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.
The fsp has a valid file descriptor when this function is called. No
need for the additional check.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The VFS flock function is called with a zero access_mask to indicate
that the share mode should be cleared. Adjust the check to directly
check the access_mask and provide a better message to indicate this
case.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The calling code already ensures that the file is open with a valid file
descriptor.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
MS-SMB2 and the smbd code refer to this field as share_access. Use the
same name in the function argument.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This implements two core changes:
* use NTTIME instead of struct timespec at the database layer
* use struct timespec { .tv_nsec = SAMBA_UTIME_OMIT } as special sentinel
value in smbd when processing timestamps
Using NTTIME at the database layer is only done to avoid storing the special
struct timespec sentinel values on disk. Instead, with NTTIME the sentinel value
for an "unset" timestamp is just 0 on-disk.
The NTTIME value of 0 gets translated by nt_time_to_full_timespec() to the
struct timespec sentinel value { .tv_nsec = SAMBA_UTIME_OMIT }.
The function is_omit_timespec() can be used to check this.
Beside nt_time_to_full_timespec(), there are various other new time conversion
functions with *full* in their name that can be used to safely convert between
different types with the changed sentinel value.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=7771
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Make the code a bit easier to read by avoiding direct access to the
statex struct.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Christof Schmitt <cs@samba.org>
Autobuild-Date(master): Tue Nov 19 22:44:53 UTC 2019 on sn-devel-184
MacOS SMB clients require that file ids are not quickly reused when
files are deleted and new files are created with the same name. Inode
numbers do not satisfy that requirement, as they will be quickly reused.
To address this problem, create a unique id from the available file
system specific metadata. As that id is larger than the available 64bit,
use a hash to generate a 64bit id for usage as fileid.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
This is no functional change, but allows to use additional metadata
later on.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
This is no functional change, but allows to use additional metadata
later on.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
This simplifies (and corrects) things in the VFS as there the caller is only
interested in whether a name is pointing to a real named stream most of the times.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
On directories with the "set group id" (SGID) bit is set, new files and
subfolders will be created with the group of the directory, and not with
the primary group of the user. Checking for free space in this case
should query the group quota for the gid of the directory.
This is the same change as the common smbd code, but since vfs_gpfs
still has its own check for available space and quotas, add it here as
well. A future goal would be to use the common free space check instead
of duplicating code in vfs_gpfs.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
This was in vfs_gpfs.c from day one, no idea why both "share_mode" and
"fsp->share_access" were used as if they were two different things.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
In GPFS, it is not possible to deny ACL or attribute access through a
SPECIAL_OWNER entry. The best that can be done is mapping this to a
named user entry, as this one can at least be stored in an ACL. The same
cannot be done for inheriting SPECIAL_OWNER entries, as these represent
CREATOR OWNER entries, and the limitation of not being able to deny
owner access to ACL or attributes remains.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14032
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
This is not functional change. It cleans up the code a bit and makes
expanding this codepath in a later patch easier.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14032
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
All supported GPFS versions now support setting WRITE and APPEND in the
ACLs independently. Remove this now unused parameter to simplify the
code.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14032
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Returning -1 in a function with bool as return value type is the same
as returning true. Change to false to indicate the error.
Detected by the help of cppcheck.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Reviewed-by: Ralph Böhme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The core smbd code implements ZERO_DATA for non-sparse files by punching
a hole and filling it again with a fallocate(FL_KEEP_SIZE) call. As GPFS
does not provide the fallocate(FL_KEEP_SIZE) call and non-sparse files
should not contain holes, block the punchhole; effectively only allowing
ZERO_DATA/punchhole calls for sparse files.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Apr 23 00:33:03 UTC 2019 on sn-devel-144
All supported versions of GPFS now support fallocate. Use the default
codepath instead of the API call. Keep the function stub as it will
be used for a check later.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The vfs_gpfs modules uses GPFS API calls that only succeed when using
the module with the GPFS file system. Add an explicit statfs check for
the file system type on connect, to make it obvious when the file system
is missing or not mounted. The check can be skipped by setting
gpfs:check_fstype to 'no'.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Finally: use the new dos_mode_at_send() in the directory enumeration
loop. This means that fetching the DOS attributes for directory entries
is done asynchronously with regard to the enumeration loop.
