gd/samba-autobuild/.git
10 months agonsswitch: use goto to have only one function return
Ralph Wuerthner [Tue, 2 Oct 2018 08:58:12 +0000 (10:58 +0200)]
nsswitch: use goto to have only one function return

Signed-off-by: Ralph Wuerthner <ralph.wuerthner@de.ibm.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos3: winbind: Remove fstring from wb_acct_info struct
Samuel Cabrero [Tue, 30 Oct 2018 17:47:16 +0000 (18:47 +0100)]
s3: winbind: Remove fstring from wb_acct_info struct

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>
10 months agovfs_fruit: let fruit_open_meta() with O_CREAT return a fake-fd
Ralph Boehme [Wed, 22 Aug 2018 13:25:26 +0000 (15:25 +0200)]
vfs_fruit: let fruit_open_meta() with O_CREAT return a fake-fd

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

10 months agovfs_fruit: don't check for delete-on-close on the FinderInfo stream
Ralph Boehme [Sat, 20 Oct 2018 21:50:32 +0000 (23:50 +0200)]
vfs_fruit: don't check for delete-on-close on the FinderInfo stream

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>
10 months agovfs_fruit: let fruit_pwrite_meta_stream also ftruncate empty FinderInfo
Ralph Boehme [Sat, 20 Oct 2018 21:40:14 +0000 (23:40 +0200)]
vfs_fruit: let fruit_pwrite_meta_stream also ftruncate empty FinderInfo

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>
10 months agovfs_fruit: pass stream size to delete_invalid_meta_stream()
Ralph Boehme [Sat, 20 Oct 2018 21:46:43 +0000 (23:46 +0200)]
vfs_fruit: pass stream size to delete_invalid_meta_stream()

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>
10 months agovfs_fruit: let fruit handle all aio on the FinderInfo metadata stream
Ralph Boehme [Wed, 17 Oct 2018 14:51:34 +0000 (16:51 +0200)]
vfs_fruit: let fruit handle all aio on the FinderInfo metadata stream

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>
10 months agovfs_fruit: do ino calculation
Ralph Boehme [Wed, 22 Aug 2018 14:49:23 +0000 (16:49 +0200)]
vfs_fruit: do ino calculation

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>
10 months agovfs_fruit: prepare fruit_pread_meta() for reading on fake-fd
Ralph Boehme [Wed, 22 Aug 2018 13:22:08 +0000 (15:22 +0200)]
vfs_fruit: prepare fruit_pread_meta() for reading on fake-fd

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>
10 months agovfs_fruit: prepare fruit_pwrite_meta() for on-demand opening and writing
Ralph Boehme [Wed, 22 Aug 2018 13:21:08 +0000 (15:21 +0200)]
vfs_fruit: prepare fruit_pwrite_meta() for on-demand opening and writing

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>
10 months agovfs_fruit: prepare struct fio for fake-fd and on-demand opening
Ralph Boehme [Wed, 22 Aug 2018 13:22:57 +0000 (15:22 +0200)]
vfs_fruit: prepare struct fio for fake-fd and on-demand opening

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>
10 months agovfs_fruit: add fio->created
Ralph Boehme [Mon, 22 Oct 2018 14:56:46 +0000 (16:56 +0200)]
vfs_fruit: add fio->created

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>
10 months agovfs_fruit: remove resource fork special casing
Ralph Boehme [Mon, 15 Oct 2018 16:38:33 +0000 (18:38 +0200)]
vfs_fruit: remove resource fork special casing

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>
10 months agovfs_fruit: add some debugging of dev/ino
Ralph Boehme [Wed, 17 Oct 2018 17:07:11 +0000 (19:07 +0200)]
vfs_fruit: add some debugging of dev/ino

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>
10 months agos4:torture/vfs/fruit: add test "empty_stream"
Ralph Boehme [Mon, 22 Oct 2018 14:21:21 +0000 (16:21 +0200)]
s4:torture/vfs/fruit: add test "empty_stream"

One to rule them all: consistently test critical operations on all
streams relevant to macOS clients: the FinderInfo stream, the Resource
Fork stream and an arbitrary stream that macOS maps to xattrs when
written to on a macOS SMB server.

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>
10 months agos4:torture/vfs/fruit: add check_stream_list_handle()
Ralph Boehme [Thu, 11 Oct 2018 15:14:50 +0000 (17:14 +0200)]
s4:torture/vfs/fruit: add check_stream_list_handle()

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>
10 months agos4:torture/util: add torture_smb2_open()
Ralph Boehme [Wed, 10 Oct 2018 16:45:56 +0000 (18:45 +0200)]
s4:torture/util: add torture_smb2_open()

This seems to be missing: a simple wrapper to just open a file without
fancy options.

