Justin Maggard [Wed, 2 Mar 2016 18:19:56 +0000 (10:19 -0800)]
s3:smbd: add negprot remote arch detection for OSX
Remote arch detection for OSX clients has been broken for some time, since
both Samba and OSX started supporting SMB2. Fix it by adding modern OSX
client detection support to the negprot remote arch detection routine.
Signed-off-by: Justin Maggard <jmaggard10@gmail.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Ralph Boehme <rb@sernet.de>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Thu Mar 3 09:03:53 CET 2016 on sn-devel-144
Justin Maggard [Wed, 2 Mar 2016 18:18:34 +0000 (10:18 -0800)]
s3:smbd: rework negprot remote arch detection
Negprot remote arch detection is very cryptic. Rework it so it's easier
to understand, and therefore more extensible, following the protocol table
in inline comments. This also allows us to remove some hacks.
Signed-off-by: Justin Maggard <jmaggard10@gmail.com> Reviewed-by: Ralph Boehme <rb@sernet.de> Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit b69b96fc14128f50a906b4f5b34934efbc94afbc)
Karolin Seeger [Wed, 24 Feb 2016 11:22:26 +0000 (12:22 +0100)]
WHATSNEW: Add release notes for Samba 4.2.9.
CVE-2015-7560 Getting and setting Windows ACLs on symlinks can change
permissions on link target.
CVE-2016-0771: Read of uninitialized memory DNS TXT handling
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Garming Sam [Tue, 15 Dec 2015 04:22:32 +0000 (17:22 +1300)]
CVE-2016-0771: tests/dns: restore formerly segfaulting test
This was on the client side, due the a strlen(NULL) on the previously
DOS-encoded TXT field. With a new IDL structure, this segfault no longer exists.
Note that both Samba and Windows return NXRRSET instead of FORMERR.
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
As part of reload services, unused service ptr structures are getting
unloaded without actually freeingup the memory associated to them.
Made changes to freeup those allocations on unload. On reload, reuse the
slots in ServicePtr global array instead of extending the array with
holes.
Also we should mark shares like IPC$ as auto loaded as they never be
exposed to users.
Signed-off-by: Hemanth Thummala <hemanth.thummala@nutanix.com> Reviewed-by: Volker Lendecke <Volker.Lendecke@SerNet.DE> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Feb 17 22:30:25 CET 2016 on sn-devel-144
Martin Schwenke [Mon, 8 Feb 2016 04:55:17 +0000 (15:55 +1100)]
ctdb-scripts: Drop use of "smbcontrol winbindd ip-dropped ..."
This is unnecessary in Samba >= 4.0 because winbindd monitors IP
address itself and no longer needs to be told when they are dropped.
The smbcontrol commands can hang if a node has recovery mode active
because smbcontrol is unable to connect to the registry. Therefore,
the smbcontrol commands should be removed.
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Wed Feb 10 14:08:17 CET 2016 on sn-devel-144
Ralph Boehme [Thu, 4 Feb 2016 14:35:06 +0000 (15:35 +0100)]
lib/tsocket: workaround sockets not supporting FIONREAD
Netlink sockets don't support querying pending bytes with ioctl(fd,
FIONREAD, ...) and would return EOPNOTSUPP, so use recvmsg() with
MSG_PEEK|MSG_TRUNC as a fallback.
The MSG_TRUNC flag to recvmsg() is Linux only, but netlink is as well,
so we're safe for now.
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Amitay Isaacs <amitay@gmail.com> Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Wed Feb 10 10:30:24 CET 2016 on sn-devel-144
Signed-off-by: Hemanth Thummala <hemanth.thummala@nutanix.com> Reviewed-by: Alexander Bokovoy <ab@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Feb 4 12:39:14 CET 2016 on sn-devel-144
Jeremy Allison [Tue, 9 Feb 2016 20:47:43 +0000 (12:47 -0800)]
s3: smbd: posix_acls: Fix check for setting u:g:o entry on a filesystem with no ACL support.
