Volker Lendecke [Thu, 3 Feb 2022 16:17:07 +0000 (17:17 +0100)]
smbd: Only file_free() a self-created fsp in create_file_unixpath()
This fixes a use-after-free in smb_full_audit_create_file() when
calling SMB_VFS_CREATE_FILE with fsp->fsp_name as smb_fname.
create_file_unixpath() has this comment:
* This is really subtle. If someone passes in an smb_fname
* where smb_fname actually is taken from fsp->fsp_name, then
* the lifetime of these objects is meant to be the same.
so it seems legitimate to call CREATE_FILE this way.
When CREATE_FILE runs into an error, create_file_unixpath() does a
file_free, which also takes fsp->fsp_name with
it. smb_full_audit_create_file() wants to log the failure including
the smb_fname after NEXT_CREATE_FILE has exited, but this will then
use the already free'ed data.
Fix by only doing the file_free() on an fsp that
create_file_unixpath() created itself.
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): Thu Feb 10 19:11:33 UTC 2022 on sn-devel-184
Volker Lendecke [Wed, 9 Feb 2022 17:03:33 +0000 (18:03 +0100)]
smbd: Introduce close_file_smb()
This does almost everything that close_file_free() does, but it leaves
the fsp around.
A normal close_file() now calls fsp_unbind_smb() twice. Functionally
this is not a problem, fsp_unbind_smb() is idempotent. The only
potential performance penalty might come from the loops in
remove_smb2_chained_fsp(), but those only are potentially large with
deeply queued smb2 requests. If that turns out to be a problem, we'll
cope with it later. The alternative would be to split up file_free()
into even more routines and make it more difficult to figure out which
of the "rundown/unbind/free" routines to call in any particular
situation.
Volker Lendecke [Wed, 2 Feb 2022 11:42:08 +0000 (12:42 +0100)]
smbd: Simplify the flow in close_file_free()
We are no longer called on base_fsp's in SHUTDOWN_CLOSE. That
simplifies the logic in the common case, we now have a linear flow for
the very often-called close_file()
Volker Lendecke [Wed, 2 Feb 2022 07:58:15 +0000 (08:58 +0100)]
smbd: No base fsps to close_file_free() from file_close_conn()
close_file_free() needs to handle base fsps specially. This can be
simplified a lot if we pass the the open files a second time in case
we encountered base_fsps that we could not immediately delete.
file_close_conn() is not our hot code path, and also we don't expect
many thousand open files that we need to walk a second time.
A subsequent patch will simplify close_file_free(), the complicated
logic is now in files.c, where it IMHO belongs because
file_set_base_fsp() are here as well.
Volker Lendecke [Tue, 1 Feb 2022 16:47:29 +0000 (17:47 +0100)]
smbd: NULL out "fsp" in close_file()
Quite a few places already had this in the caller, but not all. Rename
close_file() to close_file_free() appropriately. We'll factor out
close_file_smb() doing only parts of close_file_free() later.
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com>
Pair-Programmed-With: Andreas Schneider <asn@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Thu Feb 10 22:09:06 UTC 2022 on sn-devel-184
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com>
Pair-Programmed-With: Andreas Schneider <asn@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit a25c714c34d3e00e0f3c29d2acfa98cf9cdbc544)
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 547b4c595a8513a4be99177edbaa39ce43840f7a)
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 2fd518e5cc63221c162c9b3f8526b9b7c9e34969)
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 9f34babec7c6aca3d91f226705d3b3996792e5f1)
We have the env variable SERVER_LOG_LEVEL which allows you to change
the log level on the command line. If we force -d0 this will not work.
make test TESTS="samba" SERVER_LOG_LEVEL=10
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit 9693f7ea7383c6a51ab58b7c8255b30206f18a3b)
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 6063e8016fccbefd1c3fe378e3807c77bc04e4ec)
Autobuild-User(v4-16-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-16-test): Mon Feb 14 09:58:46 UTC 2022 on sn-devel-184
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 2a0d6c11330c40f5692dc07ed6482c7107035bd4)
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 63e00f81b5dd05b50e6ac286e87b8637a4ecd7e0)
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Wed Feb 2 18:29:08 UTC 2022 on sn-devel-184
libcli/smb: fix error checking in smb2_signing_decrypt_pdu() invalid ptext_len
When the ptext_size != m_total check fails, we call this:
status = gnutls_error_to_ntstatus(rc, NT_STATUS_INTERNAL_ERROR);
goto out;
As rc is 0 at that point we'll exit smb2_signing_decrypt_pdu()
with NT_STATUS_OK, but without copying the decrypted data
back into the callers buffer. Which leads to strange errors
in the caller.
