selftest/manage-ca: Add certificiate etc. for DC addcsmb1

Preparatory commit pre adding new addc_smb1 environment.

In order to support new addc_smb1 test environment we need new
certificate(s) in order to satisfy some tests
(e.g. samba4.blackbox.pkinit.*)

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Ralph Boehme <slow@samba.org>

script: Add testenv fileserver_smb1 to autobuild jobs

Adjust the existing autobuild jobs that currently include/exclude
test env fileserver to also use fileserver_smb1.

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: Add new test environment fileserver_smb1

fileserver_smb1 is a copy of the fileserver test environment but
with the difference that it still supports SMB1, this will allow
use to still run SMB1 tests.

Signed-off-by: Noel Power <npower@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: adjust skip_file (after adjusting s4member and rpc_proxy)

The orig skip file was created with *all* test environments by
default not support SMB1.

Signed-off-by: Noel Power <npower@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: Allow rpc_proxy & s4member also negotiate SMB1

Test environments rpc_proxy & s4member depend on ad_dc_ntvfs
so these environments additionally need to also be able to
still support SMB1.

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: disable SMB1 from testenv(s) make >=SMB2 the default

Change the values of 'min protocol' set for the various test
environments to be SMB2_02.

Servers will only offer protocols starting with the min specified in the
conf files. We don't change the client value here yet (until SMB1 is
truely gone) as we still want to be able to run SMB1 tests.

Signed-off-by: Noel Power <npower@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: Exclude SMB1 tests that would fail when SMB1 is disabled

Exclude the tests specified in the file 'skip_smb1_fails' which
contains the list of tests that would fail if min protocol >= SMB2_02

Note: a subsequent commit will change the default min protocol so
      in order to maintain bisectability we skip the tests before
      we change the default.

Signed-off-by: Noel Power <npower@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: Create a file to exclude tests that fail with SMB1

skip_smb1_fails contains tests that will fail when the min protocol
of the test environments is set to SMB2_02

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Ralph Boehme <slow@samba.org>

selftest: Allow ad_dc_ntvfs negotiate smb1 still

The default for most test envs will be changed in a future
commit(s) to be >= SMB2 then we need to still allow  ad_dc_ntvfs to
support SMB1. This will make the number of tests to port easier to
deal with. In addition to test env ad_dc_ntvfs still supporting SMB1
we need to do the same with test other environments that depend on
ad_dc_ntvfs (this will be handled in followup commits).

In addition to the above this change will ensure we don't trigger
failures for ntvfs tests when we switch to default >= SMB2 which will
make the associated skip file (added in a future commit) smaller.

Signed-off-by: Noel Power <npower@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3/selftest: Fix typo in test generator script

Introduced in a64516099936 but only makes a difference when
environments limit protocols for testing.

Reviewed-by: Ralph Boehme <slow@samba.org>
Signed-off-by: Noel Power <noel.power@suse.com>

selftest: allow SMB1 in clusteredmember env

This is a relatively new test environment that only uses very few tests
yet. Skip the env name rename dance for this env, but someone has to convert the
tests later on.

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Noel Power <npower@samba.org>

samba-tool group show: only shows global security groups, this patch makes it show all groups.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14335

Signed-off-by: Rowland Penny <rpenny@samba.org>
Reviewed-by: Björn Baumbach <bb@samba.org>
Autobuild-User(master): Björn Baumbach <bb@sernet.de>
Autobuild-Date(master): Thu Apr  2 15:27:53 UTC 2020 on sn-devel-184

dbwrap: fix possible memleak and false result check.

A cstatus != 0 or a data.dsize != sizeof(uint32_t)
does not guarantee to have no received data referenced by data.dptr.
Therefore, make sure data.dptr is free'd.
Reusing the same data structure as data input and data output parameter
can lead to wrong results, especially when the output parameters value
is used to detect errors. Create the additional local variable outdata
to prevent this issue.

Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Thu Apr  2 11:26:32 UTC 2020 on sn-devel-184

dbwrap: fix comment in code leading to wrong function parameter

Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: possible memleak in ctdbd_db_attach

A cstatus != 0 or a data.dsize != sizeof(uint32_t)
does not guarantee to have no received data referenced by data.dptr.
Therefore, make sure data.dptr is free'd.

Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: possible memleak in ctdbd_dbpath

A cstatus != 0 does not guarantee to have no received data
referenced by rdata.dptr.
Therefore, make sure rdata.dptr is free'd.

Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdbd_conn: possible memleak in ctdbd_working

A cstatus != 0 does not guarantee to have no received data
referenced by outdata.dptr.
Therefore, make sure outdata.dptr is free'd.

Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

script/autobuild.py: allow write_system_info commands to fail

These commands are just there as hints to debug possible problems.
In order to support autobuild.py on non-linux platforms we should
just ignore errors here.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Apr  2 07:36:07 UTC 2020 on sn-devel-184

script/autobuild.py: use more portable 'cp -R -a -l'

On platforms like FreeBSD 12 cp doesn't accept the long options,
using the one letter options works there and keeps working on Linux
as well.

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

script/autobuild.py: add support git worktree

.git is not always a directory, with 'git worktree' it's a file.

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

.gitlab-ci.yml: actually run ubuntu2004-samba-o3

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

s3/utils: Fix double free error with smbtree

==3632==
WORKGROUP
\\ATP
\\ATP\IPC$            IPC Service ()
==3632== Invalid read of size 8
==3632==    at 0x773C926: poptResetContext (in /usr/lib64/libpopt.so.0.0.0)
==3632==    by 0x773E5DD: poptFreeContext (in /usr/lib64/libpopt.so.0.0.0)
==3632==    by 0x10A8BC: main (smbtree.c:354)
==3632==  Address 0x16085e00 is 640 bytes inside a block of size 784 free'd
==3632==    at 0x4C2F1AD: free (vg_replace_malloc.c:530)
==3632==    by 0x773E6F7: poptFreeContext (in /usr/lib64/libpopt.so.0.0.0)
==3632==    by 0x10A84B: main (smbtree.c:342)
==3632==  Block was alloc'd at
==3632==    at 0x4C2FE45: calloc (vg_replace_malloc.c:711)
==3632==    by 0x773C79A: poptGetContext (in /usr/lib64/libpopt.so.0.0.0)
==3632==    by 0x10A829: main (smbtree.c:339)
==3632==

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14332
Signed-off-by: Noel Power <npower@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Noel Power <npower@samba.org>
Autobuild-Date(master): Wed Apr  1 18:51:43 UTC 2020 on sn-devel-184

s3:rpc_server: Fix talloc_free() with references error on server exit

Fixes talloc_free() with references when smbd exists and talloc_free the
global dcesrv context:

ERROR: talloc_free with references at ../../source3/rpc_server/rpc_config.c:73
reference at ../../source3/rpc_server/rpc_server.c:193
reference at ../../source3/rpc_server/rpc_server.c:592
...

The talloc_reference to make the dcerpc_ncacn_listen_state a parent of
the dcesrv context is not necessary as the listen state is allocated
under the endpoint, which is allocated under the dcesrv context:

full talloc report on 'struct dcesrv_context' (total  12298 bytes in 188 blocks)
    ...
    struct dcesrv_endpoint         contains    691 bytes in  11 blocks (ref 0) 0x55ae964d79a0
        struct dcerpc_ncacn_listen_state contains    160 bytes in   2 blocks (ref 0) 0x55ae964fca60
            struct tevent_fd               contains    104 bytes in   1 blocks (ref 0) 0x55ae964f36f0
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964e8a40
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964e8ac0
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964e8870
            ../../librpc/rpc/dcesrv_core.c:327 contains    104 bytes in   1 blocks (ref 0) 0x55ae964e8970
        struct dcerpc_binding          contains    108 bytes in   2 blocks (ref 0) 0x55ae964e87a0
            \pipe\lsass                    contains     12 bytes in   1 blocks (ref 0) 0x55ae964e88f0
        struct dcerpc_binding          contains    111 bytes in   2 blocks (ref 0) 0x55ae964e86d0
            \pipe\netlogon                 contains     15 bytes in   1 blocks (ref 0) 0x55ae964e8650
    ...
    struct dcesrv_endpoint         contains   1728 bytes in  27 blocks (ref 0) 0x55ae964e0db0
        struct dcerpc_ncacn_listen_state contains    160 bytes in   2 blocks (ref 0) 0x55ae964fd540
            struct tevent_fd               contains    104 bytes in   1 blocks (ref 0) 0x55ae964f3890
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964fbd70
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964f67a0
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964fff60
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964f66c0
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964eb8b0
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964eba00
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964eb200
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964eb350
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964eab70
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964eabf0
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964ea360
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964ea4b0
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964e9cb0
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964e9e00
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964e96a0
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964e97f0
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964e8430
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964e8580
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964e6250
            ../../librpc/rpc/dcesrv_core.c:360 contains    104 bytes in   1 blocks (ref 0) 0x55ae964e8070
        struct dcesrv_if_list          contains    128 bytes in   2 blocks (ref 0) 0x55ae964e7e50
            ../../librpc/rpc/dcesrv_core.c:327 contains    104 bytes in   1 blocks (ref 0) 0x55ae964e7fa0
        struct dcerpc_binding          contains    104 bytes in   2 blocks (ref 0) 0x55ae964e7ed0
            DEFAULT                        contains      8 bytes in   1 blocks (ref 0) 0x55ae964eaa10
    ....

