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.
Currently the vfs_default fs_capabilities handler calls statvfs
directly, rather than calling the vfs macro. This behaviour may cause
issues for VFS modules that delegate fs_capabilities handling to
vfs_default but offer their own statvfs hook.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13208
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Adding the fs_capabilities() hook to the CephFS VFS module avoids
fallback to the vfs_default code-path, which calls statvfs() against the
share path on the *local* filesystem.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13208
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
GCC on a Ubuntu 16.04 instance said:
[3174/4240] Compiling source3/modules/vfs_cap.c
In file included from ../source3/include/includes.h:301:0,
from ../source3/modules/vfs_fruit.c:20:
../source3/modules/vfs_fruit.c: In function
‘fruit_disk_free’:
../source3/../lib/util/debug.h:217:7: error: ‘bandsize’ may be used
uninitialized in this function [-Werror=maybe-uninitialized]
&& (dbgtext body) )
^
../source3/modules/vfs_fruit.c:6302:9: note: ‘bandsize’ was
declared here
size_t bandsize;
^
[3175/4240] Compiling source3/modules/vfs_expand_msdfs.c
[3176/4240] Compiling source3/modules/vfs_shadow_copy.c
[3177/4240] Compiling source3/modules/vfs_shadow_copy2.c
cc1: all warnings being treated as errors
Waf: Leaving directory
/home/ubuntu/autobuild/b17854/samba-o3/bin'
Build failed: -> task failed (err #1):
{task: cc vfs_fruit.c -> vfs_fruit_25.o}
make: *** [all] Error 1
As far as I can tell, it is wrong, and the bandsize variable never
gets passed uninititalised to DEBUG.
Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Ralph Boehme <slow@samba.org>
We previously removed the stream from the underlying filesystem stream
backing store when the client zeroes out FinderInfo in the AFP_AfpInfo
stream, but this causes certain operations to fail (eg stat) when trying
to access the stream over any file-handle open on that stream.
So instead of deleting, set delete-on-close on the stream. The previous
commit already implemented not to list list streams with delete-on-close
set which is necessary to implemenent correct macOS semantics for this
particular stream.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13181
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Tue Jan 9 17:09:12 CET 2018 on sn-devel-144
This is in preperation of fixing the implementation of removing the
AFP_AfpInfo stream by zeroing the FinderInfo out.
We currently remove the stream blob from the underyling filesystem
backing store, but that results in certain operations to fail on any
still open file-handle.
The fix comes in the next commit which will convert to backing store
delete operation to a set delete-on-close on the stream.
This commit adds filtering on streams that have the delete-on-close
set. It is only needed for the fruit:metadata=stream case, as with
fruit:metadata=netatalk the filtering is already done in
fruit_streaminfo_meta_netatalk().
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13181
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
No change in behaviour, just some refactoring before adding more code to
fruit_streaminfo_meta_stream() in the next commit.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13181
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Mon Jan 8 03:16:30 CET 2018 on sn-devel-144
Based-on-a-patch-by: Ralph Wuerthner <ralph.wuerthner@de.ibm.com>
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 Jan 6 04:41:24 CET 2018 on sn-devel-144
Based-on-a-patch-by: Ralph Wuerthner <ralph.wuerthner@de.ibm.com>
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Enabling fileid:algorithm = fsname_nodirs uses the hostname algorithm
for directories and thus breaks cluster lock coherence for directories.
Based-on-a-patch-by: Christian Ambach <ambi@samba.org>
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Using fileid:algorithm = hostname makes fileid generate
fileids based on the hostname. This breaks cluster lock coherence.
Based-on-a-patch-by: Christian Ambach <ambi@samba.org>
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is in preperation of adding an additional mapping function that
acts differently depending of the file type. No change in behaviour.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
When using the fsname or fsid algorithm a stat() and statfs() call is
required for all mounted file systems to generate the file_id. If e.g.
an NFS file system is unresponsive such a call might block and the smbd
process will become unresponsive. Add "fileid:fstype deny",
"fileid:fstype allow", "fileid:mntdir deny", and "fileid:mntdir allow"
options to ignore potentially unresponsive file systems.
See also https://lists.samba.org/archive/samba-technical/2016-January/111553.html
for a discussion about why this is useful.
Signed-off-by: Ralph Wuerthner <ralph.wuerthner@de.ibm.com>
Reviewed-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This can be used to configure a per client filesystem size limit on
TimeMachine shares.
It's a nasty hack but it was reportedly working well in Netatalk where
it's taken from.
Signed-off-by: Ralph Boehme <slow@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
This module allow injecting errors in vfs calls. It only implements one
case (return ESTALE from chdir), but the idea is to extend this to more
vfs functions and more errors when needed.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13189
Signed-off-by: Christof Schmitt <cs@samba.org>
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Copy the data that the child read into the caller's buffer. This can't
have been used in half a decade at least...
