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.
Every time a nodemap is contructed the node IP addresses all need to
be parsed. This isn't very productive use of CPU.
Instead, parse each string once when the nodes file is loaded. This
results in much simpler code.
This code also removes the use of ctdb_address. Duplicating the port
is pointless without an abstraction layer around ctdb_address. If
CTDB gets an incompatible transport in the future then add an
abstraction layer.
Note that the infiniband code is not updated. Compilation of the
infiniband code is already broken. Fixing it will be a separate,
properly tested effort.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Having it require a CTDB context stops ctdb_parse_address() from being
used in more generic code. Just use the existing talloc context for
memory allocations.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Have it just silently take or fail to take the lock, except on an
unexpected failure (where it should log an error).
This means that when it is called we need to keep the old behaviour
and explicitly release the lock. In do_recovery() the lock is
released and a message is printed before attempting to take the lock.
In the daemon sanity check the lock must be released in the error path
if it is actually taken.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Unlock the recovery lock file. This way knowledge of the file
descriptor isn't sprinkled throughout the code.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
It is much simpler for most cases to have a syslog backend that
doesn't need a separate CTDB-specific logging daemon. This loses the
lossy, non-blocking mode provided by logd. However, a corresponding
feature with a completely different implemention (not requiring an
extra daemon) will be re-added into the syslog backend. In an ideal
world the new implementation would be added first but unfortunately
that is hard to do because the logd code is hooked in at more than one
place.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This makes the code cleaner and allows the syslog backend to be easily
modified without affecting other code. Also do some extra clean-up,
including whitespace fixups.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This is only used by logging code and there is already a file-level
variable for this. struct ctdb_context already contains too many
things.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
As far as we know, nobody uses this and it just complicates the
logging subsystem.
Remove all ringbuffer code and documentation. Update the local
daemons startup code correspondingly.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Volker Lendecke <vl@samba.org>
Warnings are currently produced when compiling Samba and
ctdb_private.h is included. A forward enum declaration avoids the
warning.
This is a temporary measure. The log ringbuffer should be removed
soon.
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): Tue Sep 23 10:31:50 CEST 2014 on sn-devel-104
To avoid warnings when using --enable-developer, which uses
-Wmissing-prototypes.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
When CTDB receives DMASTER_REQUEST or DMASTER_REPLY packet, the specified
record needs to be updated as soon as possible to avoid inconsistent
dmaster information between nodes. During this time, queue up all calls
for that record and process them only after dmaster request/reply has
been processed.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
There is no need for a special function to free lock request and
corresponding lock context. Freeing lock request will free lock
context also.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
This makes sure that when the client context is destroyed, the lock
request goes away. If the lock requests is already scheduled, then the
lock child process will be terminated.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
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>
This avoids traversing a single pending queue which is quite expensive
when there are lots of pending lock requests. This seems to happen
quite a lot on a loaded cluster for notify_index.tdb.
Adding per database queues avoids the need to traverse pending queue
for that database if there are already the maximum number of active
lock requests.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Aug 4 20:23:45 CEST 2014 on sn-devel-104
The number of pending locks displayed in ctdb statistics are stored in
ctdb_statistics structure and not ctdb_context.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
This allows to change the maximum number of lock processes that can
be active.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
To enable TDB mutex support, set tunable TDBMutexEnabled=1.
When databases are attached for the first time, attach flags must include
TDB_MUTEX_LOCKING and TDBMutexEnabled must set to enable mutex support.
However, when CTDB attaches databases internally for recovery, it will
enable mutex support if TDBMutexEnabled is set.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Wed Jul 9 06:45:17 CEST 2014 on sn-devel-104
This prevents ctdb tool from thawing databases prematurely in
thaw/wipedb/restoredb commands if recovery is active.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
This duplicates ctdb->ctdbd_pid.
Thanks to Sumit Bose <sbose@redhat.com> for the suggestion.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
When calling sched_setscheduler() with SCHED_OTHER, the only valid
priority is 0. Nice value is "restored" anyway.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Database priority is a global property and all the nodes should have the
priority set for the databases. Just setting priority on one node can
lead to problems in the recovery as a database can be frozen at wrong
priority and then freezing database would not succeed.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Mon Apr 7 14:06:26 CEST 2014 on sn-devel-104
reloadips really expects deleted IPs to be released before completing.
