Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit afa15e6128473d3e4006f7cdc3762ab4c1cba05a)
Autobuild-User(v4-5-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-5-test): Tue May 9 14:34:10 CEST 2017 on sn-devel-144
Signed-off-by: Ralph Boehme <slow@samba.org> Signed-off-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Sat May 6 22:58:47 CEST 2017 on sn-devel-144
Ralph Boehme [Wed, 19 Apr 2017 11:12:55 +0000 (13:12 +0200)]
vfs_fruit: lp_case_sensitive() does not return a bool
lp_case_sensitive() returns an int, not a bool, so with the default
setting of "Auto" by default we set the AAPL flag
SMB2_CRTCTX_AAPL_CASE_SENSITIVE.
This caused the client to believe the volume is case sensitive where it
wasn't, leading to an error when trying to rename files changing only
the case of the name.
Also fix the existing torture test that verifies AAPL context
negotiation and actually expected the server to return "case sensitive",
while the Samba default is really "case insensitive".
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Uri Simchoni <uri@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Thu Apr 20 20:50:10 CEST 2017 on sn-devel-144
In notifyd_trigger_parser() while initializing notify_event_msg values from
notify_trigger_msg, 'when' value is ignored. So the smbd process does not get
correct 'when' value and this is causing issues during qsort in
notify_marshall_changes(). Because of this issue, smb2.notify.dir test was
failing.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=12756 Signed-off-by: Shilpa Krishnareddy <skrishnareddy@panzura.com> Reviewed-by: Volker Lendecke <vl@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Wed Apr 26 17:02:58 CEST 2017 on sn-devel-144
The last 3 patches address
BUG: https://bugzilla.samba.org/show_bug.cgi?id=12743
vfs_shadow_copy2 fails to list snapshots from shares with GlusterFS backend
pam_winbind: no longer use wbcUserPasswordPolicyInfo when authenticating
The expiry time for the specific user comes from
info->pass_must_change_time and nothing else.
The authenticating DC knows which password policy applies
to the user, that's nothing the client can do, as
domain trusts and fine-grained password policies makes
this a very complex task.
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Tue Apr 18 15:43:02 CEST 2017 on sn-devel-144
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 60af864f751706c48b8af448700bf06e33e45946)
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 600f8787e3b605c9f3e8f724c726e63157ee9efc)
Uri Simchoni [Thu, 13 Apr 2017 09:44:58 +0000 (12:44 +0300)]
vfs_acl_xattr: avoid needlessly supplying a large buffer to getxattr()
When obtaining the security descriptor via getxattr(), first try
optimistically to supply a buffer of 4K, and if that turns out
to be too small, determine the correct buffer size.
The previous behavior of falling back to a 64K buffer encountered
problem with Linux prior to version 3.6, due to pyisical memory
fragmentation. With those kernels, as long as the buffer is 8K or
smaller, getting the xattr is much less prone to failure due to
memory fragmentation.
Signed-off-by: Uri Simchoni <uri@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Apr 18 04:41:16 CEST 2017 on sn-devel-144
Ralph Boehme [Wed, 29 Mar 2017 09:13:46 +0000 (11:13 +0200)]
winbindd: trigger possible passdb_dsdb initialisation
If the passdb backend is passdb_dsdb the domain SID comes from dsdb, not
from secrets.tdb. As we use the domain SID in various places, we must
ensure the domain SID is migrated from dsdb to secrets.tdb before
get_global_sam_sid() is called the first time.
The migration is done as part of the passdb_dsdb initialisation, calling
pdb_get_domain_info() triggers it.
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 Apr 1 21:18:59 CEST 2017 on sn-devel-144
Ralph Boehme [Fri, 31 Mar 2017 14:06:18 +0000 (16:06 +0200)]
selftest: fix for wbinfo -s tests for wellknown SIDs
Rework while loop to not use a pipe as that uses a subshell for the loop
which means assigning to the variable failed is not visible in the
main script.