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 99182af4ab5a3413311e27c2a193e09babceb01c)
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 68e62962b08497da8359ddbe4324443818c05cd1)
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 Feb 8 20:24:12 UTC 2022 on sn-devel-184
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Noel Power <npower@samba.org>
Autobuild-User(master): Noel Power <npower@samba.org>
Autobuild-Date(master): Fri Feb 4 12:02:36 UTC 2022 on sn-devel-184
Jeremy Allison [Thu, 3 Feb 2022 19:15:30 +0000 (11:15 -0800)]
s3: libsmb: Add cli_dfs_target_check() function.
Strips any DFS prefix from a target name that will be passed
to an SMB1/2/3 rename or hardlink call. Returns a pointer
into the original target name after the prefix. Not yet used.
If the incoming filename is *NOT* a DFS prefix, the
original filename is returned unchanged.
Jeremy Allison [Thu, 3 Feb 2022 22:21:26 +0000 (14:21 -0800)]
s3: tests: Add a new test test_msdfs_rename() that does simple renames on MSDFS root shares.
We fail this on SMB2 for a subtle reason.
Our client code called from smbclient only sets the SMB2_HDR_FLAG_DFS flag
in the outgoing packet on the SMB2_CREATE call, and SMB2 rename does the
following operations:
SMB2_CREATE(src_path) // We set SMB2_HDR_FLAG_DFS here for a MSDFS share.
SMB2_SETINFO: SMB2_FILE_RENAME_INFO(dst_path). // We don't set SMB2_HDR_FLAG_DFS
However, from smbclient, dst_path is a MSDFS path but we don't set the flag,
so even though the rename code inside smbd will cope with a MSDFS path
(as used in the SMB1 SMBmv call) it fails as the correct flag isn't set.
Add knownfail selftest/knownfail.d/msdfs-rename.
Note we need to add the new test to "selftest/knownfail.d/smb1-tests"
as test_smbclient_s3.sh is run against the (ad_member|nt4_member)
environments first using NT1 (SMB1) protocol and then using SMB3,
but the (ad_member|nt4_member) environments don't support SMB1.
Seems a bit strange to me, but all the other SMB1 tests inside
test_smbclient_s3.sh have already been added to "selftest/knownfail.d/smb1-tests"
so just go with the test environment.
Jeremy Allison [Thu, 3 Feb 2022 21:58:28 +0000 (13:58 -0800)]
s3: tests: Add a new test test_msdfs_hardlink() that does simple hardlinks on MSDFS root shares.
We pass this already as the cmd_hardlink in smbclient doesn't
do the DFS path conversion on the hardlink target. But it's
good to have the test.
Note we need to add the new test to "selftest/knownfail.d/smb1-tests"
as test_smbclient_s3.sh is run against the (ad_member|nt4_member)
environments first using NT1 (SMB1) protocol and then using SMB3,
but the (ad_member|nt4_member) environments don't support SMB1.
Seems a bit strange to me, but all the other SMB1 tests inside
test_smbclient_s3.sh have already been added to "selftest/knownfail.d/smb1-tests"
so just go with the test environment.
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 Feb 2 21:50:31 UTC 2022 on sn-devel-184
Jule Anger [Mon, 31 Jan 2022 11:56:33 +0000 (12:56 +0100)]
VERSION: Bump version up to Samba 4.16.0rc3...
and re-enable GIT_SNAPSHOT.
