rpl.rpl_seconds_behind_master_spike uses the DEBUG_SYNC mechanism to
count how many format descriptor events (FDEs) have been executed,
to attempt to pause on a specific relay log FDE after executing
transactions. However, depending on when the IO thread is stopped,
it can send an extra FDE before sending the transactions, forcing
the test to pause before executing any transactions, resulting in a
table not existing, that is attempted to be read for COUNT.
This patch fixes this by no longer counting FDEs, but rather by
programmatically waiting until the SQL thread has executed the
transaction and then automatically activating the DEBUG_SYNC point
to trigger at the next relay log FDE.
Improve the performance of slave connect using B+-Tree indexes on each binlog
file. The index allows fast lookup of a GTID position to the corresponding
offset in the binlog file, as well as lookup of a position to find the
corresponding GTID position.
This eliminates a costly sequential scan of the starting binlog file
to find the GTID starting position when a slave connects. This is
especially costly if the binlog file is not cached in memory (IO
cost), or if it is encrypted or a lot of slaves connect simultaneously
(CPU cost).
The size of the index files is generally less than 1% of the binlog data, so
not expected to be an issue.
Most of the work writing the index is done as a background task, in
the binlog background thread. This minimises the performance impact on
transaction commit. A simple global mutex is used to protect index
reads and (background) index writes; this is fine as slave connect is
a relatively infrequent operation.
Here are the user-visible options and status variables. The feature is on by
default and is expected to need no tuning or configuration for most users.
binlog_gtid_index
On by default. Can be used to disable the indexes for testing purposes.
binlog_gtid_index_page_size (default 4096)
Page size to use for the binlog GTID index. This is the size of the nodes
in the B+-tree used internally in the index. A very small page-size (64 is
the minimum) will be less efficient, but can be used to stress the
BTree-code during testing.
binlog_gtid_index_span_min (default 65536)
Control sparseness of the binlog GTID index. If set to N, at most one
index record will be added for every N bytes of binlog file written.
This can be used to reduce the number of records in the index, at
the cost only of having to scan a few more events in the binlog file
before finding the target position
Two status variables are available to monitor the use of the GTID indexes:
Binlog_gtid_index_hit
Binlog_gtid_index_miss
The "hit" status increments for each successful lookup in a GTID index.
The "miss" increments when a lookup is not possible. This indicates that the
index file is missing (eg. binlog written by old server version
without GTID index support), or corrupt.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
A debug_sync signal could remain for the SQL thread that should have begun
a wait_for upon seeing a GTID event, but would instead see the old signal
and continue on without waiting. This broke an "idle" condition in
SHOW SLAVE STATUS
which should have automatically negated Seconds_Behind_Master. Instead,
because the SQL thread had already processed the GTID event, it set
sql_thread_caught_up to false, and thereby calculated the value of
Seconds_behind_master, when the test expected 0.
This patch fixes this by resetting the debug_sync state before creating a
new transaction which sends a GTID event to the replica
rpl_semi_sync_slave_enabled_consistent.test and the first part of
the commit message comes from Brandon Nesterenko.
A test to show how to induce the "Read semi-sync reply magic number
error" message on a primary. In short, if semi-sync is turned on
during the hand-shake process between a primary and replica, but
later a user negates the rpl_semi_sync_slave_enabled variable while
the replica's IO thread is running; if the io thread exits, the
replica can skip a necessary call to kill_connection() in
repl_semisync_slave.slave_stop() due to its reliance on a global
variable. Then, the replica will send a COM_QUIT packet to the
primary on an active semi-sync connection, causing the magic number
error.
The test in this patch exits the IO thread by forcing an error;
though note a call to STOP SLAVE could also do this, but it ends up
needing more synchronization. That is, the STOP SLAVE command also
tries to kill the VIO of the replica, which makes a race with the IO
thread to try and send the COM_QUIT before this happens (which would
need more debug_sync to get around). See THD::awake_no_mutex for
details as to the killing of the replica’s vio.
Notes:
- The MariaDB documentation does not make it clear that when one
enables semi-sync replication it does not matter if one enables
it first in the master or slave. Any order works.
Changes done:
- The rpl_semi_sync_slave_enabled variable is now a default value for
when semisync is started. The variable does not anymore affect
semisync if it is already running. This fixes the original reported
bug. Internally we now use repl_semisync_slave.get_slave_enabled()
instead of rpl_semi_sync_slave_enabled. To check if semisync is
active on should check the @@rpl_semi_sync_slave_status variable (as
before).