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>
10 months agovfs_fruit: filter empty streams
Ralph Boehme [Sat, 20 Oct 2018 12:53:50 +0000 (14:53 +0200)]
vfs_fruit: filter empty streams

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>
10 months agovfs_fruit: use check on global_fruit_config.nego_aapl for macOS specific behaviour
Ralph Boehme [Sat, 20 Oct 2018 13:28:06 +0000 (15:28 +0200)]
vfs_fruit: use check on global_fruit_config.nego_aapl for macOS specific behaviour

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>
10 months agos4:torture/vfs/fruit: enable AAPL extensions in a bunch of tests
Ralph Boehme [Mon, 22 Oct 2018 12:01:34 +0000 (14:01 +0200)]
s4:torture/vfs/fruit: enable AAPL extensions in a bunch of tests

These tests check for macOS SMB server specific behaviour. They work
currently against Samba without enabling AAPL because in vfs_fruit we're
currently don't check whether AAPL has been negotiated in one place. A
subsequent commit will change that and this commit prepares for that
change.

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>
10 months agovfs_fruit: don't unlink 0-byte size truncated streams
Ralph Boehme [Sat, 20 Oct 2018 12:54:48 +0000 (14:54 +0200)]
vfs_fruit: don't unlink 0-byte size truncated streams

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>
10 months agos4:torture/vfs/fruit: write some data to a just created teststream
Ralph Boehme [Sat, 20 Oct 2018 12:52:23 +0000 (14:52 +0200)]
s4:torture/vfs/fruit: write some data to a just created teststream

Doesn't currently make a difference, but this prepares for a later
change in vfs_fruit that will filter out empty streams (which is the
macOS 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>
10 months agos4:torture/vfs/fruit: expand test "setinfo eof stream"
Ralph Boehme [Mon, 22 Oct 2018 10:43:16 +0000 (12:43 +0200)]
s4:torture/vfs/fruit: expand test "setinfo eof stream"

o Adds checks verifying that after setting eof to 0 on a stream, a
  subsequent open gets ENOENT, before and after closing the handle that
  had been used to set eof to 0.

o Verify that a write to a handle succeeds after that handle has been
  used to set eof to 0 on 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>
10 months agovfs_fruit: update handling of read-only creation of resource fork
Ralph Boehme [Mon, 22 Oct 2018 10:32:09 +0000 (12:32 +0200)]
vfs_fruit: update handling of read-only creation of resource fork

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>
10 months agos4:torture/vfs/fruit: update test "creating rsrc with read-only access" for newer...
Ralph Boehme [Mon, 15 Oct 2018 13:17:08 +0000 (15:17 +0200)]
s4:torture/vfs/fruit: update test "creating rsrc with read-only access" for newer macOS versions

While this operation failed against older macOS versions, it passes
against versions 10.12 and newer. Update the test accordingly, a
subsequent commit will then update our implementation.

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>
10 months agos4:torture/vfs/fruit: expand existing vfs_test "null afpinfo"
Ralph Boehme [Wed, 10 Oct 2018 10:47:07 +0000 (12:47 +0200)]
s4:torture/vfs/fruit: expand existing vfs_test "null afpinfo"

This adds a check that a read on a seperate handle also sees the
previously created AFP_AfpInfo 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>
10 months agos4:torture/vfs/fruit: expand existing test "setinfo delete-on-close AFP_AfpInfo"...
Ralph Boehme [Tue, 9 Oct 2018 16:48:08 +0000 (18:48 +0200)]
s4:torture/vfs/fruit: expand existing test "setinfo delete-on-close AFP_AfpInfo" a little bit

Add a check that verifies a create on a stream gets
NT_STATUS_DELETE_PENDING after delete-on-close has been 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>
10 months agos4:torture/vfs/fruit: update test "read open rsrc after rename" to work with macOS
Ralph Boehme [Mon, 15 Oct 2018 14:24:19 +0000 (16:24 +0200)]
s4:torture/vfs/fruit: update test "read open rsrc after rename" to work with macOS

macOS SMB server seems to return NT_STATUS_SHARING_VIOLATION in this
case while Windows 2016 returns NT_STATUS_ACCESS_DENIED.

Lets stick with the Windows error code for now in the Samba fileserver,
but let the test pass against macOS.