Otherwise the recovery daemon starts failing the local IP check. The
races that follow can cause a node to be banned.
To make the error handling simple, do the actual deletion in
release_ip_callback().
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Any currently running monitor events are cancelled if any other events
are scheduled. However, this does not stop monitor events to be run
when other events are already running.
Keep track of the number of active events and schedule monitor event
only if there are no active events.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
At the moment ctdb_check_healthy() is overloaded to wait until the
first recovery is complete, handle the "startup" event and also
actually handle monitoring. This is untidy and hard to follow.
Instead, have the daemon explicitly wait for 1st recovery after the
"setup" event. When first recovery is complete, schedule a function
to handle the "startup" event. When the "startup" event succeeds then
explicitly enable monitoring.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This function has been replaced with ctdb_vfork_with_logging().
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): Thu Jan 16 04:05:35 CET 2014 on sn-devel-104
This will be used to spawn lightweight helper processes to run
eventscripts.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
This was added to support external monitoring using CTDB event scripts.
However, it was never used.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
This code can then be used to track child processes created with vfork().
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
This limit was currently a global limit and not per database. This
prevents any database freeze lock requests from getting scheduled if
the global limit was reached.
Only individual record requests should be limited and database freeze
requests should always get scheduled.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
When running a mixed version cluster, compatibility with older
versions was was broken during recent refactorisation.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
Behaves like mkdir -p.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit afe2145d91725daf1399f0a24f1cddcf65f0ec31)
No need to pass it as an extra argument to ctdb_start_daemon.
Also ensure options.public_address_list gets a nice static default.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit a3d63a9db89d08bb284b3b3a6db773422f21b477)
Also get rid of ctdb_set_event_script_dir(). It creates an
unnecessary copy of something that will be around for the lifetime of
the process.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 21b4d1aba00902f1eee0cbf4f082b0794fd5b738)
This allows ctdb_load_nodes_file() to move to ctdb_server.c and
ctdb_set_nlist() to become static.
Setting ctdb->nodes_file needs to be done early, before the nodes file
is loaded. It is now set from CTDB_BASE instead ETCDIR, so setting
CTDB_BASE also needs to be done earlier.
Unhack ctdbd_test.c - it no longer needs to define
ctdb_load_nodes_file().
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 20e705e63bd3b20837cc3ac92fdcf2a9650ccfc8)
This removes data types and structure elements related to TRANS2
persistent transaction code.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 22a253b7ccf1ff854cddf0b67969dc84d7d6a654)
The current implementation has a few flaws:
* A takeover run is called unconditionally when the timer goes even if
the recovery master role has moved. This means a node other than
the recovery master can incorrectly do a takeover run.
* The rebalancing target nodes are cleared in the setup for a takeover
run, regardless of whether the takeover run succeeds.
* The timer to force a rebalance isn't cleared if another takeover run
occurs before the deadline. Any forced rebalancing will happen in
the first takeover run and when the timer expires some time later
then an unnecessary takeover run will occur.
* If the recovery master role moves then the rebalancing data will
stay on the original node and affect the next takeover run to occur
if the recovery master role should come back to the original node.
Instead, store an array of rebalance target nodes in the recovery
master context. This is passed as an extra argument to
ctdb_takeover_run() each time it is called and is cleared when a
takeover run succeeds. The timer hangs off the array of rebalance
target nodes, which is cleared if the node isn't the recovery master.
This means that it is possible to lose rebalance data if the recovery
master role moves. However, that's a difficult problem to solve. The
best way of approaching it is probably to try to stop the recovery
master role from jumping around unnecesarily when inactive nodes join
the cluster.
The long term solution is to avoid this nonsense completely. The IP
allocation algorithm needs to cache state between runs so that it
knows which nodes have just become healthy. This also needs recovery
master stability.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit c51c1efe5fc7fa668597f2acd435dee16e410fc9)
No need for a separate one for each SRVID.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit d9c22b04d5aa7938a3965bd3144568664eb772ce)
This is an internal structure. It was moved into ctdb_private.h a
long time ago to allow unit testing. Unit test compilation was
changed shortly afterwards to make this unnecessary.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit db57261d7dc264e161659a8c547f44fbd9e88eeb)
This reverts commit 035c0d981bde8c0eee8b3f24ba8e2dc817e5b504.