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit d8fd56a8244a3010469c27eaa3b73a2c5fbbc41f)
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 167bb5ead8c7193d173fdba8a453279d422fa7ea)
Ralph Boehme [Fri, 31 Mar 2017 14:24:05 +0000 (16:24 +0200)]
selftest: wbinfo --sids-to-unix-ids tests for wellknown SIDs
This test passes even without the fix, as in sids2xids we use the
lookupnames just to determine the mapping domain, using the default
idmap domain as fallback if that fails.
Ralph Boehme [Thu, 30 Mar 2017 21:41:59 +0000 (23:41 +0200)]
winbindd: use passdb backend for well-known SIDs
On a DC well-known SIDs like S-1-1-0 (everyone) *must* be handled by the
local domain, otherwise something simple like this fails with
WBC_ERR_DOMAIN_NOT_FOUND:
$ make testenv SELFTEST_TESTENV=nt4_dc SCREEN=1
localnt4dc2$ ./bin/wbinfo --sid-to-name S-1-1-0
failed to call wbcLookupSid: WBC_ERR_DOMAIN_NOT_FOUND
Could not lookup sid S-1-1-0
On a member server asking our DC works and is what we're currently
doing, but changing it to ask passdb avoids the overhead.
Ralph Boehme [Tue, 7 Feb 2017 06:44:40 +0000 (07:44 +0100)]
vfs_fruit: resource fork open request with flags=O_CREAT|O_RDONLY
When receiving an SMB create request with read-only access mode and
open_if disposition, we end of calling the open() function with
flags=O_CREAT|O_RDONLY for the ._ AppleDouble file.
If the file doesn't exist, ie there's currently no rsrc stream, we create
it but then we fail to write the AppleDouble header into the file due to
the O_RDONLY open mode, leaving a 0 byte size ._ file.
Running this create requests against macOS SMB server yields an
interesting result: it returns NT_STATUS_OBJECT_NAME_NOT_FOUND even
though create dispotion is open_if. Another instance where the macOS SMB
server just exposes FSA behaviour (ie HFS+) and we have to adapt to be
compatible.
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Jim McDonough <jmcd@samba.org>
(cherry picked from commit d8a5565ae647352d11d622bd4e73ff4568678a7c)
Christof Schmitt [Mon, 27 Mar 2017 22:11:08 +0000 (15:11 -0700)]
winbindd: Fix password policy for pam authentication
Authenticating users from trusted domains would return the password
policy of the joined domain. Fix the code so that the password policy of
the joined domain is only returned for users from that domain.
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Fri Apr 7 00:05:02 CEST 2017 on sn-devel-144
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit ef10b43469f5b31a696259a70b3e116a350bfd3d)
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Ralph Boehme <slow@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 9671811da8ad3f91ba7bb0fa868f806bc5afe863)
Ralph Boehme [Tue, 4 Apr 2017 12:21:25 +0000 (14:21 +0200)]
winbindd: use correct domain name for failed lookupsids
What we want here is, for failed lookupsids, pass the domain name of the
SID we were trying to lookup to the idmap backend.
But as a domain member, using
state->single_domains[state->single_sids_done]
for this purpose will always be use our primary domain name (for S-1-5-21
SIDs that are not in our local SAM).
So for now use find_domain_from_sid_noinit() to find the domain from the
domain list. This can be removed when we switch idmap backend
determination to be based on domain SIDs, not names.
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Ralph Boehme <slow@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit a684df160e692710e011c4eb6795a66772025c23)
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): Tue Mar 28 17:05:27 CEST 2017 on sn-devel-144
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): Tue Mar 28 17:05:27 CEST 2017 on sn-devel-144
Signed-off-by: Uri Simchoni <uri@samba.org> Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Sun Mar 26 23:31:08 CEST 2017 on sn-devel-144
Uri Simchoni [Thu, 23 Mar 2017 12:08:26 +0000 (14:08 +0200)]
smbd: add zero_file_id flag
This flag instructs the SMB layer to report a zero on-disk
file identifier.