Signed-off-by: Jule Anger <janger@samba.org>
Autobuild-User(v4-16-test): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(v4-16-test): Mon Jan 31 15:26:29 UTC 2022 on sn-devel-184
This was removed accidentally remvoed by a7c65958a15149918415b7456d6f20ee8c9669d2 because the original code
only installed the json file if the mdssvc was built as module:
if bld.SAMBA3_IS_ENABLED_MODULE('rpc_mdssvc_module'):
bld.INSTALL_FILES(bld.env.SAMBA_DATADIR,
'mdssvc/elasticsearch_mappings.json')
Installing the json file should just depend on Elasticsearch support
being enabled, regardless of the removed module support.
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Noel Power <npower@samba.org>
Autobuild-User(master): Noel Power <npower@samba.org>
Autobuild-Date(master): Fri Jan 28 10:22:31 UTC 2022 on sn-devel-184
Jeremy Allison [Wed, 8 Dec 2021 06:15:46 +0000 (22:15 -0800)]
CVE-2021-44141: s3: torture: Add a test samba3.blackbox.test_symlink_rename.SMB1.posix that shows we still leak target info across a SMB1+POSIX rename.
Jeremy Allison [Tue, 7 Dec 2021 22:39:42 +0000 (14:39 -0800)]
CVE-2021-44141: s3: smbd: Fix a subtle bug in the error returns from filename_convert().
If filename_convert() fails to convert the path, we never call
check_name(). This means we can return an incorrect error code
(NT_STATUS_ACCESS_DENIED) if we ran into a symlink that points
outside the share to a non-readable directory. We need to make
sure in this case we always call check_name().
Jeremy Allison [Tue, 7 Dec 2021 20:56:51 +0000 (12:56 -0800)]
CVE-2021-44141: s3: torture: In test_smbclient_s3, change the error codes expected for test_widelinks() and test_nosymlinks() from ACCESS_DENIED to NT_STATUS_OBJECT_NAME_NOT_FOUND.
For SMB1/2/3 (minus posix) we need to treat bad symlinks
as though they don't exist.
This ensures ad_unpack_xattrs() is only called for an ad_type of ADOUBLE_RSRC,
which is used for parsing ._ AppleDouble sidecar files, and the buffer
ad->ad_data is AD_XATTR_MAX_HDR_SIZE bytes large which is a prerequisite for all
buffer out-of-bounds access checks in ad_unpack_xattrs().
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Mon Jan 24 16:18:34 UTC 2022 on sn-devel-184
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 43648e95a514020da4c7efa62df55d0882e3db85)
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 8da26cb6725b5d853ab481a348a3a672966715b5)
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 1dc385cb648f0c37b04f4ede6b1c96916e379b23)
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit d1a7f392a8ceef111a5d6c3d2a3bdb9dcb90db5e)
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit b489b7feda19b3c0f0fe2300f2c76d416776355b)
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 9110a8854a518befa2908c26076e17a085c5ec48)
It also fixes some corner cases regarding whitespaces in the
output, that's why we have to mark a few tests as knownfail,
they will be fixed in the next commit.
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 58b09e107cadd7fb8191822d4e7e42657b1ed4c7)
samba-bgqd daemon is started by existing Samba daemons. When running
under systemd, those daemons control systemd notifications and
samba-bgqd messages need to be silenced.
dcesrv_core: wrap gensec_*() calls in [un]become_root() calls
This is important for the source3/rpc_server code as it might
be called embedded in smbd and may not run as root with access
to our private tdb/ldb files.
Note this is only really needed for 4.15 and older, as
we no longer run the rpc_server embedded in smbd,
but we better be consistent for now.
This should be able to fix the problem the printing no longer works
on Windows 7 with 2021-10 monthly rollup patch (KB5006743).
Windows uses NTLMSSP with privacy at the DCERPC layer on top
of NCACN_NP (smb).
