clifuse: Start implementing forget()

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

clifuse: Use an empty array for holding the inode path

Cleaner these days.

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

examples: Slightly modernize printfs in teststat

long long works, but it's a bit too specific now that we have intmax_t

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

libsmb: Remove two #defines just used once

To me these two did not create clarity.

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

smbd: Modernize a few DBG statements

Changes on level from 4 to 3 in store_file_unix_basic().

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

examples: Use explicit SMBCCTX

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

lib: Avoid a tdb handle leak

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

lib: Modernize tdb_fetch_lifetime()

Just came across this, avoid mallocs.

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

libsmb: Remove unused cli_is_dos_error()

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

libsmb: Remove unused cli_dos_error()

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

libsmb: Eliminate a reader of cli->raw_status

At some point we should get rid of cli->raw_status.

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

libsmb: Remove SMBC_errno()

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

libsmb: Remove a call to SMBC_errno()

This involves converting cli_printjob_del() to NTSTATUS and thus
touches a few callers.

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

libsmb: Remove a call to SMBC_errno()

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

libsmb: Remove a call to SMBC_errno()

All returns from cacl_get() now explicitly set the errno.

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

libsmb: Remove unused cli_is_nt_error()

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

smbclient: Save lines with talloc_asprintf_addbuf()

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

libsmb: Extend cli_mknod to create NFS reparse points

Tested smbclient mkfifo manually against W2k22, the Windows NFS server
reports this as FIFO.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon Sep 25 18:08:01 UTC 2023 on atb-devel-224

libsmb: Remove reparse_symlink.c

Makes reparse_buffer_marshall static to reparse.c

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

libsmb: Use reparse_data_buffer_marshall() in py_reparse_put()

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

libsmb: Use reparse_data_buffer_marshall() in cli_symlink_send()

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

libsmb: Factor out cli_create_reparse_point() from cli_symlink()

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

libsmb: Use reparse_data_buffer_marshall() in py_reparse_symlink_put()

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

libsmb: Add reparse_data_buffer_marshall()

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

smbclient: Add mkfifo command

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

libsmb: Add sync cli_mknod() for smbclient3's use

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

s3:tests: Check if test_smbclient_kerberos.sh was successful

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Pavel Filipensky <pfilipensky@samba.org>
Autobuild-Date(master): Mon Sep 25 00:14:47 UTC 2023 on atb-devel-224

s3:tests: Fix syntax error in test_smbclient_kerberos.sh

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:tests: Fix init of samba_kdestroy in test_smbclient_kerberos.sh

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:tests: test_smbspool.sh should be calling smbspool_argv_wrapper

There is no variable named 'smbspool_argv_wrapper', the correct name is
'samba_argv_wrapper'. The smbspool_argv_wrapper was not called from this
test till now.

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Pavel Filipensky <pfilipensky@samba.org>
Autobuild-Date(master): Sun Sep 24 22:33:07 UTC 2023 on atb-devel-224

s3:tests: Fix smbspool_argv_wrapper.c

smbspool_argv_wrapper must copy also the last argument

Signed-off-by: Pavel Filipenský <pfilipensky@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

conf: Remove "smb3 unix extensions" parameter

Always offer it, it's a client thing to ask for it or not.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Thu Sep 21 17:43:23 UTC 2023 on atb-devel-224

ctdb-daemon: Call setproctitle_init()

Commit 19c82c19c009eefe975ae95c8b709fc93f5f4c39 changed the behaviour
of prctl_set_comment() so it now calls setproctitle(3bsd) by default.

In some Linux distributions (e.g. Rocky Linux 8.8), this results in
messages like this spamming the logs:

  ctdbd: setproctitle not initialized, please either call setproctitle_init() or link against libbsd-ctor.

Most Samba daemons seem to call setproctitle_init(), so do it here.

In the longer term CTDB should also switch to using lib/util's
process_set_title(), like the rest of Samba, for more flexible process
names.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15479

Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Thu Sep 21 00:46:50 UTC 2023 on atb-devel-224

smbd: Fix BZ15481

Bug: https://bugzilla.samba.org/show_bug.cgi?id=15481

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Sep 20 22:42:48 UTC 2023 on atb-devel-224

tests: Add reproducer for BZ15481

Bug: https://bugzilla.samba.org/show_bug.cgi?id=15481

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

s3: smbd: Ensure we remove any pending aio values for named pipes on forced shutdown.

Matches file and directory closes.

