smbclient: Simplify the queue for recursive listing

We now have talloc and the DLIST macros. That simplifies things a bit.

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

smbclient: Simplify do_list_helper()

Do an early return when we don't want to recurse

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

smbclient: Slightly simplify do_list()

Nonrecursive listing is just a special case of recursive
listing. do_list_helper() checks that.

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

smbclient: Do early return in do_list_helper().

Align integer types.

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

smbclient: Align integer types

gcc complained that the if-condition compared unsigned rb_size with a
signed value. Somehow through the arithmetic the uint16_t's got
promoted to integer.

Also, avoid some printf casts

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

smbclient: Align some integer types

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

vfs: Fix typos

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

libsmb: Fix a typo

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

registry3: Fix a typo

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

s3/torture: test rbtree TDB_INSERT and TDB_MODIFY flags

Confirm that record overwrite with TDB_INSERT and record insert with
TDB_MODIFY both fail with appropriate error values.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Jun 10 20:28:45 UTC 2020 on sn-devel-184

s3/torture: use stack buffer for rbtree loop

Using the stack here simplifies the error paths.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

dbwrap_rbt: support TDB_INSERT and TDB_MODIFY store flags

These flags provide insert-new and overwrite-existing record semantics
respectively.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

vfs_default: Remove an unused data member

This was added as part of 7f7ce0ec2f3e3cfb46314e5ad3ea6b5c49085f1d but
never got consumed.

Signed-off-by: Anoop C S <anoopcs@redhat.com>
Reviewed-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

Revert "s3:libsmb: add a cache for cli_session_creds_prepare_krb5()"

This reverts commit b458f8fbb7febf3584fa648128b4e1dc764059f7.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Jun 10 10:10:16 UTC 2020 on sn-devel-184

Add docs build to CI

We did not check we could actually build the HTML of the
Samba Developers guide and HTML of the manpages previously.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Wed Jun 10 07:11:59 UTC 2020 on sn-devel-184

docs-xml: Remove GNU TexInfo build

This does not build and is not a common or required format for
documentation any more.

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

docs-xml: Remove references to inkscape (not used any more, no more SVG files)

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

docs-xml: Remove final references to Samba3-HOWTO and Samba3-ByExample

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

docs-xml: Remove references to building docs using Plucker

This was for Palm OS based handheld devices, Windows Mobile devices, and other PDAs...

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

Remove "undocumented" target mentioned in configure script

This was left over from 12aed897ec688d5bc379690208e8b85158b8227f

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

util: Reallocate larger buffer if getpwuid_r() returns ERANGE

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Tue Jun  9 21:07:24 UTC 2020 on sn-devel-184

util: Fix build on FreeBSD by avoiding NSS_BUFLEN_PASSWD

NSS_BUFLEN_PASSWD is not defined on FreeBSD.  Use
sysconf(_SC_GETPW_R_SIZE_MAX) instead, as per POSIX.

Use a dynamically allocated buffer instead of trying to cram all of
the logic into the declarations.  This will come in useful later
anyway.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Volker Lendecke <vl@samba.org>

util: Simplify input validation

It appears that snprintf(3) is being used for input validation.
However, this seems like overkill because it causes szPath to be
copied an extra time.  The mostly likely protections being sought
here, according to https://cwe.mitre.org/data/definitions/20.html,
look to be DoS attacks involving CPU and memory usage.  A simpler
check that uses strnlen(3) can mitigate against both of these and is
simpler.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Volker Lendecke <vl@samba.org>

s3:libads: remove unused "GSSAPI" support

All AD servers support "GSS-SPNEGO". So we better
remove code that doesn't use gensec.