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Andrew Bartlet <abartlet@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Apr  1 10:28:43 UTC 2020 on sn-devel-184

dlz_bind9: Avoid talloc_new(NULL), use a parent variable

This will help provide a better memory tree if we ever suspect a problem
here.  The tmp_ctx varaible is always freed before the end of this
function.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

gitlab-ci: Use a shared ccache for the runners

https://docs.gitlab.com/ee/ci/caching/

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Apr  1 07:25:33 UTC 2020 on sn-devel-184

gitlab-ci: Add ccache and wget

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

smbd: remove struct privilege_paths

The last user of this had been removed by
d485c43cc74c5550d151fe728ddc53b02ab23df8.

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Mar 31 19:32:12 UTC 2020 on sn-devel-184

lib ldb: lmdb clear stale readers on write txn start

In use process failures and Bind9 shut downs leave stale entries in the
lmdb reader table.  This can result in lmdb filling it's database file, as
the free list can not be reclaimed due to the stale reader.

In this fix we call mdb_reader_check at the start of each transaction,
to free any stale readers.  As the default maximum number of readers is
127, this should not impact on performance to any great extent.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14330

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Tue Mar 31 01:26:07 UTC 2020 on sn-devel-184

s3: smbd: RIP smb_filename->original_lcomp.

Removed from struct smb_filename. You will not be missed :-).
Note that VFS ABI modified.

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): Mon Mar 30 16:23:45 UTC 2020 on sn-devel-184

s3: smbd: Remove UCF_SAVE_LCOMP flag. Note it is no longer used.

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

s3: smbd: Remove all references to original_lcomp from name copying code.

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

s3: smbd: Now we no longer use it, remove all references to original_lcomp from pathname processing code.

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

s3: smbd: Use get_original_lcomp() inside smb_file_rename_information().

Pass to rename_internals_fsp() and rename_internals().

Removes the last uses of the struct member original_lcomp
outside of the filename and struct util copying code.

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

s3: smbd: Use get_original_lcomp() inside smb2_file_rename_information().

Pass to rename_internals_fsp(). Note this is a logic change,
as the original code only set smb_fname->original_lcomp if
it was doing a stream rename. Inside rename_internals_fsp()
we only look at original_lcomp in the stream rename case, so
this code worked. However, it is much safer to always correctly
create dst_original_lcomp than pass in a NULL here. It won't
hurt if it's not actually looked at.

Removes one more use of the struct member original_lcomp.

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

s3: smbd: Use get_original_lcomp() inside reply_mv().

Pass to rename_internals().

Removes one more use of the struct member original_lcomp.

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

s3: smbd: Use get_original_lcomp() inside reply_ntrename().

Pass to rename_internals().

Removes one more use of the struct member original_lcomp.

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

s3: smbd: Inside rename_internals() wildcard case, re-purpose dst_original_lcomp.

Pass to rename_internals_fsp() instead of using smb_fname_dst->original_lcomp.

Removes one more use of the struct member original_lcomp.

