s3: torture: Add samba3.smbtorture_s3.plain.POSIX-SYMLINK-CHMOD

Shows we must protect against a null fsp handle when doing POSIX chmod on a symlink,
whether the symlink points to a real object or is dangling.

Add to knownfail for now. Commit 9722732b1867e359304594ada72ff40cd1341be5
removed the fsp == NULL protection for POSIX, and we need to put it back.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>

gpo: Open ssh config to write bytes

Reopening the existing config file fails because
we fail to open to write bytes.

Signed-off-by: David Mulder <dmulder@suse.com>
Reviewed-by: Andreas Schneider <asn@samba.org>

s3:passdb: Fix 'return 1' in secrets_store_creds()

The recently introduced function secrets_store_creds() should always
use 'return false' in case of a failure. It is not only spelling issue
since 'return 1' actually means 'return true'.

Signed-off-by: Pavel Filipensky <pavel.filipensky@gmail.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Tue Apr 20 06:42:50 UTC 2021 on sn-devel-184

lib: Fix a typo

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 Apr 19 19:07:01 UTC 2021 on sn-devel-184

auth3: Make auth3_session_info_create() static

Only used in the static artifical session creation

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

torture: Move sddl tests to python

This kind of test is better hosted in python than in C. More lines,
but the ones in source4/libcli/security/tests/sddl.c were preeetty
long...

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

librpc: Add py_descriptor_richcmp() equality function

Only a python3 version. Do we still need the python2 flavor?

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

librpc: Use GUID_buf_string() in python wrappers

No need for the talloc'ed strings

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

py_security: Avoid casts in py_random_sid()

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

libcli: Simplify sddl_encode_ace()

Use GUID_buf_string() instead of GUID_string() for encoding objects,
no need to check for NULL anymore.

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

auth3: Add an error check to auth_generic_prepare()

gensec_set_credentials() can fail

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

auth3: Remove unnecessary talloc_unlink() calls

The structures we unlinked have been talloc_reference()ed in gensec
and thus don't need the second talloc parent anymore. But this
talloc_unlink isn't necessary because tmp_ctx is free()ed a few lines
down.

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

dsdb: Slightly tune get_new_descriptor()

DBG_DEBUG only calls its arguments if required according to the debug
level. A simple talloc_new/TALLOC_FREE in the normal case should be
much cheaper than the full sddl_encode().

I just stumbled across this code, this is has not shown up in any
profiles. I just think it's cleaner this way.

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

auth3: Make load_auth_module() static

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

auth3: Remove auth_skel.c

Authentication is a very complex topic, and someone who is able to
write a custom auth module turning a struct auth_usersupplied_info
into a struct auth_serversupplied_info should be able to live without
this skeleton module.

This module also gave an example to load a secondary authentication
module via a module parameter (the call to load_module()). We have
abandoned this practice, and since the "auth methods" parameter has
gone we don't use this anymore internally.

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

create_local_token: Add error checks

add_sid_to_array_unique() only fails for ENOMEM, and other parts of
the auth stack would probably crash under ENOMEM anyway. But this is
authorization-related code that should be as clean as possible.

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

auth3: Fix a few error path memleaks in create_local_token()

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

auth3: Fix a typo

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

auth3: Simplify check_samba4_security()

First set up "server_info" in a local variable and once it's fully set
up, assign it to the out parameter "pserver_info".

Pointer dereferencing obfuscates the code for me.

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

smbd: Fix a typo

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

auth: Simplify DEBUG statements in make_auth3_context_for_ntlm()

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

lib: Remove two unused historic macros

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

vfs: Remove a call to TALLOC_ZERO()

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

vfs: Replace a call to TALLOC_ZERO()

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

lib: Replace a call to TALLOC_ZERO()

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

auth3: Fix a error path memleak

(find the missing TALLOC_FREE() in the - part of the patch...)