If we ever need this back we could use the
"gssapi_krb5_sasl" gensec module explicit
or just pass the SASL mech list to gensec.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Isaac Boukris <iboukris@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Tue Jun  9 17:24:31 UTC 2020 on sn-devel-184

talloc: Mark ptr of talloc_unlink() not as a tainted scalar

This should address a lot of issues reported by Coverity.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

s3:libsmb: add a cache for cli_session_creds_prepare_krb5()

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

s4:torture:smb2: use delete-on-close in test_rw_invalid()

We test the limits here and leave a 16TB file with zeros.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14361

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Jun  5 13:17:55 UTC 2020 on sn-devel-184

pysmbd: make sure that session unix info is filled

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14400

Signed-off-by: Björn Baumbach <bb@sernet.de>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Fri Jun  5 11:54:06 UTC 2020 on sn-devel-184

tests/pysmbd: fill session unix info in ntacl tests

Valid unix info is required.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14400

Signed-off-by: Björn Baumbach <bb@sernet.de>
Reviewed-by: Ralph Boehme <slow@samba.org>

python/samba/provision: set unix session info for user session, used for sysvol acl reset

The unix session info is required and expected by e.g. many vfs
modules. Missing unix session info leads to samba panic.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14400

Signed-off-by: Björn Baumbach <bb@sernet.de>
Reviewed-by: Ralph Boehme <slow@samba.org>

pyauth: add python binding for auth_session_info_set_unix()

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14400

Signed-off-by: Björn Baumbach <bb@sernet.de>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4-auth/unix_token: add new function auth_session_info_set_unix()

Used to fill the unix info in a struct auth_session_info similar to
auth_session_info_fill_unix().

The new auth_session_info_set_unix() receives the uid and gid for
the unix token as an parameter. It does not query the unix token from
winbind (via security_token_to_unix_token()).
This is useful to fill a user session info manually if winbind is not
available.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14400

Signed-off-by: Björn Baumbach <bb@sernet.de>
Reviewed-by: Ralph Boehme <slow@samba.org>

s4-auth/unix_token: separate out filling the unix_info elements in a struct session_info

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14400

Signed-off-by: Björn Baumbach <bb@sernet.de>
Reviewed-by: Ralph Boehme <slow@samba.org>

python: fix slow's mail address

Signed-off-by: Björn Baumbach <bb@sernet.de>
Reviewed-by: Ralph Boehme <slow@samba.org>

libsmb: Remove unused cli_smb2_qfileinfo_basic()

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 Jun  4 18:36:17 UTC 2020 on sn-devel-184

libsmb: Use cli_qfileinfo_basic() in cli_smb2_qpathinfo2()

Last user of cli_smb2_qfileinfo_basic().

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

torture3: Fix a debug message

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

libsmb: Remove unused cli_smb2_getattrE()

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

libsmb: Use cli_qfileinfo_basic() in cli_smb2_getatr()

This was the only remaining user of cli_smb2_getattrE(), and as
cli_qfileinfo_basic() now does all protocols, we can get rid of
cli_smb2_getattrE().

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

pylibsmb: Use cli_qfileinfo_basic_send() in py_smb_filesize()

This now does all protocols asynchronously

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

libsmb: Make qfileinfo_basic available for smb2 in async mode

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

libsmb: Remove unused sync version of cli_getattrE()

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

libsmb: Remove cli_getattrE() fallback from cli_qfileinfo_basic() users

cli_qfileinfo_basic() now takes care of this centrally

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

libsmb: Use SMBgetattrE in cli_qfileinfo_basic_send() if necessary

This is a behaviour change: Before this patch, independent of the actual
protocol we tried to do the trans2 getinfo call. All the remaining callers just
do a direct fallback to SMBgetattrE when that fails without even looking at the
error code. Here we deterministically decide after the negotiated protocol
which flavour to use without a fallback.

It *might* be relevant for very old embedded systems that we don't know, but if
we break something we can easily fix it.

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

net: Align integer types, move variable closer to its use

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

net: Copy hires timestamps in net_copy_fileattr()

In this routine we're using cli_ntcreate, so I would assume the hires
timestamp calls to be available. My goal is to reduce the use of the
second-resolution calls and potentially push their fallback use into
clifile.c.

Only tested manually.

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

torture3: Use cli_qfileinfo_basic() in run_posix_append()

Only Samba does the SMB1 posix extensions, and we do the "advanced"
trans-based call.

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

libsmb: Add async cli_qfileinfo_basic

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

s3: client. Move to all 32-bit attributes in smbclient.

