* add tevent_threaded_context_create() and tevent_threaded_schedule_immediate()
They add a way to pass the thread result from a helper thread into
the main event loop.
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Volker Lendecke [Fri, 12 Aug 2016 14:00:56 +0000 (16:00 +0200)]
tevent: Move rundown of the event pipe
Purely cosmetic change: This moves closing the signal/thread event pipe
to where it's opened. This prepares the eventfd support, making the
"magic" for eventfd more obvious.
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Volker Lendecke [Mon, 8 Aug 2016 09:26:37 +0000 (11:26 +0200)]
tevent: Add threaded immediate activation
This is infrastructure to improve our async r/w result handling and latency.
The pthreadpool signalling goes through a pipe. This has downsides: The main
event loop has to go through a read on the pipe before it can ship the result.
Also, it is not guaranteed by poll/epoll that the pthreadpool signal pipe is
handled with top priority. When an async pread/pwrite has finished, we should
immediately ship the result to the client, not waiting for anything else.
This patch enables tevent_immediate structs as job signalling. This means a
busy main tevent loop will handle the threaded job completion before any timed
or file descriptor events. Opposite to Jeremy's tevent_thread_proxy this is
done by a modification of the main event loop by looking at a linked list under
a central mutex.
Regarding performance: In a later commit I've created a test that does nothing
but fire one immediate over and over again. If you add a phread_mutex_lock and
unlock pair in the immediate handler, you lose roughly 25% of rounds per
second, so it is measurable. It is questionable that will be measurable in the
real world, but to counter concerns activation of immediates needs to go
through a new struct tevent_threaded_context. Only if such a
tevent_threaded_context exists for a tevent context, the main loop takes the
hit to look at the mutex'ed list of finished jobs.
This patch by design does not care about talloc hierarchies. The idea is that
the main thread owning the tevent context creates a chunk of memory and
prepares the tevent_immediate indication job completion. The main thread hands
the memory chunk together with the immediate as a job description over to a
helper thread. The helper thread does its job and upon completion calls
tevent_threaded_schedule_immediate with the already-prepared immediate. From
that point on memory ownership is again transferred to the main thread.
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Signalling the main event loop will also happen from threads soon, and
that will use the same mechanism. This also keeps the pipe open after the last
signal handler is removed. Threaded jobs will come and go very frequently, and
always setting up and tearing down the pipe for each job will be expensive.
Also, this is "just" two file descriptors, and with eventfd just one.
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Martin Schwenke [Tue, 23 Aug 2016 01:49:54 +0000 (11:49 +1000)]
ctdb-tests: Reimplement daemons_stop() using ctdbd_wrapper
The current daemons_stop() implementation uses very loose matching to
decide which processes to kill if "ctdb shutdown" hasn't worked within
1 second. This can cause ctdbd processes from other test runs to be
killed.
Instead, use ctdbd_wrapper, which uses the PID file as a last resort.
This has the advantage of never killing ctdbd processes from other
test runs. However, this also has the obvious consequence that an
interrupted test run in one directory can not have its daemons cleaned
up from a new test run in a different directory.
The 4.5.x WHATSNEW.txt is maintained in v4-5-test.
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Tue Aug 23 16:02:09 CEST 2016 on sn-devel-144
Volker Lendecke [Sun, 21 Aug 2016 15:46:16 +0000 (17:46 +0200)]
messaging_dgm: Fix signed/unsigned hickups
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 Aug 23 10:46:17 CEST 2016 on sn-devel-144
s4-torture: Enable tests for printer driver info level 8
Pair-Programmed-With: Guenther Deschner <gd@samba.org> Signed-off-by: Andreas Schneider <asn@samba.org> Signed-off-by: Guenther Deschner <gd@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Aug 23 05:00:13 CEST 2016 on sn-devel-144
s3-spoolss: Support for adding printer drivers with info level 8
We already supported getting driver info level 8 but not adding it. This
allows adding printer drivers with level. So several fields where emtpy.
Microsoft released a security update for Windows print spooler
components which requires support for driver info level 8 now. This is
needed to find out if a driver is PACKAGE_AWARE or not.
Pair-Programmed-With: Guenther Deschner <gd@samba.org> Signed-off-by: Andreas Schneider <asn@samba.org> Signed-off-by: Guenther Deschner <gd@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Uri Simchoni <uri@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon Aug 22 22:59:22 CEST 2016 on sn-devel-144
Jeremy Allison [Fri, 19 Aug 2016 23:58:39 +0000 (16:58 -0700)]
s3: libsmb: Correctly align create contexts in a create call.
SMB2 shadow copy requests are the first time we've used
create contexts in anger in this codepath. This took me
longer than I'd like to admit to find :-).
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): Sun Aug 21 22:45:33 CEST 2016 on sn-devel-144
Many years ago takeover_callback_state was used for both IP takeover
and release. Now it is only used when releasing an IP so rename it to
improve clarity.
Volker Lendecke [Tue, 20 Oct 2015 09:39:09 +0000 (11:39 +0200)]
lib: Implement poll_timeout
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Sat Aug 20 13:38:37 CEST 2016 on sn-devel-144
Volker Lendecke [Thu, 18 Aug 2016 20:00:06 +0000 (22:00 +0200)]
lib: Cleanup includes in messages_dgm
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Aug 20 09:41:33 CEST 2016 on sn-devel-144
Jeremy Allison [Thu, 18 Aug 2016 21:36:50 +0000 (14:36 -0700)]
s3: smbclient: In order to get shadow copy data over SMB1 we must call cli_shadow_copy_data() twice.
