IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
This is important as Windows clients with KB5028166 seem to
call netr_LogonGetCapabilities with query_level=2 after
a call with query_level=1.
An unpatched Windows Server returns DCERPC_NCA_S_FAULT_INVALID_TAG
for query_level values other than 1.
While Samba tries to return NT_STATUS_NOT_SUPPORTED, but
later fails to marshall the response, which results
in DCERPC_FAULT_BAD_STUB_DATA instead.
Because we don't have any documentation for level 2 yet,
we just try to behave like an unpatched server and
generate DCERPC_NCA_S_FAULT_INVALID_TAG instead of
DCERPC_FAULT_BAD_STUB_DATA.
Which allows patched Windows clients to keep working
against a Samba DC.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15418
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): Mon Jul 17 07:35:09 UTC 2023 on atb-devel-224
(cherry picked from commit dfeabce44f)
This is important as Windows clients with KB5028166 seem to
call netr_LogonGetCapabilities with query_level=2 after
a call with query_level=1.
An unpatched Windows Server returns DCERPC_NCA_S_FAULT_INVALID_TAG
for query_level values other than 1.
While Samba tries to return NT_STATUS_NOT_SUPPORTED, but
later fails to marshall the response, which results
in DCERPC_FAULT_BAD_STUB_DATA instead.
Because we don't have any documentation for level 2 yet,
we just try to behave like an unpatched server and
generate DCERPC_NCA_S_FAULT_INVALID_TAG instead of
DCERPC_FAULT_BAD_STUB_DATA.
Which allows patched Windows clients to keep working
against a Samba DC.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15418
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
(cherry picked from commit d5f1097b62)
The important change it that we expect DCERPC_NCA_S_FAULT_INVALID_TAG
for unsupported query_levels, we allow it to work with servers
with or without support for query_level=2.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15418
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
(cherry picked from commit 404ce08e90)
We don't have any documentation about this yet, but tests against
a Windows Server 2022 patched with KB5028166 revealed that
the response for query_level=2 is exactly the same as
for querey_level=1.
Until we know the reason for query_level=2 we won't
use it as client nor support it in the server, but
we want ndrdump to work.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15418
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
(cherry picked from commit 5f87888ed5)
Instead of returning the real server-side absolute path of shares and search
results, return a fake absolute path replacing the path of the share with the
share name, iow for a share "test" with a server-side path of "/foo/bar", we
previously returned
/foo/bar and
/foo/bar/search/result
and now return
/test and
/test/search/result
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15388
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
The next commit will change the Samba Spotlight server to return absolute paths
that start with the sharename as "/SHARENAME/..." followed by the share path
relative appended.
So given a share
[spotlight]
path = /foo/bar
spotlight = yes
and a file inside this share with a full path of
/foo/bar/dir/file
previously a search that matched this file would returns the absolute
server-side pato of the file, ie
/foo/bar/dir/file
This will be change to
/spotlight/dir/file
As currently the mdscli library and hence the mdsearch tool print out these
paths returned from the server, we have to change the output to accomodate these
fake paths. The only way to do this sensibly is by makeing the paths relative to
the containing share, so just
dir/file
in the example above.
The client learns about the share root path prefix – real server-side of fake in
the future – in an initial handshake in the "share_path" out argument of the
mdssvc_open() RPC call, so the client can use this path to convert the absolute
path to relative.
There is however an additional twist: the macOS Spotlight server prefixes this
absolute path with another prefix, typically "/System/Volumes/Data", so in the
example above the full path for the same search would be
/System/Volumes/Data/foo/bar/dir/file
So macOS does return the full server-side path too, just prefixed with an
additional path. This path prefixed can be queried by the client in the
mdssvc_cmd() RPC call with an Spotlight command of "fetchPropertiesForContext:"
and the path is returned in a dictionary with key "kMDSStorePathScopes". Samba
just returns "/" for this.