As the DOS attribute is typically read from an extended attribute in the
filesytem, this avoids sequentially blocking on IO. If the IO subsystem
is slow servicing these request, enabling async processing can result in
performance improvements.
A parametric option
smbd:async dosmode = true | false (default: false)
can be used to enable the new async processing.
Simulating slow IO with usleep(5000) in the synchronous and asynchronous
versions of SMB_VFS_GET_DOS_ATTRIBUTES(), the results of enumerating a
directory with 10,000 files are:
smbd:async dosmode = no:
$ time bin/smbclient -U slow%x //localhost/test -c "ls dir\*" > /dev/null
real 0m59.597s
user 0m0.024s
sys 0m0.012s
smbd:async dosmode = yes:
$ time bin/smbclient -U slow%x //localhost/test -c "ls dir\*" > /dev/null
real 0m0.698s
user 0m0.038s
sys 0m0.025s
Performance gains in real world workloads depends on whether the actual
IO requests can be merged and parallelized by the kernel. Without such
wins at the IO layer, the async processing may even be slower then the
sync processing due to the additional overhead.
The following parameters can be used to adapt async processing behaviour
for specific workloads and systems:
aio max threads = X (default: 100)
smbd:max async dosmode = Y (default: "aio max threads" * 2)
By default we have at most twice the number of async requests in flight
as threads provided by the underlying threadpool. This ensures a worker
thread that finishes a job can directly pick up a new one without going
to sleep.
It may be advisable to reduce the number of threads to avoid scheduling
overhead while also increasing "smbd:max async dosmode".
Note that we disable async processing for certain VFS modules in the VFS
connect function to avoid the overhead of triggering the sync fallback
in dos_mode_at_send(). This is done for VFS modules that implement the
sync SMB_VFS_GET_DOS_ATTRIBUTES(), but not the async version (gpfs), and
for VFS modules that don't share a real filesystem where fchdir() can be
used (ceph, gluster). It is disabled for catia, because we realized that
the catia name translation macros used on
fsps (CATIA_FETCH_FSP_[PRE|POST]_NEXT) have a bug (#13547).
We use threadpool = smb_vfs_ev_glue_tp_chdir_safe() and then
pthreadpool_tevent_max_threads(threadpool) to get the number of maximum
worker threads which matches the pool used by the low level
SMB_VFS_GETXATTRAT_[SEND|RECV] implementation in vfs_default.
This is a terrible abstraction leak that should be removed in the future
by maybe making it possible to ask a VFS function which threadpool it
uses, internally suporting chaining so VFS function FOO that internally
uses BAR can forward the question to BAR.
On a hyphotetical system that had a getxattrat(dirfd, path, ...)
syscall and at the same time doesn't support per-thread current working
directories (eg FreeBSD doesn't have the latter) but has support for
per-thread-credentials, pthreadpool_tevent_max_threads() on the
tp_chdir_safe threadpool returns 1.
So when hooking the hyphotetical getxattrat() into the async
SMB_VFS_GETXATTRAT_[SEND|RECV] implementation in an VFS module, the
implementation could use the tp_path_safe threadpool, but the SMB2
layer would use the wrong threadpool in the call to
pthreadpool_tevent_max_threads(), resulting in no parallelism.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
full_path_tos avoids the talloc call for most cases; use that instead of
talloc_asprintf.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat May 26 01:43:09 CEST 2018 on sn-devel-144
Updating the timestamps requires the path to the file, but no stream
information.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is returned for a get_winattrs call against a non-gpfs file system.
This can happen for the .. entry when listing a share on the file system
root.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Mon Dec 18 13:32:00 CET 2017 on sn-devel-144
A SMB client can deny the WRITE sharemode, but still grant the DELETE
sharemode. Mapping the requested DELETE access to WRITE access breaks
this case. Fix this by removing the incorrect mapping from DELETE access
to WRITE access.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
When trying to fetch the DOS attributes via gpfswrap_get_winattrs_path()
if the filesystem doesn't grant READ_ATTR to the file the function fails
with EACCESS.