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

auth3: Use talloc_move() instead of talloc_steal()

More recent coding style, avoid ambiguities about ownership

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

winbindd: Fix a typo

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

registry: Fix a typo

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

rpc_client: talloc_stackframe() aborts on failure

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

rpc_client: Direct struct initialization in dcerpc_winreg_enumvals()

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

printing: talloc_stackframe() aborts on failure

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

printing: Straighten winreg_get_printer() slightly

Use the common done: exit for everything. This involves initializing
the handles on the stack, but this is good practice anyway.

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

printing: Make winreg_get_printer() a bit easier to understand

This is more lines, but the FILL_STRING macro did not really gain much
in clarity for me.

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

printing: Make winreg_get_printer() a bit easier to read

EMPTY_STRING does not gain clarity over "" for me.

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

lib: Simplify tdb_fetch_int32()

With tdb_parse_record we don't need malloc/SAFE_FREE.

The semantics are a bit different from tdb_parse_uint32: We just return
-1 on error, but this could be overloaded with a valid -1 record value.

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

lib: Simplify tdb_fetch_uint32_t()

With tdb_parse_record() we don't need malloc/SAFE_FREE

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

lib: Remove unused tdb_traverse_delete_fn()

We have tdb_wipe_all() for that now.

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

lib: Fix nonempty line endings

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

lib: Fix includes in util_tdb.h

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

lib: Fix includes in strv.h

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

s3:script: Remove findsmb from default installation

This tool is the only client tool which requires perl. Distributions are
removing perl from the default installation now.

Also this is a wrapper around nmblookup which is obsolete in the AD
world. However it might still be used by someone so move it just to
examples/scripts/nmb/

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Mon Apr 19 14:37:04 UTC 2021 on sn-devel-184

s3: VFS: Remove SMB_VFS_NTIMES(), no longer used

                         ---------------
                        /               \
                       /      REST       \
                      /        IN         \
                     /        PEACE        \
                    /                       \
                    |                       |
                    |     SMB_VFS_NTIMES    |
                    |                       |
                    |                       |
                    |       13 April        |
                    |         2021          |
                    |                       |
                    |                       |
                   *|     *  *  *           | *
          _________)/\\_//(\/(/\)/\//\/\////|_)_______

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Samuel Cabrero <scabrero@samba.org>
Autobuild-Date(master): Mon Apr 19 13:19:35 UTC 2021 on sn-devel-184

s3: VFS: default: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: unityed_media: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: time_audit: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: snapper: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: shadow_copy2: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: not_implemented: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: media_harmony: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

gpfswrap: Remove wrapper for gpfs_set_times_path()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: gpfs: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: glusterfs: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: full_audit: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: fruit: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: delay_inject: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: ceph_snapshots: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: ceph: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: catia: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: cap: Remove SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: torture: Change cmd_utime to use SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: recycle: set the recycled file times using SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Use SMB_VFS_FNTIMES() instead of SMB_VFS_NTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Pass full fsp to file_ntimes()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Update file times right before closing the underlying fd

Next commits will update file_ntimes() to use handle-based SMB_VFS_FNTIMES().
Move the update_write_time_on_close() call immediately before closing the fd.

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: smbd: Use new debug macros

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: snapper: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: shadow_copy2: Implement VFS_SMB_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: gpfs: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

gpfswrap: Add wrapper for gpfs_set_times()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: glusterfs: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: fruit: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: delay_inject: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: ceph_snapshots: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: ceph: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: catia: Implement SMB_VFS_FNTIMES()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

VFS: Add SMB_VFS_FNTIMES

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

build: Do not check for unused functions futimes() and futimens()

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

s3: VFS: default: ntimes profile not ended when times not changed

Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

auth4: Remove unused auth_unix

auth_unix was in the source4/auth/ntlm for two reasons:
 - inherited from earlier Samba before the Samba4 fork
 - To support the ejs-backed SWAT (web administration tool)

Neither of these are good reasons to keep this unused code
around, there is very unlikely to be a need to support
plaintext PAM authentication in this part of the code in the
future.