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>
10 months agos4:torture/vfs/fruit: ensure a directory handle is closed in all code paths
Ralph Boehme [Wed, 31 Oct 2018 14:49:45 +0000 (15:49 +0100)]
s4:torture/vfs/fruit: ensure a directory handle is closed in all code paths

Otherwise we get a sharing violation when running against Samba and
opening the directory a second time.

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>
10 months agos4:torture/vfs/fruit: update test "stream names" to work with macOS
Ralph Boehme [Mon, 15 Oct 2018 13:39:12 +0000 (15:39 +0200)]
s4:torture/vfs/fruit: update test "stream names" to work with macOS

o create the basefile before trying to create a stream on it, otherwise
  this fails on macOS

o write something to the stream, otherwise the stream is not listed as
  macOS hides 0-byte sized streams

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>
10 months agos4:torture/vfs/fruit: update test "SMB2/CREATE context AAPL" to work against macOS
Ralph Boehme [Mon, 15 Oct 2018 13:31:21 +0000 (15:31 +0200)]
s4:torture/vfs/fruit: update test "SMB2/CREATE context AAPL" to work against macOS

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>
10 months agos4:torture/vfs/fruit: set share_access to NTCREATEX_SHARE_ACCESS_MASK in check_stream...
Ralph Boehme [Thu, 11 Oct 2018 15:13:52 +0000 (17:13 +0200)]
s4:torture/vfs/fruit: set share_access to NTCREATEX_SHARE_ACCESS_MASK in check_stream_list

Avoid sharing conflicts with other opens on the basefile.

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>
10 months agos4:torture/vfs/fruit: fix a few error checks in "delete AFP_AfpInfo by writing all 0"
Ralph Boehme [Wed, 17 Oct 2018 08:51:45 +0000 (10:51 +0200)]
s4:torture/vfs/fruit: fix a few error checks in "delete AFP_AfpInfo by writing all 0"

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>
10 months agos4:torture/vfs/fruit: skip a few tests when running against a macOS SMB server
Ralph Boehme [Fri, 19 Oct 2018 20:21:10 +0000 (22:21 +0200)]
s4:torture/vfs/fruit: skip a few tests when running against a macOS SMB server

These tests are designed to test specific vfs_fruit functionality.

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>
10 months agovfs_streams_xattr: fix open implementation
Ralph Boehme [Thu, 23 Aug 2018 10:07:20 +0000 (12:07 +0200)]
vfs_streams_xattr: fix open implementation

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>
10 months agovfs_fruit: optionally delete AppleDouble files without Resourcefork data
Ralph Boehme [Tue, 9 Oct 2018 12:54:31 +0000 (14:54 +0200)]
vfs_fruit: optionally delete AppleDouble files without Resourcefork data

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agovfs_fruit: add option "delete_empty_adfiles"
Ralph Boehme [Wed, 3 Oct 2018 10:01:00 +0000 (12:01 +0200)]
vfs_fruit: add option "delete_empty_adfiles"

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agovfs_fruit: detect empty resource forks in ad_convert()
Ralph Boehme [Tue, 2 Oct 2018 14:05:28 +0000 (16:05 +0200)]
vfs_fruit: detect empty resource forks in ad_convert()

For some reason the macOS client often writes AppleDouble files with a
non-zero sized resource fork, but the resource fork data is just
boilerplate data with the following string close to the start

  This resource fork intentionally left blank

A dump with apple_dump looks like this:

Entry ID   : 00000002 : Resource Fork
Offset     : 00000052 : 82
Length     : 0000011E : 286

-RAW DUMP--:  0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F : (ASCII)
00000000   : 00 00 01 00 00 00 01 00 00 00 00 00 00 00 00 1E : ................
00000010   : 54 68 69 73 20 72 65 73 6F 75 72 63 65 20 66 6F : This resource fo
00000020   : 72 6B 20 69 6E 74 65 6E 74 69 6F 6E 61 6C 6C 79 : rk intentionally
00000030   : 20 6C 65 66 74 20 62 6C 61 6E 6B 20 20 20 00 00 :  left blank   ..
00000040   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
00000050   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
00000060   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
00000070   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
00000080   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
00000090   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
000000A0   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
000000B0   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
000000C0   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
000000D0   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
000000E0   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
000000F0   : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 : ................
00000100   : 00 00 01 00 00 00 01 00 00 00 00 00 00 00 00 1E : ................
00000110   : 00 00 00 00 00 00 00 00 00 1C 00 1E FF FF       : ..............