- The semisync protocol conflicts in the way that the original
MySQL/MariaDB client-server protocol was designed (client-server
send and reply packets are strictly ordered and includes a packet
number to allow one to check if a packet is lost). When using
semi-sync the master and slave can send packets at 'any time', so
packet numbering does not work. The 'solution' has been that each
communication starts with packet number 1, but in some cases there
is still a chance that the packet number check can fail. Fixed by
adding a flag (pkt_nr_can_be_reset) in the NET struct that one can
use to signal that packet number checking should not be done. This
is flag is set when semi-sync is used.
- Added Master_info::semi_sync_reply_enabled to allow one to configure
some slaves with semisync and other other slaves without semisync.
Removed global variable semi_sync_need_reply that would not work
with multi-master.
- Repl_semi_sync_master::report_reply_packet() can now recognize
the COM_QUIT packet from semisync slave and not give a
"Read semi-sync reply magic number error" error for this case.
The slave will be removed from the Ack listener.
- On Windows, don't stop semisync Ack listener just because one
slave connection is using socket_id > FD_SETSIZE.
- Removed busy loop in Ack_receiver::run() by using
"Self-pipe trick" to signal new slave and stop Ack_receiver.
- Changed some Repl_semi_sync_slave functions that always returns 0
from int to void.
- Added Repl_semi_sync_slave::slave_reconnect().
- Removed dummy_function Repl_semi_sync_slave::reset_slave().
- Removed some duplicate semisync notes from the error log.
- Add test of "if (get_slave_enabled() && semi_sync_need_reply)"
before calling Repl_semi_sync_slave::slave_reply().
(Speeds up the code as we can skip all initializations).
- If epl_semisync_slave.slave_reply() fails, we disable semisync
for that connection.
- We do not call semisync.switch_off() if there are no active slaves.
Instead we check in Repl_semi_sync_master::commit_trx() if there are
no active threads. This simplices the code.
- Changed assert() to DBUG_ASSERT() to ensure that the DBUG log is
flushed in case of asserts.
- Removed the internal rpl_semi_sync_slave_status as it is not needed
anymore. The @@rpl_semi_sync_slave_status status variable is now
mapped to rpl_semi_sync_enabled.
- Removed rpl_semi_sync_slave_enabled as it is not needed anymore.
Repl_semi_sync_slave::get_slave_enabled() contains the active status.
- Added checking that we do not add a slave twice with
Ack_receiver::add_slave(). This could happen with old code.
- Removed Repl_semi_sync_master::check_and_switch() as it is not
needed anymore.
- Ensure that when we call Ack_receiver::remove_slave() that the slave
is removed from the listener before function returns.
- Call listener.listen_on_sockets() outside of mutex for better
performance and less contested mutex.
- Ensure that listening is ignoring newly added slaves when checking for
responses.
- Fixed the master ack_receiver listener is not killed if there are no
connected slaves (and thus stop semisync handling of future
connections). This could happen if all slaves sockets where would be
marked as unreliable.
- Added unlink() to base_ilist_iterator and remove() to
I_List_iterator. This enables us to remove 'dead' slaves in
Ack_recever::run().
- kill_zombie_dump_threads() now does killing of dump threads properly.
- It can now kill several threads (should be impossible but could
happen if IO slaves reconnects very fast).
- We now wait until the dump thread is done before starting the
dump.
- Added an error if kill_zombie_dump_threads() fails.
- Set thd->variables.server_id before calling
kill_zombie_dump_threads(). This simplies the code.
- Added a lot of comments both in code and tests.
- Removed DBUG_EVALUATE_IF "failed_slave_start" as it is not used.
Test changes:
- rpl.rpl_session_var2 added which runs rpl.rpl_session_var test with
semisync enabled.
- Some timings changed slight with startup of slave which caused
rpl_binlog_dump_slave_gtid_state_info.text to fail as it checked the
error log file before the slave had started properly. Fixed by
adding wait_for_pattern_in_file.inc that allows waiting for the
pattern to appear in the log file.
- Tests have been updated so that we first set
rpl_semi_sync_master_enabled on the master and then set
rpl_semi_sync_slave_enabled on the slaves (this is according to how
the MariaDB documentation document how to setup semi-sync).
