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.
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): Fri Nov 9 03:49:55 CET 2018 on sn-devel-144
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Nov 8 11:03:11 CET 2018 on sn-devel-144
Increase the default number of worker processes started by the pre-fork
process model from 1 to 4.
Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
This solves the problem that gencache never shrinks right
now. Whenever we write an entry, we now walk that entry's chain and
delete expired entries. This should be a good balance between
performance and cleanup actions: Reading is still unaffected, and
those who write pay a small penalty while keeping gencache size under
control.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
At more than one large site I've seen significant problems due to
gencache_stabilize. gencache_stabilize was mainly introduced to
survive machine crashes with the cache still being in place. Given
that most installations crash rarely and this is still a cache, this
safety is overkill and causes real problems.
With the recent changes to tdb, we should be safe enough to run on
completely corrupted databases and properly detect errors. A further
commit will introduce code that wipes the gencache.tdb if such a
corruption is detected.
There is one kind of corruption that we don't properly handle:
Orphaned space in the database. I don't have a good idea yet how to
handle this in a graceful and efficient way during normal operations,
but maybe this idea pops up at some point.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This covers key, timestamp and data. This will detect silent
corruption of gencache data after a system crash
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Two reasons:
The ascii conversion shows up on profiles.
In a further commit we will get checksums for gencache entries to
protect at hidden corruption due to a crash on the non-transactioned
gencache.tdb. Next to the timestamp this is a second field that is
gencache metadata, and I don't want to deal with a second ascii number
when at least some of the gencache values are binary already.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13673
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Christof Schmitt <cs@samba.org>
Autobuild-Date(master): Sat Nov 3 05:55:45 CET 2018 on sn-devel-144
MS-FSA states that a CREATE with FILE_DELETE_ON_CLOSE on an existing
file with READ_ONLY attribute has to return STATUS_CANNOT_DELETE. This
was missing in smbd as the check used the DOS attributes from the CREATE
instead of the DOS attributes on the existing file.
We need to handle the new file and existing file cases separately.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13673
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is the one place where we have to do another strpcy. Many of the
sid_to_fstring calls should be replacable by dom_sid_str_buf, so this
will get less.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This was used internally to mark an aio request as cancelled. As the aio
cancellation functionality has been removed, we can now also remove this
check.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13667
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
As we currently don't attempt to cancel the internal aio request, we
must ignore the SMB2 cancel request and continue to process the SMB2
request, cf MS-SM2 3.3.5.16:
If the target request is not successfully canceled, processing of the
target request MUST continue and no response is sent to the cancel
request.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13667
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
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 Nov 1 05:06:23 CET 2018 on sn-devel-144
The group enumeration backend functions try to allocate an array of
wb_acct_info structs with a number of elements equal to the number of
groups. In domains with a large number of groups this allocation may
fail due to the size of the chunk.
Found while trying to enumerate the groups in a domain with more than
700k groups.
Signed-off-by: Samuel Cabrero <scabrero@suse.de>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
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>
fio->created tracks whether a create created a 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>
Directly unlinking a file with open handles is not good, don't do 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>
Aids in debugging dev/ino mismatch failures in open_file_ntcreate.
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>
First step in achieving macOS compliant behaviour wrt to empty streams:
- hide empty streams in streaminfo
- prevent opens of empty streams
This means that we may carry 0-byte sized streams in our streams
backend, but this shouldn't really hurt.
The previous attempt of deleting the streams when an SMB setinfo eof to
0 request came in, turned out be a road into desaster.
We could set delete-on-close on the stream, but that means we'd have to
check for it for every write on a stream and checking the
delete-on-close bits requires fetching the locking.tdb record, so this
is expensive and I'd like to avoid that overhead.
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>
Ensure any non MS compliant protocol behaviour targetted at supporting
macOS clients are only effective if the client negotiated AAPL.
Currently this only guards the resource fork which only macOS client are
going to use, but subsequent commits add more this at this place.
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 caused all sort of havoc with subsequent SMB request that acted on
the handle of the then deleted backend storage (file or blob, depending
on the used streams module).
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>
macOS SMB server versions supports this since 10.12, so we adapt our
behaviour.
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>
Since a long time the modules's open function happily returned success
when opening a non existent stream without O_CREAT.
This change fixes it to return -1 and errno=ENOATTR if
o get_ea_value() returns NT_STATUS_NOT_FOUND (eg mapped from
getxattr() = -1, errno=ENOATTR) and
o flags doesn't contain O_CREAT
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>