Currently the mdscli library doesn't issue this Spotlight RPC
request (fetchPropertiesForContext), so this is added in this commit. In the
end, all search result paths are stripped of the combined prefix
kMDSStorePathScopes + share_path (from mdssvc_open).
eg
kMDSStorePathScopes = /System/Volumes/Data
share_path = /foo/bar
search result = /System/Volumes/Data/foo/bar/dir/file
relative path returned by mdscli = dir/file
Makes sense? :)
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15388
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
sl_pack_alloc() does the buffer allocation that previously all callers of
sl_pack() did themselves.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15388
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Just reduce indentation of the code handling the success case. No change in
behaviour.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15388
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
This is handled by the NDR code transparently.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15388
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
d is talloc_free()d at the end of the functions and the buffer was later used
after beeing freed in the DCERPC layer when sending the packet.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15388
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Prepare for the "path" being a fake path and not the real server-side
path where we won't be able to vfs_stat_fsp() this fake path. Luckily we already
got stat info for the object in mds_add_result() so we can just pass stat info
from there.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15388
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Make it available for other components
Bug: https://bugzilla.samba.org/show_bug.cgi?id=15207
Signed-off-by: Volker Lendecke <vl@samba.org>
(backported from commit d905dbddf8)
[slow@samba.org: subdir_of() didn't exist yet in 4.16 so this just adds it]
Change the dalloc_value_for_key() function to require an additional final
argument which denotes the expected type of the value associated with a key. If
the types don't match, return NULL.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15341
Signed-off-by: Ralph Boehme <slow@samba.org>
Sends a maliciously crafted packet where the value in a key/value style
dictionary for the "scope" key is a simple string object whereas the server
expects an array. As the server doesn't perform type validation on the value, it
crashes when trying to use the "simple" object as a "complex" one.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15341
Signed-off-by: Ralph Boehme <slow@samba.org>
A malicious client could send a packet where subcount is zero, leading to a busy
loop because
count -= subcount
=> count -= 0
=> while (count > 0)
loops forever.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15340
Signed-off-by: Ralph Boehme <slow@samba.org>
Send a maliciously crafted packet where a nil type has a subcount of 0. This
triggers an endless loop in mdssvc sl_unpack_loop().
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15340
Signed-off-by: Ralph Boehme <slow@samba.org>
We already copy at most sizeof(request.data.auth_crap.lm_resp) bytes to the
lm_resp buffer, but we don't cap the length indicator.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15072
Signed-off-by: Ralph Boehme <slow@samba.org>
With WBFLAG_BIG_NTLMV2_BLOB being set plus lm_resp_len too large you
can crash winbind. We don't independently check lm_resp_len
sufficiently.
Discovered via Coverity ID 1504444 Out-of-bounds access
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15072
Signed-off-by: Volker Lendecke <vl@samba.org>
The winbindd_dual_pam_auth_crap() function will be converted to a local
RPC call handler and it won't receive a winbindd_cli_state struct. Move
the checks accessing this struct to the parent.
Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit 74a511a8ea)
This avoids sending new or reset passwords in the clear
(integrity protected only) from samba-tool in particular.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15315
Signed-off-by: Rob van der Linde <rob@catalyst.net.nz>
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Joseph Sutton <josephsutton@catalyst.net.nz>
* CVE-2023-0614 Not-secret but access controlled LDAP attributes can be discovered (bug 15270)
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Joseph Sutton <josephsutton@catalyst.net.nz>
[abartlet@samba.org Adapted to LDB 2.5 series in Samba 4.16]
Setting the LDB_HANDLE_FLAG_UNTRUSTED tells the acl_read module to operate on this request.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Joseph Sutton <josephsutton@catalyst.net.nz>
The chain for transitive evaluation does consider ACLs, avoiding the disclosure of
confidential information.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Joseph Sutton <josephsutton@catalyst.net.nz>
This will allow our dsdb helper search functions to mark the new
request as untrusted, forcing read ACL evaluation (per current behaviour).