Remove knownfail.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15423

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Sep 20 02:43:18 UTC 2023 on atb-devel-224

s3: torture: Add a new SMB2 test: SMB2-PIPE-READ-ASYNC-DISCONNECT

Shows the server crashes if we open a named pipe, do an async read
and then disconnect.

Adds knownfail:

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15423

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: named pipe writes are async. Use the same logic as for named pipe transacts to avoid crashes on shutdown.

Noticed by Metze.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15423

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: named pipe reads are async. Use the same logic as for named pipe transacts to avoid crashes on shutdown.

Noticed by Metze.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15423

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Add some DEVELOPER-only code to panic if the destructor for an aio_lnk is called and the associated fsp doesn't exist.

Make this DEVELOPER-only as it walks the entire open
file list on every file close (with associated aio).

This helps catch really subtle problems with orphaned
aio lnk structs.

Reproducer test case to follow.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15423

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Now we have proved hardlink_internals() doesn't use src_dirfsp and dst_dirfsp, remove the parameters.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Tue Sep 19 19:51:47 UTC 2023 on atb-devel-224

s3: smbd: hardlink_internals() never looks at src_dirfsp or dst_dirfsp.

Show this by sending NULL in all cases.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Now we have shown dst_dirfsp is always NULL, remove the parameter from rename_internals().

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: As rename_internals() calls rename_internals_fsp(), show we can pass dst_dirfsp as NULL here too.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Now we've proved dst_dirfsp parameter is always NULL, remove the parameter from rename_internals_fsp().

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: rename_internals_fsp() has to reopen the parent directory of the target as a pathref to check permissions.

So it never looks at any passed in dst_dirfsp.

Prove this by passing NULL.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: add some basic testing for the io_uring vfs module

We're now able to build it on all linux systems and
the ci runners have at least a 5.4 kernel. That's
all the current vfs_io_uring requires.

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): Sun Sep 17 18:04:18 UTC 2023 on atb-devel-224

s3:param: Remove unused lp_set_cmdline()

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Thu Sep 14 22:30:06 UTC 2023 on atb-devel-224

s3:client: Use lpcfg_set_cmdline()

This lp_set_cmdline() leaks memory and we want to get rid of it.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:libsmb: Use lpcfg_set_cmdline()

Signed-off-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:netapi: Fix a leak in libnetapi_net_init()

Allow libnetapi_net_init() to be called more than once without
leaking libnetapi_ctx allocated on a previous call, which is
currently the case in the `net rpc` code.

Signed-off-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

libnetapi: Use lpcfg_set_cmdline()

Replace lp_set_cmdline() with lpcfg_set_cmdline() in netapi.c.
For this purpose:
1. Add loadparm_context to the libnetapi_ctx because we need it
   in libnetapi_set_debuglevel() and libnetapi_set_logfile().
2. Move loadparm_context creation from libnetapi_net_init()
   to libnetapi_init() and add the lp_ctx parameter to the former.

Signed-off-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:utils: Use lpcfg_set_cmdline() in smbpasswd

Signed-off-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:torture: Use lpcfg_set_cmdline()

Signed-off-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

s3:notifyd: Use lpcfg_set_cmdline()

Signed-off-by: Pavel Kalugin <pkalugin@inno.tech>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

build: Reduce heimdal_no_error_flags to a more minimal set

Heimdal does not seem to give as many errors as in the past, so we
can reduce the set of errors we ignore warnings for.

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

build: Remove -Wno-error=missing-field-initializers from Heimdal build

This allows this warning (error) to be used in Samba for cases where a
non-designated initialiser is under-specified.  We can do this now as
the GCC versions we build on do not regard foo = {} as an error.

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

s4:dsdb: Fix comment

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

s3:winbindd: Add zero digit to literal

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

dsgetdcname: Remove excess zero digits from literals

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

fsrvp.idl: Remove excess zero digit from literal

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

tests/krb5: Rename ‘client_claims’ to ‘claims_metadata’

This variable is used not only for client claims, but for device claims
as well.

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

s3:net: Check return value of data_blob_talloc()

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

s4:kdc: Don’t prepend useless colon to MIT KDC logging messages

If we pass an empty string as the ‘whoami’ parameter, MIT’s logging
facilities will prepend a mysterious colon to the message. Printing
“mitkdc: ” ought at least to be more sensible, and perhaps more closely
to match our behaviour prior to commit
dd8138236bec3635c25e5b482b7a14faa0a9c36b.