This is a premature optimization. Record can bounce between nodes
very quickly if it is a contended record. There is no need to hold a
record on a node unnecessarily. In case record contention becomes bad,
enabling sticky records on a database is a better idea.
Conflicts:
include/ctdb_private.h
server/ctdb_tunables.c
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit ac417b0003f0116f116834ad2ac51482d25cfa0d)
The structure cannot be removed without adding support for marshalling keys
for hot records.
This reverts commit 26a4653df594d351ca0dc1bd5f5b2f5b0eb0a9a5.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 023ca2e84f5ed064a288526b9c2bc7e06674dd81)
The code for deadlock detection and killing smbd process causing deadlock
has been removed and replaced with external debug script.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 2211cd94bea266547d3e6f167d3160a6b23bec88)
Send the ctdb_db_statistics directly instead of first copying it to
duplicate ctdb_db_statistics_wire structure. This simplifies the
implementation of the control to get database statistics.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 26a4653df594d351ca0dc1bd5f5b2f5b0eb0a9a5)
The log messages in verify_remote_ip_allocation() are confusing
because they don't include the PNN of the problem node, because it is
not known in this function.
Add the PNN of the node being verified as a function argument and then
shuffle the log messages around to make them clearer.
Also fold 3 nested if statements into just one.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit f0942fa01cd422133fc9398f56b4855397d7bc86)
This is like ctdb_fatal() but exits cleanly without dumping core or
generating a backtrace.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit c0a9456692c88a7a5542cd893d8f326524d3f94e)
When this function is called, we are already committed to banning
and there is no point in failing this function. In case, freezing of
databases fails, it will be fixed from recovery daemon.
(This used to be ctdb commit bb178338658b4ae32382a1f62f7c21cee1d4878f)
If this function fails due to memory errors, there is no way to recover.
The best course of action is to abort.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 46efe7a886f8c4c56f19536adc98a73c22db906a)
This adds more serialisation to the startup, ensuring that the
"startup" event runs after everything to do with the first recovery
(including the "recovered" event).
Given that it now takes longer to get to the "startup" state, the
initscript needs to wait until ctdbd gets to "first_recovery".
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit ed6814ff0a59ddbb1c1b3128b505380f60d7aeb7)
This allows states, including startup and shutdown states, to be
clearly tracked. This doesn't include regular runtime "states", which
are handled by node flags.
Introduce new functions ctdb_set_runstate(), runstate_to_string() and
runstate_from_string().
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 8076773a9924dcf8aff16f7d96b2b9ac383ecc28)
These functions were used in locking child process to do the locking. With
locking helper, these are not required.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit c660f33c3eaa1b4a2c4e951c1982979e57374ed4)
This really needs to be per-node. The rename is because nodes with
this tunable switched on should drop IPs if they become unhealthy (or
disabled in some other way).
* Add new flag NODE_FLAGS_NOIPHOST, only used in recovery daemon.
* Enhance set_ipflags_internal() and set_ipflags() to setup
NODE_FLAGS_NOIPHOST depending on setting of NoIPHostOnAllDisabled
and/or whether nodes are disabled/inactive.
* Replace can_node_servce_ip() with functions can_node_host_ip() and
can_node_takeover_ip(). These functions are the only ones that need
to look at NODE_FLAGS_NOIPTAKEOVER and NODE_FLAGS_NOIPHOST. They
can make the decision without looking at any other flags due to
previous setup.
* Remove explicit flag checking in IP allocation functions (including
unassign_unsuitable_ips()) and just call can_node_host_ip() and
can_node_takeover_ip() as appropriate.
* Update test code to handle CTDB_SET_NoIPHostOnAllDisabled.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 1308a51f73f2e29ba4dbebb6111d9309a89732cc)
It isn't used, superceded by "ipreallocated".
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit c2bb8596a8af6406ef50e53953884df9d6246a96)
This is an alternative to using ctdb_run_eventscripts() that can be
used when in recovery.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 27a44685f0d7a88804b61a1542bb42adc8f88cb1)
This in preparation of turning the vacuming on the lmaster into
into a two phase process:
- First the node sends the list of records to be vacuumed
to all other nodes with this new RECEIVE_RECORDS control.