Since 4.0.x we add 2 additional ACE entries, one SMB_ACL_USER
and SMB_ACL_GROUP to match the existing SMB_ACL_USER_OBJ and
SMB_ACL_GROUP_OBJ entries. The two additional entries break
the simple "must have 3 entries" check done inside convert_canon_ace_to_posix_perms().
Replace this with a more complete test.
Problem and initial fix provided by <tcleamy@ucdavis.edu>
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Uri Simchoni <uri@samba.org>
Autobuild-User(master): Uri Simchoni <uri@samba.org>
Autobuild-Date(master): Thu Feb 11 11:14:53 CET 2016 on sn-devel-144
Signed-off-by: Jose A. Rivera <jarrpa@samba.org>
Pair-programmed-with: Ira Cooper <ira@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Feb 5 04:37:43 CET 2016 on sn-devel-144
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Karolin Seeger <kseeger@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Feb 3 21:25:54 CET 2016 on sn-devel-144
Christian Ambach [Wed, 30 Dec 2015 20:25:13 +0000 (21:25 +0100)]
s3:utils/smbget fix recursive download
get_auth_data is called multiple times (once for the directory listing and then
for every file to be downloaded). Save the obtained values across multiple calls
to make smbclient use the correct username for each download.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=6482 Signed-off-by: Christian Ambach <ambi@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit ec802d27ce4dc6dd9b5e5ebd6992f90364d855a2)
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Michael Adam <obnox@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 192f1516c378ae942d14921bfcc1e11173da36e6)
We need a wrapper for smbspool to be able to authenticate with Kerberos.
This needs to replace the cups smb backend. The permission need to be
0700 and the owner root.
Note that Kerberos support is broken in CUPS 2.1.2 maybe earlier
versions. It works with 1.6.3.
Signed-off-by: Andreas Schneider <asn@cryptomilk.org> Reviewed-by: Michael Adam <obnox@samba.org> Reviewed-by: Guenther Deschner <gd@samba.org> Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Mon Jan 25 19:58:46 CET 2016 on sn-devel-144
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): Sun Jan 24 01:20:52 CET 2016 on sn-devel-144
Ralph Boehme [Wed, 25 Nov 2015 14:23:26 +0000 (15:23 +0100)]
s3:smbd: Ignore initial allocation size for directory creation
We reject directory creation with an initial allocation size > 0 with
NT_STATUS_ACCESS_DENIED. Windows servers ignore the initial allocation
size on directories.
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): Wed Jan 20 12:12:12 CET 2016 on sn-devel-144
Signed-off-by: Christian Ambach <ambi@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Jan 6 08:58:56 CET 2016 on sn-devel-144
The fix fails to correctly treat a corner case where the share
path is "/". This case is important for some real world
scenarios, notably the use of the glusterfs VFS module:
For the share path "/", the newly introduced checks deny all
operations in the share.
This change fixes the checks for the corner case.
The point is that the assumptions on which the original
checks are based are not true for the rootdir "/" case.
This is the case where the rootdir starts _and ends_ with
a slash. Hence a subdirectory does not continue with a
slash after the rootdir, since the candidate path has
been normalized.
This fix just omits the string comparison and the
next character checks in the case of rootdir "/",
which is correct because we know that the candidate
path is normalized and hence starts with a '/'.
The patch is fairly minimal, but changes indentation,
hence best viewed with 'git show -w'.
A side effect is that the rootdir="/" case needs
one strncmp less.
Pair-Programmed-With: Jose A. Rivera <jarrpa@samba.org>
Signed-off-by: Michael Adam <obnox@samba.org> Signed-off-by: Jose A. Rivera <jarrpa@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Thu Dec 24 00:57:31 CET 2015 on sn-devel-144
Ralph Boehme [Mon, 24 Aug 2015 15:45:14 +0000 (17:45 +0200)]
vfs_streams_xattr: fix and simplify streams_xattr_get_name()
streams_xattr_get_name() fails to chop off the stream type in case
config->store_stream_type is false and the passed stream name contains a
stream type.