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Since the introduction of the vfs_aio_fork:erratic_testing_mode this
crashed reliably, as we had two different structs behind
SMB_VFS_HANDLE_SET_DATA. I had always believed that due to the fact that
we have specific aio_fork tests in our autobuild, this would have been
tested. But it was not, because the share definition missed the the "aio
read/write size = 1" to actually use the async code in vfs_aio_fork.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
We don't do that in tdb either, and the mmap/memcpy prototypes don't
have it either
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
VFS_ADD_FSP_EXTENSION() now returns a properly typed pointer.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Wed Dec 6 03:43:48 CET 2017 on sn-devel-144
This fixes the problem that conversion failed with
fruit:metadata=stream. Before we were calling ad_set() which stores the
metadata in the Netatalk compatible format.
Rewrite to fully go through the VFS by calling SMB_VFS_CREATE_FILE() and
SMB_VFS_PWRITE().
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13155
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): Wed Nov 29 08:38:06 CET 2017 on sn-devel-144
Add PRINTF_ATTRIBUTE() to do_log(). This removes
picky compiler warning about printf with variable
format string, and adds compiler checks for the format
strings supplied to do_log. This in turn spurred some
warnings which are fixed.
Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
init_search_op is about to be removed from the VFS in
a following commit. In the meantime, removing it poses
no issue because he underlying impementation is a no-op.
Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This VFS function is about to be removed in a following commit.
In the meantime, not handling it by vfs_media_harmony poses no
issue because the underlying implenentation is a no-op.
Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This VFS function is about to be removed. It can be
removed by a separate commit because both the glusterfs
and the default implementations are no-ops.
Signed-off-by: Uri Simchoni <uri@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Triggered by https://bugzilla.samba.org/show_bug.cgi?id=13128 I think
this module should go. Once Linux aio will do what Samba needs, this
might be worth another look.
What we should instead do soon is support Linux preadv2 and the
RWF_NOWAIT flag to avoid the thread context switch whenever possible.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13125
The glfs.h header file has always resided under glusterfs/api/ in the
standard include directory. The glusterfs-api.pc file adds the unneeded
-I${includedir}/glusterfs compiler option. This option will be removed
from future versions of the pkg-config file.
This change can safely be backported to older versions if there is a
need to have them build against glusterfs-3.13 or newer.
URL: https://review.gluster.org/18576
CC: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Niels de Vos <ndevos@redhat.com>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Thu Nov 9 22:37:30 CET 2017 on sn-devel-144
Add a NFS4 ACL backend that stores the ACL blob in an XDR encoded xattr,
by default in "security.nfs4acl_xdr".
This backend is enabled by setting "nfs4acl_xattr:encoding = xdr" in a
share definition.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This allows take-ownership to work if the user has SEC_STD_WRITE_OWNER.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The vfs_nfs4acl_xattr VFS module is supposed to work the same as
vfs_acl_xattr|tdb with "ignore system acls" set to true. That is,
filesystem permissions should never restrict access and the actual
access checks are done by smbd in userspace.
To better cope with POSIX mode changes via other protocols (eg NFS) or
local filesystem access, add the following tweaks:
o validate ACL blob: if POSIX mode is not 0777/0666 discard the ACL blob
from the xattr and synthesize a default ACL based on the POSIX mode
o when setting an ACL, check and reset POSIX mode to 0777/0666
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This ensures we can always fetch the ACL xattr blob when we wanted,
unrestricted of filesystem permissions or Linux xattr security namespace
restrictions.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is in preperation of modularizing the storage backend. Currently we
store the NFS4 ACL as an IDL/NDR encoded blob in a xattr.
Later commits will add a different backend storing the NFS4 ACL as an
XDR encoded blob in a xattr.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We only open the underlying file if the open access mode contains
FILE_READ_DATA|FILE_WRITE_DATA|FILE_APPEND_DATA|FILE_EXECUTE
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Refactor the code in preperation of factoring out ACL blob to smb4acl
and vice versa mapping functions.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
README.Coding adjustments, DEBUG macro modernisation, variable name
sanitizing. No change in behaviour.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This changes the way ACL inheritance is achieved in this
module.
Previously the module recursed to the next parent directory until the
share root was reached or a directory with an ACL xattr. If the share
root didn't contain an ACL xattr either a default ACL would be used.
This commit removed this recursive scanning and replaces it with the
same mechanism used by vfs_acl_xattr: by setting "inherit acls = yes"
just let smbd do the heavy lefting and inheritance.
For any file without ACL xattr we still synthesize a default ACL,
leveraging the existing default ACL function used by vfs_acl_xattr.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
No change in behaviour, all option defaults are set to the original
behaviour.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Move interesting functions to the top of the vfs_fn_pointers struct, no
change in behaviour.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