Once with 'get_names = false' to get the size, then again with 'get_names = true'
to get the data or a Windows server fails to return valid info. Samba doesn't have this bug.
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): Fri Aug 19 23:59:35 CEST 2016 on sn-devel-144
Douglas Bagnall [Wed, 11 May 2016 23:38:09 +0000 (11:38 +1200)]
netcmd/domain: improve error message grammar
Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz> Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Autobuild-User(master): Douglas Bagnall <dbagnall@samba.org>
Autobuild-Date(master): Fri Aug 19 17:30:39 CEST 2016 on sn-devel-144
Christof Schmitt [Thu, 18 Aug 2016 21:01:50 +0000 (14:01 -0700)]
winbindd: Introduce helper function for winbindd_cache.tdb directory
Also add a comment why the file is placed in the state directory.
Signed-off-by: Christof Schmitt <cs@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Fri Aug 19 13:30:23 CEST 2016 on sn-devel-144
pvfs_open win10 fix, need return SMB2_CREATE_TAG_QFID
Signed-off-by: kkhaike <kkhaike@gmail.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Uri Simchoni <uri@samba.org>
Autobuild-User(master): Uri Simchoni <uri@samba.org>
Autobuild-Date(master): Fri Aug 19 09:35:15 CEST 2016 on sn-devel-144
Signed-off-by: Uri Simchoni <uri@samba.org> Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Thu Aug 18 18:58:22 CEST 2016 on sn-devel-144
Martin Schwenke [Thu, 18 Aug 2016 02:57:33 +0000 (12:57 +1000)]
ctdb-ipalloc: Fix cumulative takeover timeout
Commit c40fc62642ff5ac49b75e9af49c299e33dbc9073 runs the IP allocation
algorithm after calculating the timeout offset. If the algorithm
takes a long time then there may be no attempt to release or take over
IPs.
Instead, reset the timeout just before the RELEASE_IP stage if an
early jump to IPREALLOCATED was not taken.
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Christof Schmitt <cs@samba.org> Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Thu Aug 18 06:43:02 CEST 2016 on sn-devel-144
Jeremy Allison [Wed, 17 Aug 2016 17:49:50 +0000 (10:49 -0700)]
s3: vfs: shadow_copy2: Replace all uses of (p-name) with len_before_gmt.
p and name don't change, and we've already calculated this length.
Part of the effort to make the code inside vfs_snapper.c that does
the same thing more similar (we can't make these functions identical
due to the 'snapdir' use case).
Martin Schwenke [Mon, 1 Aug 2016 06:00:55 +0000 (16:00 +1000)]
ctdb-tests: Validate that TAKE_IP works with IP already on an interface
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Thu Aug 18 02:50:16 CEST 2016 on sn-devel-144
Martin Schwenke [Tue, 2 Aug 2016 02:18:15 +0000 (12:18 +1000)]
ctdb-recoverd: Don't directly release rogue IP addresses
This is inconsistent with the rest of the local IP verification. It
should notice problems but not try to fix them directly. Like other
cases, it should use an IP takeover run to try to fix the problem. In
this case the address might have just been added and an out-of-band
RELEASE_IP might cause conflicts (i.e. "another change is in flight")
with a scheduled IP takeover run.
Martin Schwenke [Fri, 29 Jul 2016 09:48:37 +0000 (19:48 +1000)]
ctdb-daemon: Avoid referencing NULL pointer due to unknown old interface
This doesn't currently happen but it will in a subsequent commit.
That commit and this one could be squashed but then the functional
change gets lost in amongst this one.
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Fri, 29 Jul 2016 09:41:01 +0000 (19:41 +1000)]
Revert "When adding an ip at runtime, it might not yet have an iface assigned to it, so ensure that the next takover_ip call will fall through to accept the ip and add it."
If the IP address is on an interface then it won't help to pretend
that it isn't. This will simply cause a takeip event, which will fail
because the address can't be added. Note that the IP address isn't
necessarily new - something unexpected may have happened.
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Wed, 17 Aug 2016 04:43:05 +0000 (14:43 +1000)]
ctdb-daemon: Drop special case handling for new IP already on interface
The address may already be assigned to another node, so this is wrong.
It also leaves the interface unknown.
This is better left to code that handles rogue IP addresses. A
takeover run should correctly takeover the address if it is assigned
to this node or release it if it is assigned to another node. Coming
soon...
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Thu, 11 Aug 2016 04:07:44 +0000 (14:07 +1000)]
ctdb-daemon: Use release_ip_post() when releasing all IP addresses
This has the advantage of using common code. Also, if there was
previously a failed attempt to release the IP address as part of a
delete, then this will finish processing the delete.
Extra care needs to be taken when a VNN is actually deleted.
Martin Schwenke [Thu, 11 Aug 2016 03:57:43 +0000 (13:57 +1000)]
ctdb-daemon: Factor out new function release_ip_post()
This contains the cleanup that needs to be done after an IP address is
released from an interface.
state->vnn is set to the return value from release_ip_post(), which is
either the original VNN, or NULL if it was deleted. This allows
correct handling of the in-flight flag in the destructor for state.
Martin Schwenke [Thu, 11 Aug 2016 03:41:12 +0000 (13:41 +1000)]
ctdb-daemon: Do not copy address for RELEASE_IP message
If there's an allocation failure then the implicit early return in
CTDB_NO_MEMORY_VOID() means that no reply is sent to the control.
ctdb_daemon_send_message() makes a copy of the data, so don't copy it
here and remove an unnecessary chance of failure.
projects / thirdparty / samba.git / log
