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.
This was a bad API, and it was used in a buggy way: In
messaging_dispatch_rec we always did the defer, we referenced the
destination pid, not the source. In messaging_send_iov this is the right
thing to do to reference the destination, but when we have arrived in
messaging_dispatch_rec we should compare source and destination.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Using the new msg_source program with 1.500 instances against a single
msg_sink I found the msg_source process to spawn two worker threads for
synchronously sending the data towards the receiving socket. This should
not happen: Per destination node we only create one queue. We strictly
only add pthreadpool jobs one after the other, so a single helper thread
should be perfectly sufficient.
It turned out that under heavy overload the main sending thread was
scheduled before the thread that just had finished its send() job. So
the helper thread was not able to increment the pool->num_idle variable
indicating that we don't have to create a new thread when the new job
is added.
This patch moves the signalling write under the mutex. This means that
indicating readiness via the pipe and the pool->num_idle variable happen both
under the same mutex lock and thus are atomic. No superfluous threads anymore.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
With this pair of programs I did some performance tests of the messaging
system. Guess what -- I found two bugs :-)
See the subsequent patches.
With 1500 msg_source processes I can generate message overload: A
Intel(R) Xeon(R) CPU L5640 @ 2.27GHz
can receive roughly 100k messages per second. When using
messaging_read_send/recv user/system time is roughly even, a bit more
work done in user space. When using messaging_register, due to less
malloc activity, user space chews a lot less.
By the way: 1.500 helper threads in a blocking sendto() against a single
datagram socket reading as fast as it can (with epoll_wait in between)
only drove the loadavg to 12 on a 24-core machine. So I guess unix domain
datagram sockets are pretty well protected against overload. No thundering
herd or so. Interestingly "top" showed msg_sink at less than 90% CPU,
although it was clearly the bottleneck. But that must be a "top" artifact.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Now we check the cleanup when conflicts happen, not when we first open
the file. This means we don't have to re-open the connection to make
cleanup happen.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is now only called during brl_forall. It does not really hurt if we list
dead processes here. If the upper layers really care, they can filter it out
themselves. The real lock conflicts are not removed on-demand.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This avoids the need to do sweeping validate_lock_entries calls
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Tidy-up of code obsoleted by fixes for bug #10773 (SECINFO_PROTECTED_DACL is not ignored).
We now never pass SECINFO_PROTECTED_DACL in security_information flags to this layer.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=10773
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Aug 22 11:26:57 CEST 2014 on sn-devel-104
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 Aug 22 05:20:43 CEST 2014 on sn-devel-104
... much nicer than PRIu64
Also, append a \n. Makes it better readable when looking at the lockfile
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Bug: https://bugzilla.samba.org/show_bug.cgi?id=10773
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Aug 22 02:52:50 CEST 2014 on sn-devel-104
Sometimes Windows clients doesn't filter SECINFO_[UN]PROTECTED_[D|S]ACL flags
before sending the security_information to the server.
security_information = SECINFO_PROTECTED_DACL| SECINFO_DACL
results in a NULL dacl being returned from an GetSecurityDecriptor
request. This happens because posix_get_nt_acl_common()
has the following logic:
if ((security_info & SECINFO_DACL) && !(security_info & SECINFO_PROTECTED_DACL)) {
... create DACL ...
}
I'm not sure if the logic is correct or wrong in this place (I guess it's
wrong...).
But what I know is that the SMB server should filter the given
security_information flags before passing to the filesystem.
[MS-SMB2] 3.3.5.20.3 Handling SMB2_0_INFO_SECURITY
...
The server MUST ignore any flag value in the AdditionalInformation field that
is not specified in section 2.2.37.
Section 2.2.37 lists:
OWNER_SECURITY_INFORMATION
GROUP_SECURITY_INFORMATION
DACL_SECURITY_INFORMATION
SACL_SECURITY_INFORMATION
LABEL_SECURITY_INFORMATION
ATTRIBUTE_SECURITY_INFORMATION
SCOPE_SECURITY_INFORMATION
BACKUP_SECURITY_INFORMATION
Bug: https://bugzilla.samba.org/show_bug.cgi?id=10773
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
A SMB server should only care about specific SECINFO flags
and ignore others e.g. SECINFO_PROTECTED_DACL.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=10773
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Set a max charge for SMB2 connections so that larger request sizes can
be used and more requests can be in flight.