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

s3: smbd: Add 'const char *dst_original_lcomp' parameter to rename_internals()

Pass through the existing smb_fname_dst->original_lcomp
parameter so no logic change. Preparing to remove the
use of original_lcomp.

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

s3: smbd: Reformatting caller of rename_internals() to make it easer to see changed parameters.

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

s3: smbd: Add a dst_original_lcomp parameter to rename_internals_fsp().

Currently passed in as dst_fname->original_lcomp in all callers
but will eventually be converted to allow original_lcomp to be
removed from struct smb_filename.

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

s3: smbd: Reformatting callers of rename_internals_fsp() to make it easer to see changed parameters.

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

s3: smbd: Remove use of smb_fname->original_lcomp from call_trans2findfirst().

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

s3: smbd: Remove use of smb_fname->original_lcomp from smb2 query directory.

Use get_original_lcomp() call. Removes one use of smb_fname->original_lcomp.

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

s3: smbd: Add utility function get_original_lcomp().

Gets the last component from a client passed-in
filename. Copes with @GMT snapshot and MS-DFS names.

Preparing to remove original_lcomp from
struct smb_filename.

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

s3: smbd: Extract large directory case normalization code into a utility function normalize_filename_case().

It is done before the original_lcomp is extracted
from the client pathname, so the new function that
will allow us to remove original_lcomp also needs
this logic.

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

nsswitch: fix use-after-free causing segfault in _pam_delete_cred

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14327

Guenther

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Mon Mar 30 13:01:20 UTC 2020 on sn-devel-184

s4:selftest: run samba.tests.krb5.simple_tests against ad_dc_default

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Mar 27 19:54:25 UTC 2020 on sn-devel-184

python/tests/krb5: add simple_tests.py with the first simple test

This just demonstrates that the infrastructure works:-)

I'm running this as:

  SERVER=172.31.9.188 DOMAIN=W2012R2-L6 REALM=W2012R2-L6.BASE \
  USERNAME=administrator PASSWORD=A1b2C3d4 SERVICE_USERNAME="w2012r2-188" \
  python/samba/tests/krb5/simple_tests.py

Pair-Programmed-With: Isaac Boukris <iboukris@samba.org>

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Isaac Boukris <iboukris@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

python/tests/krb5: add raw_testcase.py as the base for our Kerberos protocol testing

Pair-Programmed-With: Isaac Boukris <iboukris@samba.org>

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Isaac Boukris <iboukris@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

python/tests/krb5: modify rfc4120.asn1 in order to generate pyasn1 code

The pyasn1 bindings are generated by pyasn1gen.py from
https://github.com/kimgr/asn1ate.git

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

python/tests/krb5: add rfc4120.asn1

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

python/tests/krb5: add support for Cksumtype.MD5

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

s4:selftest: run samba.tests.krb5.kcrypto test

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

python/tests/krb5: convert kcrypto.py to python3-cryptography and a few Samba helpers

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

python/tests/krb5: add crypto.py from greghudson/pyk5 as kcrypto.py

This is crypto.py of commit f0612aa908062fb239d1c3873595e7204ae1691d
from https://github.com/greghudson/pyk5.git

This will be used in order to do raw protocol testing against
[MS-KILE] KDCs.

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

Signed-off-by: Isaac Boukris <iboukris@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>

python/tests: let usage.py be more verbose on errors

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

bootstrap: add python3-pyasn1/python3-cryptography for kerberos testing

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

bootstrap: add ubuntu2004 Ubuntu Focal Fossa (development branch)

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>

selftest: we no longer need '--use-dns-faking' on Linux and FreeBSD

For now I'll leave it arround for others I haven't tested.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Fri Mar 27 10:39:32 UTC 2020 on sn-devel-184

selftest: use 10.53.57.0/8 instead of 127.0.0.1/8

This makes our testing much more realistic and allows
the removal of some knowfail entries.

