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.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
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): Thu May 30 22:12:50 UTC 2019 on sn-devel-184
No need to mmap() anyway, the xattr data is already available in ad->ad_data.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We now have an fsp that we can use, so we can get rid of mmap() and
sys_pread()/sys_pwrite().
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
A first step in converting all raw syscalls to use proper VFS functions. All
existing users of the raw system filedescriptor continue to use the fd from
fsp->fh for now.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The subsystem consumers have been reworked in the previous commits, so this is
not used anymore. ad_init() doesn't need a handle argument anymore due to this,
remove it as well.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This was only needed to get the resourcefork size via the ad_* AppleDouble
function. This is now done with a fstat on the low level xattr fd (remember,
this is Solaris only code...), so we can remove the xattr special casing from
the AppleDouble functions.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This works as well, using an fstat() on the filehandle to get the size. This is
tested by the torture test "vfs.fruit.SMB2/CREATE context AAPL".
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is a genuine bug, but luckily this would only impact configs which nobody
uses:
fruit:metadata = netatalk
fruit:resource = stream
With the above configuration the switch in readdir_attr_rfork_size() would hit
the default case and so always report resource forks as 0 bytes in size.
All deployment that I've seen that use fruit:resource=stream also use
fruit:metadata=stream, so the switch takes FRUIT_META_STREAM case which runs the
correct code readdir_attr_rfork_size_stream().
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Otherwise, if SMB_VFS_UNLINK() is called for an AppleDouble path "._file", we
try to delete "._._file" which doesn't make sense. AppleDouble files don't have
AppleDouble themselves.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Luckily the missing else has the same control flow due to the previous if and
else blocks calling return.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This adds a helper function that checks whether the last component of a path is
an AppleDouble sidecar file with "._" name prefix.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 10.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 9.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 8.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 7.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 5.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Continuing to ignore a possible error for now, this is in an error codepath
anyway.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 4.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 3.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 2.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On the course of removing ad_handle from struct adouble, step 1.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13968
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This moves the trigger points where AppleDouble file conversion is run by
ad_convert() from deep down the callchain in ad_read_rsrc_adouble() to high
level VFS entry points.
Currently ad_convert() will be triggered as part of open_file_ntcreate(...,
"file:AFP_AfpResource", ...): after SMB_VFS_OPEN() has been called with O_CREAT,
what created the file, we call SMB_VFS_FSTAT() on the just created
filehandle. This ends up in ad_convert(), finds the resource fork empty and thus
deletes the file.
This commit moves calling of the conversion funtion to the high level VFS entry
points where the converted metadata is needed:
o for directory enumerations SMB_VFS_READDIR_ATTR() is called to fill in the
repurposed fields in the directory entry metadata
o obviously for SMB_VFS_CREATE_FILE() on an macOS stream
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13958
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This exhibited itself as a problem with OFD locks reported
as:
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13770
However, due to underlying bugs in the vfs_fruit
code the file locks were not being properly applied.
There are two problems in fruit_check_access().
Problem #1:
Inside fruit_check_access() we have:
flags = fcntl(fsp->fh->fd, F_GETFL);
..
if (flags & (O_RDONLY|O_RDWR)) {
We shouldn't be calling fcntl(fsp->fh->fd, ..) directly.
fsp->fh->fd may be a made up number from an underlying
VFS module that has no meaning to a system call.
Secondly, in all POSIX systems - O_RDONLY is defined as
*zero*. O_RDWR = 2.
Which means flags & (O_RDONLY|O_RDWR) becomes (flags & 2),
not what we actually thought.
Problem #2:
deny_mode is *not* a bitmask, it's a set of discrete values.
Inside fruit_check_access() we have:
if (deny_mode & DENY_READ) and also (deny_mode & DENY_WRITE)
However, deny modes are defined as:
/* deny modes */
define DENY_DOS 0
define DENY_ALL 1
define DENY_WRITE 2
define DENY_READ 3
define DENY_NONE 4
define DENY_FCB 7
so if deny_mode = DENY_WRITE, or if deny_mode = DENY_READ
then it's going to trigger both the if (deny_mode & DENY_READ)
*and* the (deny_mode & DENY_WRITE) conditions.
These problems allowed the original test test_netatalk_lock code to
pass (which was added for BUG: https://bugzilla.samba.org/show_bug.cgi?id=13584
to demonstrate the lock order violation).