The remote nodes should store the lmaster's empty current copy.
- Only those records that could be stored on all other nodes
are processed further. They are send to all other nodes with
the TRY_DELETE_RECORDS control as before for deletion.
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-By: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit e397702e271af38204fd99733bbeba7c1db3a999)
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit 1e989894764e4cd1d551c44784d91cb295cd790d)
It really is internal.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit abb64f62efaa70df4b87c030b96300eafd98e6a3)
Default is not to create a pid file.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit 996e74d3db0c50f91b320af8ab7c43ea6b1136af)
Must be called by all child processes.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit 59b019a97aad9a731f9080ea5be14d0dbdfe03d6)
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 2e92deef5221ee651028ef87138b3113f1fece91)
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit e691df43d20871468142c8fb83f7c7303c4ec307)
When CTDB is busy with lots of smbd, CTDB was spending too much time in
daemon_check_srvids() which searches a list of srvids in the registered
message handlers. Using a hash based index significantly improves the
performance of search in a linked list.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 3e09f25d419635f6dd679b48fa65370f7860be7d)
Commit a82d3ec12f0fda16d6bfa8442a07595de897c10e broke fetching from
the log ringbuffer. The solution there is still generally good: there
is no need to keep the ringbuffer in children created by
ctdb_fork()... except for those special children that are created to
fetch data from the ringbuffer!
Introduce a new function ctdb_fork_no_free_ringbuffer() that does
everything ctdb_fork() needs to do except free the ringbuffer (i.e. it
is the old ctdb_fork() function). The new ctdb_fork() function just
calls that function and then frees the ringbuffer in the child.
This means all callers of ctdb_fork() have the convenience of having
the ringbuffer freed. There are 3 special cases:
* Forking the recovery daemon. We want to be able to fetch from the
ringbuffer there.
* The ringbuffer fetching code. Change the 2 calls in this code (main
daemon, recovery daemon) to call ctdb_fork_no_free_ringbuffer()
instead.
While we're here, clear the log ringbuffer when the recovery deamon is
forked, since it will contain a copy of the messages from the main
daemon.
Note to self: always test... even the most obvious patches... ;-)
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 00db5fa00474f8a83f1aa3b603fd756cc9b49ff4)
Use an environment variable instead. This just means that the
initscript exports CTDB_DEBUG_HUNG_SCRIPT and the code checks for the
environment variable.
The justification for this simplification is that more debug options
will be arriving soon and we want to handle them consistently without
needing to add a command-line option for each. So, the convention
will be to use an environment variable for each debug option.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 0581f9a84e58764d194f4e04064c2c5b393c348b)
The only allocation against this context is by
ctdb_fork_with_logging(). This memory is freed by ctdb_log_handler()
anyway. There should be no memory leak.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 501461cc3e132d4adee9e91b5d4513a26bae2846)
When CTDB is shut down and monitoring has been stopped, monitor_context
gets freed and all the callback states hanging off it. This includes
callback state for current_monitor, if the current monitor event has
not yet finished. As a result, when the shutdown event is called,
current_monitor->callback state is not NULL, but it's actually freed
and it's a dangling reference.
So before executing callback function and freeing callback state check
if ctdb->monitor->monitor_context is not NULL.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 7d8546ee4353851f0543d0ca2c4c67cb0cc75aea)
When CTDB is shutting down, recovery daemon is stopped, but the
event that checks if recovery daemon is still alive is not destroyed.
So recovery master is restarted during shutdown if CTDB daemon takes
longer to shutdown.
There are two processes that check if recovery daemon is working.
1. ctdb_check_recd() - which checks every 30 seconds if the recovery
daemon process exists.
2. ctdb_recd_ping_timeout() - which is triggered when recovery daemon
fails to ping CTDB daemon.
Both the events are periodic and need to be destroyed when shutting down.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 746168df2e691058e601016110fae818c6a265c3)
This effectively reverts d96cb02c2c24f9eabbc53d3d38e90dea49cff3e0
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 496387a585b2c5778c808cf02b8e1435abde4c3e)
Samba versions 3.6.x and older do not set the database priority.