See b16362fab65d0700bd6a8cf6569a9e21c7e6b069 for some
context on the historical use case.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Apr 19 07:07:01 UTC 2021 on sn-devel-184

allow tests to be run against a PAM-less build

Indexing the config hash table fails for PAM related values:

    Traceback (most recent call last):
      File "/src/samba/samba/selftest/tests.py", line 49, in <module>
        pam_set_items_so_path = config_hash["PAM_SET_ITEMS_SO_PATH"]
    KeyError: 'PAM_SET_ITEMS_SO_PATH'
    Error creating recipe from python3 /src/samba/samba/selftest/tests.py| at /src/samba/samba/selftest/selftest.pl line 645.

which prevents the test suite from running when built
--without-pam. Access those values using the get() method
instead.

Signed-off-by: Philipp Gesang <philipp.gesang@intra2net.com>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Fri Apr 16 10:27:41 UTC 2021 on sn-devel-184

lib/util: fix timespec normalization

When fixing up timespec structs, negative values for the ns part
should be taken into account. Also, the range for a valid ns part
is [0, 1000000000), not [0, 1000000000].

Signed-off-by: Philipp Gesang <philipp.gesang@intra2net.com>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

auth4: Remove sync check_password from auth_operations

Remove complexity in the data structures, and pushes the async-ness
one level down.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

auth4: Make auth_sam pseudo-async

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

auth4: Make auth_unix pseudo-async

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

auth4: Make auth_developer pseudo-async

This is a simpler approach to really just wrap the code.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

auth4: Make auth_anonymous pseudo-async

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

waf: Check correctly if gnutls has been compiled with fips mode support

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Tue Apr 13 19:17:56 UTC 2021 on sn-devel-184

add .gitlab-ci-coverage.yml for a scheduled build

This will be used by the https://gitlab.com/samba-team/samba
configuration, while https://gitlab.com/samba-team/devel/samba
will still use .gitlab-ci.yml (via the legacy .gitlab-ci-private.yml).

The key point is the usage of the more powerful n1-standard-2
runners for testing.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Tue Apr 13 09:33:14 UTC 2021 on sn-devel-184

script/autobuild.py: split samba-ad-dc-4* tests into two

As single job they used more than 1h.

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

script/autobuild.py: split samba-nopython out of samba-minimal-smbd again

This was using more than 1h as a single job.

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

script/autobuild.py: split samba-no-opath into two tests

This was is basically a combination of 'samba-nt4' and
'samba-fileserver'.

As a single job it used more than 1h only for testing,
while the samba-no-nopath-build uses ~ 10mins (with a filled ccache).
Now we have two test jobs with ~ 30mins.

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

script/autobuild.py: skip lcov step for samba-fips

This doesn't really work and only generates an empty samba-fips.info
file.

Someone familiar with gcov/lcov should look at this and fix it.

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

.gitlab-ci-main.yml: specify the image only by SAMBA_CI_JOB_IMAGE

That way we can construct the url just in one place,
we can also add SAMBA_CI_JOB_IMAGE to the ccache identifier.

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

.gitlab-ci*.yml: only use gitlab.org shared runners if possible

We no longer fallback to our private runner, lets see how that works
out...

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

bootstrap/.gitlab-ci.yml: make sure we force gitlab.com runners for now

We've just added our own runners with 'docker' and 'gce'.

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

.gitlab-ci-main.yml: build coverity using --with-cluster-support

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

.gitlab-ci.yml: move the content to .gitlab-ci-main.yml

We introduce an indirection from
gitlab-ci.yml via .gitlab-ci-default.yml to .gitlab-ci-main.yml

We do that in order to introduce a .gitlab-ci-coverage.yml later
as that will have to use different settings in future.

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

.gitlab-ci.yml: move samba-ad-back{1,2} and samba-schemaupgrade to shared runners

This seems to work quite reliable now.

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

script/autobuild.py: split samba-ad-dc-backup into samba-ad-back{1,2}

This will make it possible to run them in parallel (hopefully on shared
gitlab runners).

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