This patch refactors the fruit_check_access()
code to be much simpler (IMHO) to understand.
Firstly, pass in the SMB1/2 share mode, not old
DOS deny modes.
Secondly, read all the possible NetAtalk locks
into local variables:
netatalk_already_open_for_reading
netatalk_already_open_with_deny_read
netatalk_already_open_for_writing
netatalk_already_open_with_deny_write
Then do the share mode/access mode checks
with the requested values against any stored
netatalk modes/access modes.
Finally add in NetATalk compatible locks
that represent our share modes/access modes
into the file, with an early return if we don't
have FILE_READ_DATA (in which case we can't
write locks anyway).
The patch is easier to understand by looking
at the completed patched fruit_check_access()
function, rather than trying to look at the
diff.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Böhme <slow@samba.org>
https://bugzilla.samba.org/show_bug.cgi?id=13725
We cannot always rely on vfs_default to close the fake fds. This mostly is
relevant when used with another non-local VFS filesystem module such as
gluster.
Guenther
Signed-off-by: Günther Deschner <gd@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Dec 21 07:20:49 CET 2018 on sn-devel-144
This helps avoiding a NULL dereference on systems where additional
patches modify the following condition in open_file()
if ((open_access_mask & (FILE_READ_DATA|FILE_WRITE_DATA|FILE_APPEND_DATA|FILE_EXECUTE)) ||
(!file_existed && (local_flags & O_CREAT)) ||
((local_flags & O_TRUNC) == O_TRUNC) ) {
to
if ((open_access_mask & (FILE_READ_DATA|FILE_WRITE_DATA|FILE_APPEND_DATA|FILE_EXECUTE|DELETE_ACCESS)) ||
(!file_existed && (local_flags & O_CREAT)) ||
((local_flags & O_TRUNC) == O_TRUNC) ) {
Ie addtionally check open_access_mask against DELETE_ACCESS. As a result
opens with DELETE_ACCESS go through the code that does an fd_open() plus
a subsequent fstat().
That will trigger a crash in fruit_fstat_meta_stream() when a client
wants to delete a file for deletion. When we open base file for delete,
we call open_streams_for_delete() which internally calls create-file
with NTCREATEX_OPTIONS_PRIVATE_STREAM_DELETE which prevents opening of
the base_fsp. Voila, combined with the change described above you get a
NULL deref.
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): Sun Dec 2 07:52:34 CET 2018 on sn-devel-144
This is the final step in implementing the needed macOS semantics on the
FinderInfo stream: as long as the client hasn't written a non-zero
FinderInfo blob to the stream, there mustn't be a visible filesystem
entry for other openers.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
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): Thu Nov 1 01:14:23 CET 2018 on sn-devel-144
macOS SMB server doesn't filter out the FinderInfo stream if it has
delete-on-close set.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
fruit_streaminfo currently filters out the FinderInfo stream is
delete-on-close is set. We set it here internally, but the client may
also set it over SMB. Turns out that the macOS SMB server does NOT
filter out FinderInfo stream with delete-on-close set, so we must change
the way filtering is done in fruit_streaminfo.
Filtering is now done based on the FinderInfo stream being 0-bytes large which
is why I'm adding the ftruncate here.
No idea why the tests that check the filtering passed the commits
leading up to this one, but if you revert this commit after applying the
whole patchset, the "delete AFP_AfpInfo by writing all 0" test will fail.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
delete_invalid_meta_stream() is meant to guard against random data being
present in the FinderInfo stream. If the stream size is 0, it's likely a
freshly created stream where no data has been written to yet, so don't
delete it.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This will be required to support using fake fds for the FinderInfo
metadata stream.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
As we'll start returning fake fds in open shortly, we can't rely on the
next module to calculat correct inode numbers for streams and must take
over that responsibility.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
If the read on the stream fails we may have hit a handle on a just
created stream (fio->created=true) with no data written yet.
If that's the case return an empty initialized FinderInfo blob.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This avoid creating files or blobs in our streams backend when a client
creates a stream but hasn't written anything yet. This is the only sane
way to implement the following semantics:
* client 1: create stream "file:foo"
* client 2: open stream "file:foo"
The second operation of client 2 must fail with NT_STATUS_NOT_FOUND.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Not used for now, that comes in the subsequent commits.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13646
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