We can safely discard this Resource Fork data.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agovfs_fruit: add option "wipe_intentionally_left_blank_rfork"
Ralph Boehme [Wed, 3 Oct 2018 10:01:00 +0000 (12:01 +0200)]
vfs_fruit: add option "wipe_intentionally_left_blank_rfork"

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos4:torture: add test for AppleDouble ResourceFork conversion
Ralph Boehme [Thu, 4 Oct 2018 12:28:15 +0000 (14:28 +0200)]
s4:torture: add test for AppleDouble ResourceFork conversion

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos3:selftest: list vfs testssuites one per line
Ralph Boehme [Thu, 4 Oct 2018 11:47:20 +0000 (13:47 +0200)]
s3:selftest: list vfs testssuites one per line

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agodocs:vfs_fruit: add "delete_empty_adfiles" option
Ralph Boehme [Thu, 4 Oct 2018 16:22:31 +0000 (18:22 +0200)]
docs:vfs_fruit: add "delete_empty_adfiles" option

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agodocs:vfs_fruit: add "wipe_intentionally_left_blank_rfork" option
Ralph Boehme [Tue, 2 Oct 2018 14:31:15 +0000 (16:31 +0200)]
docs:vfs_fruit: add "wipe_intentionally_left_blank_rfork" option

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13642

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos3:smbd: remove "0x" string prefix from dev/ino
Ralph Boehme [Wed, 31 Oct 2018 14:16:18 +0000 (15:16 +0100)]
s3:smbd: remove "0x" string prefix from dev/ino

We used %llu as conversion specifier which results in a decimal number
being printed, so remove the misleading "0x" prefix.

While at it, I'll change %llu to the terse %ju.

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agothird_party: Update nss_wrapper to version 1.1.5
Andreas Schneider [Wed, 31 Oct 2018 07:44:08 +0000 (08:44 +0100)]
third_party: Update nss_wrapper to version 1.1.5

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agonetcmd: Add 'samba-tool group stats' command
Tim Beale [Thu, 18 Oct 2018 04:08:32 +0000 (17:08 +1300)]
netcmd: Add 'samba-tool group stats' command

With large domains it's hard to get an idea of how many groups there
are, and how many users are in each group, on average. However, this
could have a big impact on whether a problem can be reproduced or not.

This patch dumps out some summary information so that you can get a
quick idea of how big the groups are.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Autobuild-User(master): Douglas Bagnall <dbagnall@samba.org>
Autobuild-Date(master): Wed Oct 31 03:40:41 CET 2018 on sn-devel-144

10 months agonetcmd: Include num-members in 'samba-tool group list --verbose'
Tim Beale [Thu, 18 Oct 2018 03:59:24 +0000 (16:59 +1300)]
netcmd: Include num-members in 'samba-tool group list --verbose'

This adds an easy way for users to see (via samba-tool) how many members
are in various groups, without querying the members for each individual
group.

For example, you could pipe this output to grep to check for groups with
zero or one members (i.e. historic groups that may no longer make
sense).

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
10 months agoselftest: Specify different DB backends for restored testenvs
Tim Beale [Fri, 26 Oct 2018 00:21:48 +0000 (13:21 +1300)]
selftest: Specify different DB backends for restored testenvs

Vary the DB backend that we use for the renamed DCs. The labdc and
renamedc are fairly similar, so let's have each of them use a different
backend.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
10 months agonetcmd: Add backend-store option to domain backup/rename cmds
Tim Beale [Wed, 24 Oct 2018 20:03:53 +0000 (09:03 +1300)]
netcmd: Add backend-store option to domain backup/rename cmds

Currently the online/rename backup files always use the default backend
(TDB) and there is no way to change this.

This patch adds the backend-store option to the backup commands so that
you can create a backup with an MDB backend, if needed.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
10 months agotraffic_replay: logger was ignoring smb.conf log-level
Tim Beale [Mon, 22 Oct 2018 21:19:38 +0000 (10:19 +1300)]
traffic_replay: logger was ignoring smb.conf log-level

We were trying to access the debug-level (in python C bindings) before
the smb.conf had been loaded and actually set the debug-level. So it
would default to zero, regardless of what was in the smb.conf.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
10 months agotraffic_replay: Change print() to use logger()
Tim Beale [Tue, 16 Oct 2018 22:26:28 +0000 (11:26 +1300)]
traffic_replay: Change print() to use logger()

This reduces noise, so the messages only come out if you specify
--debug.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
10 months agotraffic_replay: Generate users faster by writing to local DB
Tim Beale [Thu, 11 Oct 2018 01:47:28 +0000 (14:47 +1300)]
traffic_replay: Generate users faster by writing to local DB

We can create user accounts much faster if the LDB connection talks
directly to the local sam.ldb file rather than going via LDAP. This
patch allows the 'host' argument to the tool to be a .ldb file (e.g.
"/usr/local/samba/private/sam.ldb") instead of a server name/IP.