It also means the testing with network namespaces on Linux
can use the same addresses as our socket wrapper testing.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: pass ipv4 and ipv6 listener addresses to dns_hub.py

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: let realm_to_ip_mappings() return stable (sorted) results

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: {offlinebackupdc,restoredc,customdc} are disconnected from other environments

These restore copies of others and better use their own resolv.conf
and use '--use-samba-tool --no-credentials' for samba_dnsupdate in order
to avoid talking to the real environments.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: correctly use RESOLV_CONF from the DC environment or "no_resolv.conf"

It's important to have the correct $resolv_conf variable within
provision(), because it also sets "RESOLV_WRAPPER_CONF" if needed,
instead of just setting "RESOLV_CONF".

There's also no point in creating an resolv.conf with the ip addresses
for other roles than "active directory domain controller".

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: always explicitly use RESOLV_CONF

By default point RESOLV_CONF to a non-existing file and
use the per environment RESOLV_CONF explicitly where needed.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: split out Samba::mk_resolv_conf() helper

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: we have a global $ENV{UID_WRAPPER} = 1, don't set it again

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: let get_env_for_process() include RESOLV_CONF

This is needed for all environments not just "samba".

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: handle proc{killdom,limit}.samba.example.com in dns_hub

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: allow dns_hub.py to do forwarding to ipv6 servers

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: allow dns_hub.py to listen on more than one address

This makes it possible to serve ipv4 and ipv6 at the same time.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

selftest: make sure dns_hub.py exits on SIGINT

Otherwise we have a deadlock in the python threading
that prevents out EOF detection on stdin to work
if someone aborts 'make test' with strg+c.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

samba_dnsupdate: fix delete (samba-tool) message

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

samba_dnsupdate: fix --no-credentials handling

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

python/netcmd/dns: improve exception handling in cmd_add_record()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

python/netcmd/dns: improve exception handling in cmd_delete_record()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

librpc/tests: make use of replace.h in test_ndr*.c

This fixes the build on FreeBSD 12.1 and maybe other systems.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

s4:torture/ldb: fix the build on FreeBSD

clock_t is 'int' instead' of 'long' there.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

s3:util_sec: fix the build on non-linux platforms

This fixes a regression introduced by
"util_sec.c: Move __thread variable to global scope"
(5a80f399b51221fb0b8661f30d940ca24e1ce627).

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

smbd: use a helper variable in filename_convert_internal()

Avoids the complicated pointer-to-pointer dereferencing.

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Mar 27 02:41:39 UTC 2020 on sn-devel-184

smbd: use filename_convert() in smb_file_rename_information()

Replaces direct calls to resolve_dfspath_wcard() and filename_convert(). On the
way to consolidate all callers of pathname processing onto filename_convert().

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

smbd: modernize DEBUG in unix_convert()

I'm also consolidating on level 10 instead of 5 for "debug" level messages, as
that what's tend to be the canonical debug level these days.

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

smbd: turn (*orig_path != '/') assert into a normal if check

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

smbd: consolidate comments in unix_convert()

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

smbd: modernize unix_convert() replacing True with true

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

smbd: modernize unix_convert() replacing False with false

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

smbd: initialize end to NULL in unix_convert()

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

smbd: leave start at NULL, it's not used before being set to smbfname->basename

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

tdbtorture: Use ARRAY_DEL_ELEMENT()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Thu Mar 26 16:22:00 UTC 2020 on sn-devel-184

ldb: Use ARRAY_DEL_ELEMENT() in ldb_dn_set_extended_component()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

ldb: Use ARRAY_DEL_ELEMENT() in ldb_dn_extended_filter()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

ldb: Use ARRAY_DEL_ELEMENT() in ldb_kv_index_del_value()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

ldb: Use ARRAY_DEL_ELEMENT() in ldb_kv_msg_delete_element()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

libcli: Use ARRAY_DEL_ELEMENT() in security_descriptor_acl_del()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

dsdb: Use ARRAY_DEL_ELEMENT() in dirsync_filter_entry()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

vfs_fruit: Use ARRAY_DEL_ELEMENT() in del_fruit_stream()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

vfs_fruit: Use ARRAY_DEL_ELEMENT() in filter_empty_rsrc_stream()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

libsmb: Use ARRAY_DEL_ELEMENT() in remove_duplicate_addrs2()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

smbd: Use ARRAY_DEL_ELEMENT() in merge_default_aces()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

smbd: Use ARRAY_DEL_ELEMENT() in smbd_smb1_blocked_locks_cleanup()

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>