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

s3: torture: Add simple tests to samba3.base.attr to check INVALID_PARAMETER returned on SMB1 out-of-range attribute use.

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

s3: libsmb: cli_setatr() and async version. Don't allow attribtes > 16 bits on an old protocol send.

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

s3: libsmb: cli_unlink() and async version. Don't allow attribtes > 16 bits on an old protocol send.

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

s3: libsmb: Change cli_unlink() and async version to take a uint32_t attribute.

Fix the callers.

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

s3: libsmb: Change cli_setfileinfo_ext() and async version to take a uint32_t attr.

Fix all callers.

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

s3: libsmb: Change cli_setpathinfo_ext() to take 32-bit attributes.

Fix the callers. Note the special casing of mapping (uint16)-1 -> (uint32_t)-1
in SMBC_setatr() where we can't change the ABI.

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

s3: libsmb: Change cli_setatr() and async version to use 32-bit attributes.

Fix the callers. Only sets 16 bits on the wire for this level.

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

s3: libsmb: Make cli_list_trans() and cli_list take a 32-bit attribute.

All sub-functions take 32-bits, and all callers now are
compatible with 32-bit attributes.

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

s3: libsmb: Make cli_list_old() take 32-bit attribute.

Only uses 16 on the wire. No change needed in callers.

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

s3: libsmb: Change cli_smb2_setatr() to use 32-bit attributes.

The SMB_SET_FILE_BASIC_INFO info level this uses in SMB2
sets 32-bit attributes, so don't use SSVAL, use SIVAL.

No change needed in callers as implicit casts work fine.

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

s3: Fix dfs-reparse testing code to use 32-bit attributes. Change variable names mode -> attr.

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

s3: libsmb: Change cli_qpathinfo1() and async version to return a 32-bit attribute.

Only used in torture.

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

s3: libsmb: Change cli_qpathinfo3() to return a 32-bit attribute.

Fix all callers.

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

s3: libsmb: Change cli_qpathinfo2() and async versions to return a 32-bit attribute.

Fix all callers.

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

s3: libsmb: Change cli_getatr() and async versions to return a 32-bit attribute.

Fix all callers.

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

s3: libsmb: Rename uint16_t attr -> sattr (short attributes).

This will make it easier to add 32-bit attr value later.

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

s3: libsmb: Change cli_getattrE() and async versions to return a 32-bit attribute.

Fix all callers.

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

s3: libsmb: Change cli_qfileinfo_basic() to return 32-bit attributes.

Fix all callers.

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

s3: libsmb: Change cli_smb2_qpathinfo2() to return a 32 bit attribute.

Fix all callers. All libsmb SMB2 calls now return 32 bit attributes.

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

s3: libsmb: Fix cli_smb2_getatr() to return a 32 bit attribute.

Fix all callers.

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

s3: libsmb: Change cli_smb2_getattrE() to return 32 bit attributes.

Fix all callers.

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

s3: libsmb: Change cli_smb2_qfileinfo_basic() and all users to return a uint32_t attribute.

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

s3: smbclient: Rename mode parameter to attr.

This is a DOS attribute, not a UNIX mode.

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

s3: libsmb: clifile.c: Rename all uint16_t *attr parameters to uint16_t *pattr.

Will make it easier to move to 32 bit attributes later.

No logic changes, just a rename.

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

s3: libsmb: clirap. Rename all mode/ *mode parameters to attr and pattr.

They are attributes, not unix modes.

Once they have the right name, we can start moving all uses to uint32_t.

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

s3: libsmb: cli_smb2_qfileinfo_basic()/cli_smb2_qpathinfo2(). Rename a parameter from mode -> pattr.

Also rename attr pointers to pattr.

They are pointers to attributes.

Only parameter name changes, no logic changes.

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

s3: libsmb: Change size of finfo->attr to uint32_t.

That's what modern servers return.

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

s3: libsmb: SMBC_chmod_ctx(), rename mode -> attr when setting attributes.

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

s3: libsmb: Internally rename to use the name attr instead of mode for attributes.

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

s3: libsmb: Internally rename to use the name attr instead of mode for attributes.