Signed-off-by: Ross Lagerwall <rosslagerwall@gmail.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Aug 21 17:31:11 CEST 2014 on sn-devel-104
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Ira Cooper <Ira@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Thu Aug 21 14:58:37 CEST 2014 on sn-devel-104
Bug: https://bugzilla.samba.org/show_bug.cgi?id=10751
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Aug 21 11:36:55 CEST 2014 on sn-devel-104
default/ctdb/utils/smnotify/gen_xdr.c: In function ‘xdr_status’:
default/ctdb/utils/smnotify/gen_xdr.c:11:20: warning: unused variable ‘buf’ [-Wunused-variable]
register int32_t *buf;
^
When generating the code, change it to assign the variable to itself.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Thu Aug 21 07:11:02 CEST 2014 on sn-devel-104
Seeing these with -Wall:
../server/ctdb_call.c:1117:3: warning: dereferencing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing]
record_flags = *(uint32_t *)&c->data[c->keylen + c->datalen];
^
memcpy() seems to be the easiest way to get fix these. The
alternative would be to use unmarshalling functions.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
We really should extricate these from source3 and into some common
code. However, just copy them for now to help get rid of a lot of
warnings.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
The (return) value of FAILED() is a constant 1. However, it is never
used, so the compiler complains when run with -Wall:
lib/replace/test/os2_delete.c: In function ‘cleanup’:
lib/replace/test/os2_delete.c:39:163: warning: right-hand operand of comma expression has no effect [-Wunused-value]
FAILED("system");
So just get remove the ", 1" since it is the bit that does nothing and
is never used.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Wed Aug 20 16:54:31 CEST 2014 on sn-devel-104
Signed-off-by: Ralph Boehme <rb@sernet.de>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Aug 18 20:08:32 CEST 2014 on sn-devel-104
Add wrapper functions that connect two trees with sharenames taken
from passed option.
Signed-off-by: Ralph Boehme <rb@sernet.de>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Add a utility function that takes an option name as parameter and then
uses the value of the option 'torture:NAME' as share name in a tree
connect.
Signed-off-by: Ralph Boehme <rb@sernet.de>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
This module provides enhanced compatibility with Apple SMB clients and
interoperability with a Netatalk 3 AFP fileserver.
The module intercepts the OS X special streams "AFP_AfpInfo" and
"AFP_Resource" and handles them in a special way. All other named
streams are deferred to vfs_streams_xattr.
The OS X client maps all NTFS illegal characters to the Unicode
private range. This module optionally stores the charcters using their
native ASCII encoding.
Open modes are optionally checked against Netatalk AFP share modes.
The "AFP_AfpInfo" named stream is a binary blob containing OS X
extended metadata for files and directories. This module optionally
reads and stores this metadata in a way compatible with Netatalk 3
which stores the metadata in an EA "org.netatalk.metadata". Cf
source3/include/MacExtensions.h for a description of the binary blobs
content.
The "AFP_Resource" named stream may be arbitrarily large, thus it
can't be stored in an EA on most filesystem. ZFS on Solaris is an
exception to the rule, because it there EAs can be of any size and EAs
are first-class filesystem objects that can be used with normal file
syscalls like open(), read(), write(), fcntl() asf. This module stores
the AFP_Resource stream in an AppleDouble file, prepending "._" to the
filename. On Solaris and ZFS the stream is optionally stored in an EA
"org.netatalk.ResourceFork".
Signed-off-by: Ralph Boehme <rb@sernet.de>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
AFP_BackupTime value must be 0x80000000 and all existing defines use
native byte order, not byte swapped.
Signed-off-by: Ralph Boehme <rb@sernet.de>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
The fix missed one instance, as autobuild has just told me...
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Mon Aug 18 17:42:00 CEST 2014 on sn-devel-104
Previously, --without-dmapi would still autodetect and link a useable dmapi
library. This change allows to build without dmapi support even when a dmapi
library is found.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=10369
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Michael Adam <obnox@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Currently tdbtorture prints the test details (processes, loops, etc.)
from the first forked child process.
When run with -n 1 (one process), the test is run from within the
parent and no details are printed. This change ensures that they are.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Mon Aug 18 13:26:32 CEST 2014 on sn-devel-104
We go to a great deal of effort to avoid administrators posting their
passwords in Samba logs, and one of the ways we do that is to remove
them from internal ldif dumps Samba produces while operating as an AD
DC.
clearTextPassword is not a real attribute, but it functions as one for
an input path.
Change-Id: Iaacf3354fc9bfff18d6774f49b17a9ba962347d5
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Sat Aug 16 01:05:07 CEST 2014 on sn-devel-104
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 Aug 15 20:34:56 CEST 2014 on sn-devel-104
Signed-Off-By: Jelmer Vernooij <jelmer@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Fri Aug 15 05:47:51 CEST 2014 on sn-devel-104