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 0651fa474cd68b18d8eb9bdc7c4ba5b847ba9ad9)
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(v4-16-test): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(v4-16-test): Fri Jan 28 11:17:33 UTC 2022 on sn-devel-184
Björn Jacke [Sun, 23 Jan 2022 11:35:22 +0000 (12:35 +0100)]
s4:librpc: raise log level for failed connection attempts
this keeps the log files silent when other DCs are currently not running. We
saw frequent NT_STATUS_HOST_UNREACHABLE messages at log level 0 for now.
https://bugzilla.samba.org/show_bug.cgi?id=11537
Signed-off-by: Bjoern Jacke <bjacke@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Sun Jan 23 12:51:44 UTC 2022 on sn-devel-184
Pair-Programmed-With: Andreas Schneider <asn@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> 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): Sat Jan 22 00:27:52 UTC 2022 on sn-devel-184
Pair-Programmed-With: Andreas Schneider <asn@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Pair-Programmed-With: Andreas Schneider <asn@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Pair-Programmed-With: Andreas Schneider <asn@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Pair-Programmed-With: Andreas Schneider <asn@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Pair-Programmed-With: Andreas Schneider <asn@samba.org>
Signed-off-by: Pavel Filipenský <pfilipen@redhat.com> Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
wafsamba: Add our own implmentation to generate the clangdb
Pair-Programmed-With: Andreas Schneider <asn@samba.org> Signed-off-by: Andreas Schneider <asn@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org>
This generates an incomplete database where defines and includes are missing.
Pair-Programmed-With: Andreas Schneider <asn@samba.org> Signed-off-by: Andreas Schneider <asn@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org>
Volker Lendecke [Mon, 17 Jan 2022 09:49:13 +0000 (10:49 +0100)]
build: Without getrandom() require gnutls 3.7.2
gnutls before 3.7.2 and without getrandom() will open /dev/urandom at library
initialization time before main() is run. We use closefrom(3) in samba-bgqd and
samba-dcerpd, which closes /dev/urandom, which then breaks gnutls. On system
with getrandom(), no file descriptor is opened and gnutls 3.7.2+ will open and
close /dev/urandom whenever it needs to access it.
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): Fri Jan 21 21:42:08 UTC 2022 on sn-devel-184
Volker Lendecke [Thu, 20 Jan 2022 11:23:43 +0000 (12:23 +0100)]
libcli/dns: Fix TCP fallback
A customer has come across a DNS server that really just cuts a SRV
reply if it's too long. This makes the packet invalid according to
ndr_pull and according to wireshark. DNS_FLAG_TRUNCATION is however
set. As this seems to be legal according to the DNS RFCs, we need to
hand-parse the first two uint16's and look whether DNS_FLAG_TRUNCATION
is set.
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): Thu Jan 20 18:01:41 UTC 2022 on sn-devel-184
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): Thu Jan 20 14:19:02 UTC 2022 on sn-devel-184
s4:dsdb/vlv_pagination: fix segfault in vlv_results()
It can happen that the vlv_results() failes, e.g. due to
LDB_ERR_TIME_LIMIT_EXCEEDED, if that happens we should not
dereference ares->response, if ares is NULL.
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Jan 20 10:04:39 UTC 2022 on sn-devel-184
s4:dsdb/paged_results: fix segfault in paged_results()
It can happen that the paged_results() failes, e.g. due to
LDB_ERR_TIME_LIMIT_EXCEEDED, if that happens we should not
dereference ares->response, if ares is NULL.
We also should not call ldb_module_done() if paged_results()
fails, as it was already called.
HEIMDAL: move code from source4/heimdal* to third_party/heimdal*
This makes it clearer that we always want to do heimdal changes
via the lorikeet-heimdal repository.
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Autobuild-User(master): Joseph Sutton <jsutton@samba.org>
Autobuild-Date(master): Wed Jan 19 21:41:59 UTC 2022 on sn-devel-184
Joseph Sutton [Sun, 5 Dec 2021 22:10:01 +0000 (11:10 +1300)]
s4:torture: Adapt LSA tests to newer Heimdal version
The Heimdal upgrade results in some changes that affect these tests. The
cname is now non-NULL in certain circumstances, the IO counts are
different due to a change between the ordering of capaths and referrals,
some requests no longer fail, and referral tickets are not cached
anymore, and so cannot be checked.
NOTE: THIS COMMIT WON'T COMPILE/WORK ON ITS OWN!
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org>
projects / thirdparty / samba.git / log