According to [MS-SMB2] 3.3.5.9.9, the reported on-disk file ID
SHOULD be unique. However, macOS clients seem to expect it to be
unique over time as well, like the HFS+ CNID. Reporting a file ID
of 0 seems to instruct the Mac client not to trust the server-reported
file ID.
Signed-off-by: Tom Schulz <schulz@adi.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 59229276bcf5e2b7fa0ddf3ceb6fd3adccc01f9a)
During revoking readonly delegations, if one of the nodes disappears, then
there is no point re-trying revoking readonly delegation. The database
needs to be recovered before the revoke operation can succeed. So retry
only after a grace period.
Signed-off-by: Amitay Isaacs <amitay@gmail.com> Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Fri Mar 17 14:05:57 CET 2017 on sn-devel-144
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 Mar 15 20:04:32 CET 2017 on sn-devel-144
In the GNU C Library, "makedev" is defined by <sys/sysmacros.h>. For
historical compatibility, it is currently defined by <sys/types.h> as
well, but it is planned to remove this soon.
Ralph Boehme [Tue, 7 Mar 2017 15:27:39 +0000 (16:27 +0100)]
s3/smbd: fix deferred open with streams and kernel oplocks
I noticed smbd can get stuck in an open() call with kernel oplocks
enabled and named streams (provided by vfs_streams_xattr):
- client opens a file and with an exclusive oplock
- client starts writing to the file
- client opens an existing stream of the file
- the smbd process gets stuck in an open()
What happens is:
we had setup a locking.tdb record watch in defer_open(), the watch was
triggered, we reattempted the open and got stuck in a blocking open
because the oplock holder (ourselves) hadn't given up the oplock yet.
Cf e576bf5310bc9de9686a71539e9a1b60b4fba5cc for the commit that added
the kernel oplock retry logic. tldr: with kernel oplocks the first open
is non-blocking, but the second one is blocking.
Detailed analysis follows.
When opening a named stream of a file, Samba internally opens the
underlying "base" file first. This internal open of the basefile suceeds
and does *not* trigger an oplock break (because it is an internal open
that doesn't call open() at all) but it is added as an entry to the
locking.tdb record of the file.
Next, the stream open ends up in streams_xattr where a non-blocking
open() on the base file is called. This open fails with EWOULDBLOCK
because we have another fd with a kernel oplock on the file.
So we call defer_open() which sets up a watch on the locking.tdb record.
In the subsequent error unwinding code in open_file_ntcreate() and
callers we close the internal open file handle of the basefile which
also removes the entry from the locking.tdb record and so *changes the
record*.
This fires the record watch and in the callback defer_open_done() we
don't check whether the condition (oplock gone) we're interested in is
actually met. The callback blindly reschedules the open request with
schedule_deferred_open_message_smb().
schedule_deferred_open_message_smb() schedules an immediate tevent event
which has precedence over the IPC fd events in messaging, so the open is
always (!) reattempted before processing the oplock break message.
As explained above, this second open will be a blocking one so we get
stuck in a blocking open.
It doesn't help to make all opens non-blocking, that would just result
in a busy loop failing the open, as we never process the oplock break
message (remember, schedule_deferred_open_message_smb() used immediate
tevent events).
To fix this we must add some logic to the record watch callback to check
whether the record watch was done for a kernel oplock file and if yes,
check if the oplock state changed. If not, simply reschedule the
deferred open and keep waiting.
This logic is only needed for kernel oplocks, not for Samba-level
oplocks, because there's no risk of deadlocking, the worst that can
happen is a rescheduled open that fails again in the oplock checks and
gets deferred again.
Just roll our own deferred record instead of calling defer_open() and
also set up timer that, as a last resort, catches stuck opens and just
exits for now.
projects / thirdparty / samba.git / log