In most cases, the traffic_replay tool wants to run on a remote device
(because the point of it is to send traffic to the DC). However, the
--generate-users-only is one case where the tool can be run locally,
directly on the test DC. (The traffic_replay user generation is handy
for standalone testing, because it also handles assigning group
memberships to the generated user accounts).

Note that you also need to use '--option="ldb:nosync = true"' to get
the improvement in performance.

Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
10 months agodsdb group audit tests: log_membership_changes extra tests
Gary Lockyer [Wed, 24 Oct 2018 21:52:55 +0000 (10:52 +1300)]
dsdb group audit tests: log_membership_changes extra tests

Add extra tests to ensure better test coverage of log_membership_changes

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Tue Oct 30 20:20:26 CET 2018 on sn-devel-144

10 months agodsdb group audit tests: check_version improve diagnostics
Gary Lockyer [Thu, 25 Oct 2018 01:38:31 +0000 (14:38 +1300)]
dsdb group audit tests: check_version improve diagnostics

Change check_version to display the expected, actual along with the
line and name of the failing test, rather than the line in check_version

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
10 months agodsdb group audit tests: check_timestamp improve diagnostics
Gary Lockyer [Thu, 25 Oct 2018 00:28:09 +0000 (13:28 +1300)]
dsdb group audit tests: check_timestamp improve diagnostics

Change check_timestamp to display the expected, actual along with the
line and name of the failing test, rather than the line in
check_timestamp.

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
10 months agodsdb group audit: align dn_compare with memcmp
Gary Lockyer [Wed, 24 Oct 2018 21:52:27 +0000 (10:52 +1300)]
dsdb group audit: align dn_compare with memcmp

Rename the parameter names and adjust the  return codes from dn_compare
so that:
dn_compare(a, b) =>

LESS_THAN means a is less than b.
GREATER_THAN means a is greater than b.

Thanks to metze for suggesting the correct semantics for dn_compare

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13664

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
10 months agodsdb group_audit: Test to replicate BUG 13664
Gary Lockyer [Tue, 23 Oct 2018 04:14:34 +0000 (17:14 +1300)]
dsdb group_audit: Test to replicate BUG 13664

The group audit code incorrectly logs member additions and deletions.

Thanks to metze for the debugging that isolated the issue, and for
suggesting the fix to dn_compare.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13664

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
10 months agodsdb: Add comments explaining the limitations of our current backlink behaviour
Andrew Bartlett [Tue, 30 Oct 2018 02:56:43 +0000 (15:56 +1300)]
dsdb: Add comments explaining the limitations of our current backlink behaviour

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Tim Beale <timbeale@catalyst.net.nz>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Tue Oct 30 10:32:51 CET 2018 on sn-devel-144

10 months agos4:samldb: internally use extended dns while changing the primaryGroupID field
Stefan Metzmacher [Fri, 24 Aug 2018 13:33:49 +0000 (15:33 +0200)]
s4:samldb: internally use extended dns while changing the primaryGroupID field

This is important, otherwise we'll loose the <SID=> component of the
linked attribute.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:repl_meta_data: add support for DSDB_CONTROL_DBCHECK_FIX_LINK_DN_SID
Stefan Metzmacher [Fri, 12 Oct 2018 13:56:18 +0000 (15:56 +0200)]
s4:repl_meta_data: add support for DSDB_CONTROL_DBCHECK_FIX_LINK_DN_SID

This will be used by dbcheck in the next commits.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_la_rep...
Stefan Metzmacher [Fri, 12 Oct 2018 16:43:25 +0000 (18:43 +0200)]
s4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_la_replace()

This will simplify further changes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_la_del...
Stefan Metzmacher [Fri, 12 Oct 2018 16:43:25 +0000 (18:43 +0200)]
s4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_la_delete()

This will simplify further changes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:repl_meta_data: add missing \n to a DEBUG message in replmd_modify_la_add()
Stefan Metzmacher [Fri, 12 Oct 2018 17:34:08 +0000 (19:34 +0200)]
s4:repl_meta_data: add missing \n to a DEBUG message in replmd_modify_la_add()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_la_add()
Stefan Metzmacher [Fri, 12 Oct 2018 16:43:25 +0000 (18:43 +0200)]
s4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_la_add()