This can cause deadlock between Samba and CTDB since the locking order
of database will be different. A hack was added for automatic promotion
of priority for specific databases to avoid deadlock. This code should
not be invoked with Samba version 4.x which correctly specifies the
priority for each database.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit 4a9e96ad3d8fc46da1cd44cd82309c1b54301eb7)
This fixes a bug where wrong variable is checked.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit f81e9add466b1d9b2796c09c6ba63b77296ea149)
RECLOCK is for recovery lock in CTDB. Do not override the meaning for
tracking locks on databases. Database lock latency has nothing to do
with recovery lock latency.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 54e24a151d2163954e5a2a1c0f41a2b5c19ae44b)
If any of the nodes fail takeover run (either due to timeout or failure
to complete within takeover_timeout interval) from main loop, recovery
master will give up trying takeover run with following message:
"Unable to setup public takeover addresses. Try again later"
And as a side-effect the monitoring is disabled on all the nodes. Before
ctdb_takeover_run() is called from main loop, monitoring get disabled via
startrecovery event. Since ctdb_takeover_run() fails, it never runs
recovered event and monitoring does not get re-enabled.
In main_loop, ctdb_takeover_run() is called with a takeover_fail_callback.
This callback will get called if any of the nodes fail in handling
takeip/releaseip/ipreallocated events in ctdb_takeover_run().
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit a5c6bb1fffb8dc3960af113957a1fd080cc7c245)
These support getting and clearing logs from the ring-buffer in the
recovery daemon.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit cbca233d1e03b2410e0bb63b936328d4a8b3c7b4)
This introduces a consistent API for handling locks on single record, complete
db or all dbs. The locks are taken out in a child process. In cases of timeout,
find the processes that currently hold the lock and log.
Callback functions for locking requests take locked boolean to indicate
whether the lock was successfully obtained or not.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit 1af99cf0de9919dd89af1feab6d1bd18b95d82ff)
Currently these functions are implemented only for Linux.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
(This used to be ctdb commit be4051326b0c6a0fd301561af10fd15a0e90023b)
We've seen this function report "Unknown family, 0" and then CTDB
disappeared without a trace. If we can reproduce it then this might
help us to debug it.
The idea is that you do something like the following in /etc/sysconfig/ctdb:
export CTDB_EXTERNAL_TRACE="/etc/ctdb/config/gcore_trace.sh"
When we hit this error than we call out to gcore to get a core file so
we can do forensics. This might block CTDB for a few seconds.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 7895bc003f087ab2f3181df3c464386f59bfcc39)
There's a race here where release and takeover events for an IP can
run at the same time. For example, a "ctdb deleteip" and a takeover
initiated by the recovery daemon. The timeline is as follows:
1. The release code registers a callback to update the VNN. The
callback is executed *after* the eventscripts run the releaseip
event.
2. The release code calls the eventscripts for the releaseip event,
removing IP from its interface.
The takeover code "updates" the VNN saying that IP is on some
iface.... even if/though the address is already there.
3. The release callback runs, removing the iface associated with IP in
the VNN.
The takeover code calls the eventscripts for the takeip event,
adding IP to an interface.
As a result, CTDB doesn't think it should be hosting IP but IP is on
an interface. The recovery daemon fixes this later... but it
shouldn't happen.
This patch can cause some additional noise in the logs:
Release of IP 10.0.2.133/24 on interface eth2 node:2
recoverd:We are still serving a public address '10.0.2.133' that we should not be serving. Removing it.
Release of IP 10.0.2.133/24 rejected update for this IP already in flight
recoverd:client/ctdb_client.c:2455 ctdb_control for release_ip failed
recoverd:Failed to release local ip address
In this case the node has started releasing an IP when the recovery
daemon notices the addresses is still hosted and initiates another
release. This noise is harmless but annoying.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit bfe16cf69bf2eee93c0d831f76d88bba0c2b96c2)
Stops the behaviour where unhealthy nodes can host IPs when there are
no healthy nodes. Set this to 1 when an immediate complete outage is
preferred when all nodes are unhealthy. The alternative
(i.e. default) can lead to undefined behaviour when the shared
filesystem is unavailable.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit a555940fb5c914b7581667a05153256ad7d17774)
Change this to instead preallocate , by default, 10MByte chunks to the data buffer.
