:winexe: Fully initialize EXPLICIT_ACCESS

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15752

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Pavel Filipensky <pfilipensky@samba.org>
Autobuild-Date(master): Wed Nov 13 15:47:40 UTC 2024 on atb-devel-224

(cherry picked from commit 50f3273e8283c9dabeca3c899b90ad831657b12a)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Thu Nov 14 14:06:52 UTC 2024 on atb-devel-224

libcli/auth: make use of netlogon_creds_cli_check_transport() in more places

This was somehow missing in commit
7a5ad9f64a905f5744430c6e0796c646baf9432e

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15425

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Nov  7 09:14:33 UTC 2024 on atb-devel-224

(cherry picked from commit f340dce6546a22d857cad440f8afaee9815dbdb1)

Autobuild-User(v4-21-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-21-test): Wed Nov 13 10:05:15 UTC 2024 on atb-devel-224

libcli/auth: split out netlogon_creds_cli_check_transport()

This will make it easier to implement netr_ServerAuthenticateKerberos()
later...

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15425

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 7a5ad9f64a905f5744430c6e0796c646baf9432e)

libcli/auth: let netlogon_creds_copy() copy all scalar elements

This version is good for now, as we want it to be backportable.
For master we'll add a ndr_deepcopy_struct() helper in order
to avoid future problems.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15425

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 8edbdd65ef78e3f26357d0254b58db3120a32880)

s4:librpc/rpc: make use of netlogon_creds_client_verify()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15425

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 132629ee3a9b73d0888d1110e4d0a45ded778e5a)

libcli/auth: make use of netlogon_creds_client_verify()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15425

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
(cherry picked from commit 1a5984ac6312b204b51590057b8327cf4698383b)