This will simplify further changes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_handle...
Stefan Metzmacher [Fri, 12 Oct 2018 16:43:25 +0000 (18:43 +0200)]
s4:repl_meta_data: pass down struct replmd_replicated_request to replmd_modify_handle_linked_attribs()

This will simplify further changes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agoblackbox/dbcheck-links: Test broken links with missing <SID=...> on linked attributes
Stefan Metzmacher [Fri, 12 Oct 2018 13:56:18 +0000 (15:56 +0200)]
blackbox/dbcheck-links: Test broken links with missing <SID=...> on linked attributes

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agodbchecker: Fix missing <SID=...> on linked attributes
Stefan Metzmacher [Fri, 12 Oct 2018 13:56:18 +0000 (15:56 +0200)]
dbchecker: Fix missing <SID=...> on linked attributes

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agodbchecker: improve verbose output of do_modify()
Stefan Metzmacher [Mon, 8 Oct 2018 15:14:28 +0000 (17:14 +0200)]
dbchecker: improve verbose output of do_modify()

This makes it easier to debug dbcheck problems.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:dsdb: add DSDB_CONTROL_DBCHECK_FIX_LINK_DN_SID oid
Stefan Metzmacher [Mon, 8 Oct 2018 15:13:52 +0000 (17:13 +0200)]
s4:dsdb: add DSDB_CONTROL_DBCHECK_FIX_LINK_DN_SID oid

This will be used to fix missing <SID=> components in future.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agotestprogs/blackbox: add samba4.blackbox.test_primary_group test
Stefan Metzmacher [Tue, 16 Oct 2018 13:16:18 +0000 (15:16 +0200)]
testprogs/blackbox: add samba4.blackbox.test_primary_group test

This demonstrates the bug, that happens when the primaryGroupID
of a user is changed.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13418

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agos4:dsdb: fix comment on DSDB_CONTROL_DBCHECK_FIX_LINK_DN_NAME
Stefan Metzmacher [Mon, 8 Oct 2018 15:13:13 +0000 (17:13 +0200)]
s4:dsdb: fix comment on DSDB_CONTROL_DBCHECK_FIX_LINK_DN_NAME

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agoschema_samba4.ldif: add allocation of DSDB_CONTROL_DBCHECK_FIX_LINK_DN_NAME
Stefan Metzmacher [Mon, 8 Oct 2018 13:35:52 +0000 (15:35 +0200)]
schema_samba4.ldif: add allocation of DSDB_CONTROL_DBCHECK_FIX_LINK_DN_NAME

This was already allocated in source4/dsdb/samdb/samdb.h with
commit 22208f52e6096fbe9413b8ff339d9446851e0874.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
10 months agotdb: Make record deletion circular-chain safe
Volker Lendecke [Wed, 24 Oct 2018 12:31:34 +0000 (14:31 +0200)]
tdb: Make record deletion circular-chain safe

Before this patch we had 3 loops walking a hash chain to delete
records:

tdb_do_delete() to find the predecessor of the record that was to be
deleted. tdb_count_dead(), the name says it all and tdb_purge_dead()
to give back all dead records from a chain to the freelist.

This patch introduces tdb_trim_dead that walks a hash chain just
once. While it does so it counts the number of dead records, and all
records beyond tdb->max_dead_records are moved to the freelist.

Normal record deletion now works by always marking a record as dead in
step 1 and then calling tdb_trim_dead. This is made safe against
circular chains by doing the slow chain walk only in the case when we
did not delete a dead record during our walk.

It changes our dynamics a bit:

When deleting a record with non-zero max_dead_records, now we always
leave that number of records around when deleting, doing a blocking
lock on the freelist when we found too many dead records.

Previously when exceeding max_dead_records we wiped all dead records
to start accumulating them from scratch, assuming we could lock the
freelist in a nonblocking fashion.

The net effect for an uncontended freelist is the same: In
tdb_allocate() we still completely hand over all dead records to the
freelist when we could lock it, it just happens later than without
this patch.

This means for a lightly loaded system we will potentially leave more
dead records around in databases like locking.tdb. However, on a
heavily loaded system we become more predictable: If the freelist is
so heavily contended that across many deletes we can't get hold of it,
previously we accumulated more dead records than max_dead_records
would allow. This is a really lowlevel tradeoff that is likely hard to
measure, but to me becoming more deterministic without sacrificing too
much parallelism (we keep more dead records around) is worth trying.