This significantly reduces the number of potential reallocate and move operations that may be required.
Create a tunable to override/change how much preallocation should be used.
(This used to be ctdb commit 1f262deaad0818f159f9c68330f7fec121679023)
Add tunables to control when to log these instances and allow it to be completely turned off by setting the threshold to 0
(This used to be ctdb commit 9ed58fef4991725f75509433496f4d5ffae0ae87)
Break this debug and datacollection out into an external script to make it easier to modify what data we need to collect.
For now we only collect a pstree so we can see what part of the script we hung in.
S1037271
(This used to be ctdb commit 6e68797af67bee36f2bad045f94806e7e98f27e9)
Wrap all creation of child processes inside ctdb_fork() which is used to track all processes we have spawned.
Capture SIGCHLD to track also which child processes have terminated.
Wrap kill() inside ctdb_kill() and make sure that we never send a !0 signal to a child process pid that has already terminated (and might have been replaced with a
(This used to be ctdb commit f73a4b1495830bcdd094a93732a89dd53b3c2f78)
and also update the "read public address file" to not check if the address exists already locally when we read if from the child process, to stop it
from spamming the logs with "We already host ..."
messages
(This used to be ctdb commit 334ea830f1bf33419f4a1e78f23afd41a852d0f4)
Also add a method to use the recovery master/daemon to reload the public ips on all nodes in the cluster.
Reloading the public ips on all node sin the cluster is only suported if all nodes in the cluster are available and healthy.
(This used to be ctdb commit 05603e914f8c12618d7e06943c0f7df207f645b0)
The implementation of DisableIPFailover got intermingled with
--nopublicipcheck. This just looks wrong - Ronnie must have been
having a bad day. :-)
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 5083b266dd68b292c4275505f3d1b878dbf12f11)
This can improve performance and stop clients from having to chase a rapidly migrating/bouncing record
(This used to be ctdb commit d0d98f7e45e5084b81335b004d50bddc80cdc219)
This can improve performance slightly on certain workloads where smbds frequently read from the same record
(This used to be ctdb commit 035c0d981bde8c0eee8b3f24ba8e2dc817e5b504)
Instead, tie them together via referencing a permanent linked list hung off the ctdb structure.
(This used to be ctdb commit a95c02da6c67dc4bd8716b75318a4188301df6f9)
there are some child processes where we do not create a connection to the main daemon (switch_from_server_to_client()) because it is expensive to set up and we normally might not need to talk to the daemon at all via a domainsocket.
but we might want to still call to ctdb_ltdb_store() from such chil processes.
(This used to be ctdb commit 9e372a08c40087e6b5335aa298e94d88273566a5)
By this, the original CTDB_CONTROL_TRAVERSE_START control that is
used by e.g. samba's smbstatus, is not changed, so that samba
continues working without code change.
The CTDB_CONTROL_TRAVERSE_START currently just adds the "withemptyrecords"
flag to the state and processon on as CTDB_CONTROL_TRAVERSE_START_EXT.
(This used to be ctdb commit 8281bb210858ed04992eacea7f6d02261e0fc1b1)
Add a new tunable that changes the mode how persistent databases are recovered.
RecoveryPDBBySeqNum
When set to 1, persistent databases will be recovered in whole from the node which
has the highest "__db_sequence_number__" record.
This record is managed by samba for those databases where we do persistent writes and have
inter-record relations.
For these databases we do not want the usual "blend records from all nodes based
on individual record RSN" but instead a mode where we pick one instance of the persistent database.
If no node was found with a "__db_sequence_number__" record at all, we fail back to the original "recover records independently based on record RSN".
Some persistent databases do not contain record interrelations and as such does not
contain this special record at all.
(This used to be ctdb commit 502150c764298a9fa8c4d8aa445bf7d85d4ee9dc)
These were intentionally not static so they could be linked to in unit
test programs. However, using the CCAN-style unit tests where
relevant code is just included, this is no longer necessary.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit d0e9e8554614bd49ffb9ec3509feaa0e80d0f65d)
Move identical copies of ctdb_null_func(), ctdb_fetch_func(),
ctdb_fetch_with_header_func() from ctdb_client.c and
ctdb_ltdb_server.c to somewhere common.