Eg when the passed in stream name is ":mystream:$DATA", but
config->store_stream_type is false, we must generate a xattr name of
"mystream" or "user.mystream".
Signed-off-by: Ralph Boehme <slow@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Oct 16 23:27:01 CEST 2015 on sn-devel-104
Ralph Boehme [Sat, 19 Dec 2015 10:10:54 +0000 (11:10 +0100)]
s4:torture:vfs_fruit: update AFP_AfpInfo IO tests
When reading from the AFP_AfpInfo stream, OS X ignores the offset from
the request and always reads from offset=0.
The offset bounds check has a off-by-1 bug in OS X, so a request
offset=60 (AFP_AfpInfo stream has a ficed size of 60 bytes), len=1
returns 1 byte from offset 0 insteaf of returning 0.
Ralph Boehme [Sat, 19 Dec 2015 10:06:19 +0000 (11:06 +0100)]
vfs_fruit: fix offset and len handling for AFP_AfpInfo stream
When reading from the AFP_AfpInfo stream, OS X ignores the offset from
the request and always reads from offset=0.
The offset bounds check has a off-by-1 bug in OS X, so a request
offset=60 (AFP_AfpInfo stream has a ficed size of 60 bytes), len=1
returns 1 byte from offset 0 insteaf of returning 0.
Ralph Boehme [Fri, 18 Dec 2015 16:18:41 +0000 (17:18 +0100)]
s4:torture:vfs_fruit: add explicit cleanup of testfiles
smb2_deltree() doesn't work with OS X (looks like OS X doesn't handle
FILE_NON_DIRECTORY_FILE correctly). As a workaround, use explicit
cleanup of all testfiles and directories.
Ralph Boehme [Thu, 17 Dec 2015 15:51:10 +0000 (16:51 +0100)]
s4:torture:vfs_fruit: use AFPINFO_STREAM_NAME
I got erratic results from OS X SMB server with AFPINFO_STREAM
(":AFP_AfpInfo:$DATA") in some tests. Using AFPINFO_STREAM_NAME
(just the ":AFP_AfpInfo" part) instead fixed this.
Ralph Boehme [Thu, 17 Dec 2015 12:31:12 +0000 (13:31 +0100)]
s4:torture:vfs_fruit: tweak check_stream_list()
Modify check_stream_list() to open the basefile (or directory) itself
insteaf of having the callers pass in a filehandle. Removes some code
duplication in the callers.
Ralph Boehme [Wed, 16 Dec 2015 12:13:55 +0000 (13:13 +0100)]
smbd: make "hide dot files" option work with "store dos attributes = yes"
When using "store dos attributes = yes", the function that reads the
attributes from the xattr get_ea_dos_attribute() will overwrite the
attribute previously set for "hide dot files".
According to smb.conf, "store dos attributes = yes" should only
overwrite the "map XXX" options, but not "hide dot files".
Jeremy Allison [Tue, 8 Dec 2015 00:32:57 +0000 (16:32 -0800)]
lib/param: add a fixed unified lpcfg_string_{free,set,set_upper}() infrastructure
This reduces the memory footprint of empty string options.
smbd -d1 -i with 1400 shares in smb.conf under x64 valgrind massif before this
patch has 7,703,392 bytes peak memory consumption and after this patch
3,321,200 bytes.
Ralph Boehme [Thu, 22 Jan 2015 09:00:15 +0000 (10:00 +0100)]
s3:smbd: file_struct: seperate POSIX directory rename cap from POSIX open
We need more fine grained control over which POSIX semantics we'd like
to enable per file handle. Currently POSIX_FLAGS_OPEN is a kitchensink
for all kinds of stuff like:
projects / thirdparty / samba.git / log