But according to MS-FSA 2.1.5.1.2.1 "Algorithm to Check Access to an
Existing File" FILE_LIST_DIRECTORY on a directory implies
FILE_READ_ATTRIBUTES for directory entries.
So if the user can open the parent directory for reading this implies
FILE_LIST_DIRECTORY and we can safely call gpfswrap_get_winattrs_path()
with DAC_OVERRIDE_CAPABILITY.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=12944
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Wed Aug 9 01:21:14 CEST 2017 on sn-devel-144
We need to migrate all pathname based VFS calls to use a struct
to finish modernising the VFS with extra timestamp and flags parameters.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Richard Sharpe <realrichardsharpe@gmail.com>
We need to migrate all pathname based VFS calls to use a struct
to finish modernising the VFS with extra timestamp and flags parameters.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Richard Sharpe <realrichardsharpe@gmail.com>
../source3/modules/vfs_gpfs.c: In function ‘gpfsacl_sys_acl_set_fd’:
../source3/modules/vfs_gpfs.c:1280:6: error: passing argument 2 of ‘gpfsacl_sys_acl_set_file’ from incompatible pointer type [-Werror]
SMB_ACL_TYPE_ACCESS, theacl);
^
../source3/modules/vfs_gpfs.c:1235:12: note: expected ‘const struct smb_filename *’ but argument is of type ‘char *’
static int gpfsacl_sys_acl_set_file(vfs_handle_struct *handle,
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Jun 3 05:52:32 CEST 2017 on sn-devel-144
We need to migrate all pathname based VFS calls to use a struct
to finish modernising the VFS with extra timestamp and flags parameters.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
We need to migrate all pathname based VFS calls to use a struct
to finish modernising the VFS with extra timestamp and flags parameters.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
We need to migrate all pathname based VFS calls to use a struct
to finish modernising the VFS with extra timestamp and flags parameters.
Requires a few extra cleanups in calling code.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
We need to migrate all pathname based VFS calls to use a struct
to finish modernising the VFS with extra timestamp and flags parameters.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Not currently used - no logic changes inside.
This will make it possible to pass down a long-lived talloc
context from the loading function for modules to use instead
of having them internally all use talloc_autofree_context()
which is a hidden global.
Updated all known module interface numbers, and added a
WHATSNEW.
Signed-off-by: Jeremy Allison <jra@samba.org>
Signed-off-by: Ralph Böhme <slow@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Apr 22 01:17:00 CEST 2017 on sn-devel-144
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Sat Dec 17 12:58:07 CET 2016 on sn-devel-144
This is now handled by the vfs_gpfs_(f)get_dos_attributes. Getting rid
of this in the stat VFS functions is a huge performance saver. perf
report found that in a kernel copy workload smbd was spending
considerable CPU time in vfs_gpfs_(f|l)stat -> gpfs_get_winattrs.
Most of the time the VFS stat caller is not interested in the btime. The
SMB frontend processing around btime is designed to fetch btime together
with DOS attributes via dos_mode() in all places that need these
attributes. That's the way it is implemented in the default VFS module
and that's what vfs_gpfs now does as well for performance reasons.
This makes vfs_gpfs_fstat a null op and I'm therefor removing it.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>
This paves the way for removing btime updates from the stat VFS
functions.
This way we behave like the default VFS module where DOS attributes and
btime are fetched from the same backing store and the frontend is
designed around using dos_mode() -> SMB_VFS_GET_ATTRIBUTES to update
both attributes as necessary in the SMB processing.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>
The patchset `git log -2 3031815f982e365be50148564d47d7d5afab46e0`
missed a change to vfs_gpfs_is_offline() which is now merely a helper
function that returns true or false and mustn't call into the VFS.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Mon Oct 17 21:28:12 CEST 2016 on sn-devel-144
The previous commit removed all callers of this, so lets remove it.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Tue Oct 11 14:44:03 CEST 2016 on sn-devel-144