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

s3: libsmb: In struct file_info rename mode -> attr.

Cleanup.  It's never been a UNIX mode, always a DOS attribute field.
Make that explicit.

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

examples: clifuse: Info level SMB_FIND_FILE_BOTH_DIRECTORY_INFO encodes attibutes as a uint32, not a uint8.

Cast to a uint16_t for now after pulling the information
as finfo->mode is currently only 16 bits.

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

Update SECURITY.md to point to security releases

Signed-off-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 Jun  3 15:55:10 UTC 2020 on sn-devel-184

s3: VFS: default. Ensure read_dfs_pathat() returns stat info.

Remove the knownfail.d/msdfs-attr file.

Everything now passes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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): Wed Jun  3 06:19:21 UTC 2020 on sn-devel-184

s3: VFS: shadow_copy2. Ensure read_dfs_pathat() returns stat info.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: VFS: gluster. Ensure read_dfs_pathat() returns stat info.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: VFS: ceph. Ensure read_dfs_pathat() returns stat info.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: VFS: catia. Ensure read_dfs_pathat() returns stat info.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: VFS: cap. Ensure read_dfs_pathat() returns stat info.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: VFS: Change the function signature for SMB_VFS_READ_DFS_PATHAT() to take a non-const smb_filename.

Otherwise there's no good way to return proper stat(2) information
for a DFS link without making assumptions it's a symlink store.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: torture: Add test for getting attibutes on an MSDFS link.

Mark as knownfail for now.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: torture: Add a MSDFS-ATTRIBUTE test.

Framework to drive comes next.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: libsmb: Info level SMB_FIND_EA_SIZE encodes attibutes as a uint16, not a uint8.

We will need this to detect FILE_ATTRIBUTE_REPARSE_POINT in a later commit.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: libsmb: Info level SMB_FIND_INFO_STANDARD encodes attibutes as a uint16, not a uint8.

We will need this to detect FILE_ATTRIBUTE_REPARSE_POINT in a later commit.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: libsmb: Info level SMB_FIND_FILE_BOTH_DIRECTORY_INFO encodes attibutes as a uint32, not a uint8.

Cast to a uint16_t for now after pulling the information
as finfo->mode is currently only 16 bits.

We will need this to detect FILE_ATTRIBUTE_REPARSE_POINT in a later commit.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14391

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

s3: libsmb: Info level SMB2_FIND_ID_BOTH_DIRECTORY_INFO encodes attibutes as a uint32, not a uint8.

Fix the SMB2 parsing code.

Cast to a uint16_t for now after pulling the information
as finfo->mode is currently only 16 bits.

We will need this to detect FILE_ATTRIBUTE_REPARSE_POINT in a later commit.

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

s3/rpc_server: remove unnecessary srv_fss_agent.h header

The srv_fssa_start() / _cleanup() functions are called via the
DCESRV_INTERFACE_FILESERVERVSSAGENT_INIT / SHUTDOWN_SERVER hooks,
so needn't be public.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Reviewed-by: Samuel Cabrero <scabrero@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Wed Jun  3 03:57:12 UTC 2020 on sn-devel-184

smbd: simplify uid_entry_in_group()

conn->session_info is always valid since
1fa05e5c76e718142326915f8671299217a093df and preceeding commits.

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): Sat May 30 02:40:22 UTC 2020 on sn-devel-184

libcli: simplify an if condition

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

s3:auth: fill in info3 domain name in passwd_to_SamInfo3()

This is needed for make_session_info_from_username() to fill in the domain
name. The callchain goes like this:

make_session_info_from_username()
-> make_server_info_pw()
-> passwd_to_SamInfo3()

Other callers of passwd_to_SamInfo3() are also affected. A subsequent change
will let set_conn_force_user_group() user the domainname from the session_info
returned from make_session_info_from_username() to create substitution
context. Without this fix domain name would be NULL.

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

smbd: RIP conn->origpath

conn->origpath is always a duplicate of conn->connectpath.

The only function that sets conn->connectpath is set_conn_connectpath() and
everywhere it's called, there's a subsequent talloc_strdup() into
conn->origpath.

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