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): Tue Oct 30 02:48:38 CET 2018 on sn-devel-144

10 months agotdb: Do early RDONLY error check for tdb_delete
Volker Lendecke [Thu, 25 Oct 2018 13:59:48 +0000 (15:59 +0200)]
tdb: Do early RDONLY error check for tdb_delete

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agotdb: Purge dead records whenever we block the freelist
Volker Lendecke [Thu, 25 Oct 2018 13:55:29 +0000 (15:55 +0200)]
tdb: Purge dead records whenever we block the freelist

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agotdb: Don't delete dead records in traverse
Volker Lendecke [Tue, 23 Oct 2018 11:40:34 +0000 (13:40 +0200)]
tdb: Don't delete dead records in traverse

The next commit will change the handling of dead records, removing the
"tdb_do_delete" function. As traverses should not happen in normal
operations, dead records from them should be rare, and relying on
traverses to remove them is a very bad idea IMHO.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agotdb: Align an integer type
Volker Lendecke [Wed, 24 Oct 2018 05:26:49 +0000 (07:26 +0200)]
tdb: Align an integer type

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agotdb: Fix a typo
Volker Lendecke [Thu, 25 Oct 2018 18:53:52 +0000 (20:53 +0200)]
tdb: Fix a typo

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agovfs_fruit: remove check for number of xattrs from ad_convert_xattr
Ralph Boehme [Fri, 19 Oct 2018 10:15:42 +0000 (12:15 +0200)]
vfs_fruit: remove check for number of xattrs from ad_convert_xattr

Turns out that there exist AppleDouble files with an extended FinderInfo
entry that includes the xattr marshall buffer, but the count of xattrs
in the buffer is just zero.

We do want to discard this extended FinderInfo entry and convert it to a
simple fixed size FinderInfo entry, so remove the check.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13649

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agopython: do not use "is" for string equality
Douglas Bagnall [Fri, 26 Oct 2018 07:25:59 +0000 (20:25 +1300)]
python: do not use "is" for string equality

This is not always going to work, and is not guaranteed to be
consistent even between minor versions.

Here is a simple counterexample:

>>> a = 'hello'
>>> a is 'hello'
True
>>> a is 'hello'.lower()
False
>>> a == a.lower()
True

Possibly it always works for the empty string, but we cannot rely
on that.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Mon Oct 29 23:13:36 CET 2018 on sn-devel-144

10 months agopython/samdb: properly use property()
Douglas Bagnall [Thu, 25 Oct 2018 09:09:59 +0000 (22:09 +1300)]
python/samdb: properly use property()

Python's property() function works like this:

property([getter[, setter[, delete[, doc]]]])

but we have been forgetting the delete function, or rather setting it
to be a string. A string is not callable and is unlikely to succeed at
deleting the property.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
10 months agos3:winbind: Check return code of initialize_password_db()
Andreas Schneider [Mon, 29 Oct 2018 18:45:58 +0000 (19:45 +0100)]
s3:winbind: Check return code of initialize_password_db()

See https://retrace.fedoraproject.org/faf/reports/1577174/

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13668

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agolib:socket: If returning early, set ifaces
David Mulder [Wed, 24 Oct 2018 16:55:02 +0000 (10:55 -0600)]
lib:socket: If returning early, set ifaces

Prevents a segfault in load_interfaces() when total interfaces == 1.
Fixes regression caused by da68a1b2f417ec82ea4ed3e7a4d867cef8ca8f93.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13665

Signed-off-by: David Mulder <dmulder@suse.com>
Reviewed-by: Ralph Böhme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Sun Oct 28 00:35:35 CEST 2018 on sn-devel-144

10 months agolib: Remove unused tdb_trans_* functions
Volker Lendecke [Tue, 25 Sep 2018 15:59:20 +0000 (08:59 -0700)]
lib: Remove unused tdb_trans_* functions

The transactions have all moved to dbwrap

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos3:torture: Rename the test file and remove it if it exists
Andreas Schneider [Fri, 19 Oct 2018 13:24:14 +0000 (15:24 +0200)]
s3:torture: Rename the test file and remove it if it exists

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos3:torture: Don't use the same testdir twice
Andreas Schneider [Fri, 19 Oct 2018 12:38:34 +0000 (14:38 +0200)]
s3:torture: Don't use the same testdir twice

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos3:selftest: Fix test names of smbtorture_s3.plain
Andreas Schneider [Thu, 18 Oct 2018 12:00:13 +0000 (14:00 +0200)]
s3:selftest: Fix test names of smbtorture_s3.plain

The env name will be appended. There is no need to have it twice. Can't
we remove the tests againa ad_dc_ntvfs completely?