This is in the context of wanting to run CCAN-style tests where most
of the ctdbd code is just included in the test program.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 126cb0d369b2b1aed63801dc4ba0554399e8b7e4)
When multiple clients fetch the same record concurrently, send only one single
fetch across the network and deferr all other fetches locally.
This improves performance for hot records and reduces cpu load on ctdb.
(This used to be ctdb commit 82d6946ad8b3348e8b9d3d971f24925ade02d1be)
Attempt to reconnect to ctdbd on fetch while it is unreachable.
We must provide our own queue callback wrapper, as ctdb_client_read_cb()
exits on transport failure.
(This used to be ctdb commit 28df6fbf1273b8d095a2bc38dca6a6c35c5c31bd)
This will make it much easier to root-cause problems such as
S1029023
when an external application deleted the interface while it is still is in use by ctdbd.
(This used to be ctdb commit 9abf9c919a7e6789695490e2c3de56c21b63fa57)
check that the actual interface exist, print error and fail startup if the interface does not exist.
(This used to be ctdb commit cd33bbe6454b7b0316bdfffbd06c67b29779e873)
When set to 0, clients will not be able to attach to databases
via the db_attach control. This might can be useful for maintenance
where ctdb should be kept running but clients should not be able
to modify databases.
(This used to be ctdb commit ddfeecda87955b4e46777599f678e6926d37f4c4)
let all databases default to not support this until enabled through this control
(This used to be ctdb commit 908a07c42e5135a3ba30a625fc4f4e4916de197a)
This triggers a child process to be created to perform the actual potentially blocking calls that are required.
(This used to be ctdb commit 7d575ee92c95bc4aab78a33bc1aac7ff0811ab3a)
Once the contexts are freed, the deferred calls are re-issued to the input packet processing functions again.
This is needed when/if a CALL can not currently be processed by the main engine due to the record being locked down for revoking of all delegations.
The data is passed through several layers of callbacks, and finally a timed event callback to ensure that the processing of the packet will be restarted again at the topmost eventloop, avoinding event loop nesting.
(This used to be ctdb commit cc6f78efcfa3b8caeffbd68018e6dfbf81488dce)
Assume all databases will support readonly mode for now and se thte flag for all databases. At later stage we will add support to control on a per database level whether delegations will be supported or not.
(This used to be ctdb commit 502f86f79944df4bac9094f716e54110c511dc24)
Move struct ctdb_public_ip_list to ctdb_private.h and put some
definitions for some functions from ctdb_takeover.c there. This
allows those functions to be called from unit tests.
Add ctdb_takeover_tests.c and the Makefile support to build it.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 9d34be0233edf3bc022345c0494c4b2a4d7f8480)
The current non-deterministic IP allocation algorithm balances IPs
across the whole cluster. It does not consider different
interfaces/VLANs/subnets, so these different groups of IPs aren't
generally well balanced.
This adds the LCP2 algorithm for IP allocation and allows it to be
enabled by setting the "LCP2PublicIPs" tunable to 1.
The LCP2 algorithm calculates the imbalance of a node by totalling the
squares of the distances between each IP on the node. The IP distance
is defined as the length longest common prefix (LCP) of bits that is
found when comparing 2 IPs. The imbalance of a cluster is the maximum
imbalance for any node. At each step the algorithm selects an
allocation to the IP/node combination that results in the choosing the
allocation that best reduces the imbalance of the cluster.
The implementation splits out the IP allocation part of
ctdb_takeover_run() into new function ctdb_takeover_run_core(), and
then extracts out the basic IP assignment code into new functions
basic_allocate_unassigned() and basic_failback(). 3 new functions
lcp2_init(), lcp2_allocate_unassigned() and lcp2_failback() implement
the LCP2 algorithm, and are hooked into ctdb_takeover_run_core().
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 61fc7fbd0235469df22deb6581c6bd47e30bc0be)
This is realized by adding a ctdb_ltdb_store_fn function pointer to the db
context and filling it in the attach procedure for non-persistent dbs.
(This used to be ctdb commit df49ec44de80affa5ccc637dec12a20a26e8706e)