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

s4:wrepl_server: Correctly read ‘type’ element

winsdb_message() stores this element as hexadecimal, which format
ldb_msg_find_attr_as_uint() cannot cope with. Permit this element to be
in either decimal or hexadecimal format.

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

ldb: Prefer explicit initialization to ZERO_STRUCT()

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

s4:kdc: Remove ks_is_tgs_principal()

This function is a near‐duplicate of smb_krb5_principal_is_tgs().

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

testprogs: Fix script usage lines

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

s4:setup: Fix script usage line

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

s4:selftest: Fix script usage lines

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

s3:script: Fix script usage lines

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

nsswitch: Fix script usage

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

lib:krb5_wrap: Remove Heimdal‐only krb5_princ_component() implementation

This implementation is no longer called: using a variable of static
storage duration as a conduit for return values is only asking for
trouble.

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

s4:rpc_server: Switch to using smb_krb5_princ_component()

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

s4:kdc: Switch to using smb_krb5_princ_component()

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

s4:dsdb: Switch to using smb_krb5_princ_component()

This function has the handy feature of being able to be called twice in
succession without mysteriously breaking your code. Now, doesn’t that
sound useful?

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

lib:krb5_wrap: Add Heimdal‐specific smb_krb5_princ_component() implementation

This implementation doesn’t rely on a variable of static storage
duration being used as a conduit for the return value.

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

lib:krb5_wrap: Add smb_krb5_princ_component()

For now this function is a mere wrapper round krb5_princ_component(),
but one whose interface allows for a more sensible implementation.

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

lib:krb5_wrap: Have smb_krb5_principal_get_realm() check the return values of intermediate functions

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

lib:krb5_wrap: Eliminate redundant code from smb_krb5_sockaddr_to_kaddr()

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

s4:kdc: Fix leaks

Create a temporary memory context on which to allocate things.

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

s4:kdc: Be sure not to pass a NULL pointer into strcmp()

To do so is to invoke undefined behaviour.

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

s4:kdc: Prefer explicit initialization to ZERO_STRUCTP()

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

s4:kdc: Make RODC ID checks easier to understand with more clearly‐named variables

No change in behaviour.

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

s4:kdc: Rename ‘status’ variables to ‘reply_status’

This makes it clearer what these variables are used for, and avoids
confusion with the similarly‐named ‘nt_status’ variables — also used in
these functions.

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

s4:kdc: Have samba_krbtgt_is_in_db() take a const KDC entry

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

s4:kdc: Have samba_krbtgt_is_in_db() return a krb5_error_code

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

s4:kdc: Make pac_blobs_remove_blob() never fail

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

s4:kdc: Make ‘struct pac_blobs’ memory handling safer and more consistent

Having pac_blobs::type_blobs be managed both by talloc and by the
‘pac_blobs’ structure itself (with pac_blobs_destroy()) is very prone to
error. So is the current situation of having the other ‘pac_blobs’
functions each take in a memory context.

Improve these circumstances by requiring ‘pac_blobs’ to be managed by
talloc. Now the other functions can dispense with their ‘mem_ctx’
parameters, being instead able to allocate on to the ‘pac_blobs’
structure itself. pac_blobs_init() no longer must be a separate
function; inline it into pac_blobs_from_krb5_pac(). pac_blobs_destroy(),
being no longer of use, can go too.

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

s4:kdc: Don’t corrupt pac_blobs structure if talloc_realloc() fails

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

s4:kdc: Add common out path to pac_blobs_from_krb5_pac()

This helps to make error‐checking and cleanup more systematic.

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

s4:kdc: Introduce a temporary talloc context in samba_kdc_update_pac()

This avoids allocating working structures on to a potentially long‐lived
context.

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

s4:kdc: Use temporary memory context in samba_kdc_verify_pac()

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

s4:kdc: Fix leak

Introduce a temporary memory context and allocate working structures on
to it.

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

s4:kdc: Allocate variables on to more suitable memory context

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

s4:kdc: Increment PAC_DEVICE_INFO::domain_group_count only after SID has been successfully added

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

s4:kdc: Directly zero‐initialize PAC_DOMAIN_GROUP_MEMBERSHIP structure

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

s4:kdc: Assign RID and attribute together

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

s4:kdc: Don’t corrupt domain groups structure if talloc_realloc() fails

Introduce a temporary variable instead of assigning the result of
talloc_realloc() directly to samr_RidWithAttributeArray::rids. In this
way we avoid having a structure with a non‐zero ‘count’ but with ‘rids’
set to the NULL pointer.

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

s4:kdc: Check for overflow when adding a domain group SID

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