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
10 months agos3:utils/smbget fix recursive download with empty source directories
Christian Ambach [Tue, 23 Oct 2018 18:05:04 +0000 (20:05 +0200)]
s3:utils/smbget fix recursive download with empty source directories

Bug: https://bugzilla.samba.org/show_bug.cgi?id=13199
Signed-off-by: Christian Ambach <ambi@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Fri Oct 26 09:58:07 CEST 2018 on sn-devel-144

10 months agos3:utils/smbget add error handling for mkdir() calls
Christian Ambach [Mon, 22 Oct 2018 14:28:21 +0000 (16:28 +0200)]
s3:utils/smbget add error handling for mkdir() calls

Signed-off-by: Christian Ambach <ambi@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
10 months agos3:script/tests reduce code duplication
Christian Ambach [Mon, 22 Oct 2018 14:22:00 +0000 (16:22 +0200)]
s3:script/tests reduce code duplication

Signed-off-by: Christian Ambach <ambi@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
10 months agopytests/samba3sam: unshadow and fix a search_non_mapped test
Douglas Bagnall [Thu, 11 Oct 2018 22:52:49 +0000 (11:52 +1300)]
pytests/samba3sam: unshadow and fix a search_non_mapped test

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
Autobuild-User(master): Noel Power <npower@samba.org>
Autobuild-Date(master): Fri Oct 26 00:50:37 CEST 2018 on sn-devel-144

10 months agopython/tests/ou: unshadow a test
Douglas Bagnall [Thu, 11 Oct 2018 01:45:58 +0000 (14:45 +1300)]
python/tests/ou: unshadow a test

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agosamba-tool tests: fix bytes/str issue in masked test
Douglas Bagnall [Wed, 24 Oct 2018 22:40:53 +0000 (11:40 +1300)]
samba-tool tests: fix bytes/str issue in masked test

This test will be revealed to the world in the next commit.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agopython/tests/raw_protocol: reveal shadowed test via disambiguation
Douglas Bagnall [Thu, 11 Oct 2018 01:42:40 +0000 (14:42 +1300)]
python/tests/raw_protocol: reveal shadowed test via disambiguation

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agos4/dsdb/pytest/ldap_schema: remove duplicate test
Douglas Bagnall [Thu, 11 Oct 2018 04:00:46 +0000 (17:00 +1300)]
s4/dsdb/pytest/ldap_schema: remove duplicate test

The tests differ in two lines, thus:

     def test_subClassOf(self):
     -        """ Testing usage of custom child schamaClass
     +        """ Testing usage of custom child classSchema

[...]

   -governsId: 1.3.6.1.4.1.7165.4.6.2.6.3.""" + str(random.randint(1, 100000)) + """
   +governsId: 1.3.6.1.4.1.7165.4.6.2.6.7.""" + str(random.randint(1, 100000)) + """

with the governsId OID changed because it was noticed they were colliding.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agoldb/tests/py/api: reveal shadowed casefold test
Douglas Bagnall [Thu, 11 Oct 2018 03:47:44 +0000 (16:47 +1300)]
ldb/tests/py/api: reveal shadowed casefold test

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agopython/tests/kcc_utils: disambiguate/unshadow a test
Douglas Bagnall [Thu, 11 Oct 2018 03:15:43 +0000 (16:15 +1300)]
python/tests/kcc_utils: disambiguate/unshadow a test

Pair-programmed-with: Garming Sam <garming@catalyst.net.nz>
Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agopython/tests/lsa_string: remove duplicate method
Douglas Bagnall [Thu, 11 Oct 2018 00:50:06 +0000 (13:50 +1300)]
python/tests/lsa_string: remove duplicate method

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agopython/gp_parse/gp_inf: remove unused variables
Douglas Bagnall [Thu, 25 Oct 2018 09:34:31 +0000 (22:34 +1300)]
python/gp_parse/gp_inf: remove unused variables

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>
10 months agopython/gp_parse/gp_inf: remove shadowed method
Douglas Bagnall [Thu, 11 Oct 2018 00:23:01 +0000 (13:23 +1300)]
python/gp_parse/gp_inf: remove shadowed method

The 'from_xml()' definition is replaced a few lines down

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Noel Power <npower@samba.org>