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Joseph Sutton <josephsutton@catalyst.net.nz>
[abartlet@samba.org adapted due to Samba 4.16 and lower
not having the patches for CVE-2022-32743]
In the unlikely case that someone adds a confidential indexed attribute
to the schema, LDAP search expressions on that attribute could disclose
information via timing differences. Let's not use the index for searches
on confidential attributes.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Redaction may be expensive if we end up needing to fetch a security
descriptor to verify rights to an attribute. Checking the search scope
is probably cheaper, so do that first.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
This makes it less likely that we forget to handle a case.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
If the AS_SYSTEM control is present, we know we have system privileges,
and have no need to call dsdb_module_am_system().
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Add a hook, acl_redact_msg_for_filter(), in the aclread module, that
marks inaccessible any message elements used by an LDAP search filter
that the user has no right to access. Make the various ldb_match_*()
functions check whether message elements are accessible, and refuse to
match any that are not. Remaining message elements, not mentioned in the
search filter, are checked in aclread_callback(), and any inaccessible
elements are removed at this point.
Certain attributes, namely objectClass, distinguishedName, name, and
objectGUID, are always present, and hence the presence of said
attributes is always allowed to be checked in a search filter. This
corresponds with the behaviour of Windows.
Further, we unconditionally allow the attributes isDeleted and
isRecycled in a check for presence or equality. Windows is not known to
make this special exception, but it seems mostly harmless, and should
mitigate the performance impact on searches made by the show_deleted
module.
As a result of all these changes, our behaviour regarding confidential
attributes happens to match Windows more closely. For the test in
confidential_attr.py, we can now model our attribute handling with
DC_MODE_RETURN_ALL, which corresponds to the behaviour exhibited by
Windows.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
[abartlet@samba.org adapted due to Samba 4.17 and lower
not having the patches for CVE-2020-25720 and 4.16 and lower
not having the patches for CVE-2022-32743 ]
These variables are often used together, and it is useful to have the
setup code in one place.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
[abartlet@samba.org adapted to the use of
acl_check_access_on_attribute as
acl_check_access_on_attribute_implicit_owner is
only in Samba 4.18 and newer]
This function parses a SID from an ldb_message, similar to
samdb_result_dom_sid(), but does it without allocating anything.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
[abartlet@samba.org Adapted for simple conflicts due to
56297449f9 trimming
trailing whitespace]
DSDB_CONTROL_CALCULATED_DEFAULT_SD_OID was added in commit
08187833fe.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
DSDB_CONTROL_FORCE_ALLOW_VALIDATED_DNS_HOSTNAME_SPN_WRITE_OID was added
to source4/dsdb/samdb/samdb.h in commit
c2ab1f4696.
DSDB_EXTENDED_SCHEMA_LOAD was added in commit
1fd4cdfafa.
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 672ec6135f)
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
[abartlet@samba.org This required as context for the above bug]
The object returned by schema_format_value() is a bytes object.
Therefore the search expression would resemble:
(lastKnownParent=<GUID=b'00000000-0000-0000-0000-000000000000'>)
which, due to the extra characters, would fail to match anything.
Fix it to be:
(lastKnownParent=<GUID=00000000-0000-0000-0000-000000000000>)
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
In extended_dn_fix_filter() we had:
req->op.search.tree = ldb_parse_tree_copy_shallow(req, req->op.search.tree);
which overwrote the parse tree on an existing ldb request with a fixed
up tree. This became a problem if a module performed another search with
that same request structure, as extended_dn_in would try to fix up the
already-modified tree for a second time. The fixed-up tree element now
having an extended DN, it would fall foul of the ldb_dn_match_allowed()
check in extended_dn_filter_callback(), and be replaced with an
ALWAYS_FALSE match rule. In practice this meant that <GUID={}> searches
would only work for one search in an ldb request, and fail for
subsequent ones.
Fix this by creating a new request with the modified tree, and leaving
the original request unmodified.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Change all uses of ldb_kv_filter_attrs() to use
ldb_filter_attrs_in_place() instead. This function does less work than
its predecessor, and no longer requires the allocation of a second ldb
message. Some of the work is able to be split out into separate
functions that each accomplish a single task, with a purpose to make the
code clearer.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15270
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