- Error text "Master server does not have semi-sync enabled" has been
replaced with "Master server does not support semi-sync" for the
case when the master supports semi-sync but semi-sync is not
enabled.
Other things:
- Some trivial cleanups in Repl_semi_sync_master::update_sync_header().
- We should in 11.3 changed the default value for
rpl-semi-sync-master-wait-no-slave from TRUE to FALSE as the TRUE
does not make much sense as default. The main difference with using
FALSE is that we do not wait for semisync Ack if there are no slave
threads. In the case of TRUE we wait once, which did not bring any
notable benefits except slower startup of master configured for
using semisync.
Co-author: Brandon Nesterenko <brandon.nesterenko@mariadb.com>
This solves the problem reported in MDEV-32960 where a new
slave may not be registered in time and the master disables
semi sync because of that.
While a replica may be reading events from the
primary, the primary is killed. Left to its own
devices, the IO thread may or may not stop in
error, depending on what it is doing when its
connection to the primary is killed (e.g. a
failed read results in an error, whereas if the
IO thread is idly waiting for events when the
connection dies, it will enter into a reconnect
loop and reconnect). MDEV-32168 changed the test
to always wait for the reconnect, thus breaking
the error case, as the IO thread would be stopped
at a time of expecting it to be running.
The fix is to manually stop/start the IO thread
to ensure it is in a consistent state.
Note that rpl_domain_id_filter_master_crash.test
will need additional changes after fixing MDEV-33268
Reviewed By:
============
Kristian Nielsen <knielsen@knielsen-hq.org>
This patch augments Gtid_log_event with the user thread-id.
In particular that compensates for the loss of this info in
Rows_log_events.
Gtid_log_event::thread_id gets visible in mysqlbinlog output like
#231025 16:21:45 server id 1 end_log_pos 537 CRC32 0x1cf1d963 GTID 0-1-2 ddl thread_id=10
as 64 bit unsigned integer.
While the size of Gtid event has grown by 8-9 bytes
replication from OLD <-> NEW is not affected by it.
This work was started by the late Sujatha Sivakumar.
Brandon Nesterenko took it over, reviewed initial patches and extended
the work.
Reviewed-by: <andrei.elkin@mariadb.com>
Field_new_decimal::store_value(const my_decimal*, int*)
Analysis
========
When rpl applier is unpacking a before row image, Field::reset() will be
called before setting a field to null if null bit of the field is set in
the row image. For Field_new_decimal::reset(), it calls
Field_new_decimal::store_value() to reset the value. store_value() asserts
that the field is in the write_set bitmap since it thinks the field is
updating.
But that is not true for the row image generated in FULL_NODUP
mode. In the mode, the before image includes all fields and the after
image includes only updated fields.
Fix
===
In the case unpacking binlog row images, the assertion is meaningless.
So the unpacking field is marked in write_set temporarily to avoid the
assertion failure.
Item::val_str() sets the Item::null_value flag, so call it before checking
the flag, not after.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
fix test cases like
... prepare the test, create a table or a user
create event ... on schedule every 1 second ... do something ...
... verify that something was done
... cleanup drop event, table or user
here the event scheduler can manage to start the worker for the
second execution of the event, then cleanup happens
and the worker realizes that it cannot change the security context
(as the user is dropped) or cannot insert (as the table is dropped).
followup for 279b0db8c6 (that removed suppression of /Event Schedule/)
AKA rpl.rpl_parallel, binlog_encryption.rpl_parallel fails in
buildbot with timeout in include
A replication parallel worker thread can deadlock with another
connection running SHOW SLAVE STATUS. That is, if the replication
worker thread is in do_gco_wait() and is killed, it will already
hold the LOCK_parallel_entry, and during error reporting, try to
grab the err_lock. SHOW SLAVE STATUS, however, grabs these locks in
reverse order. It will initially grab the err_lock, and then try to
grab LOCK_parallel_entry. This leads to a deadlock when both threads
have grabbed their first lock without the second.
This patch implements the MDEV-31894 proposed fix to optimize the
workers_idle() check to compare the last in-use relay log’s
queued_count==dequeued_count for idleness. This removes the need for
workers_idle() to grab LOCK_parallel_entry, as these values are
atomically updated.
Huge thanks to Kristian Nielsen for diagnosing the problem!
Reviewed By:
============
Kristian Nielsen <knielsen@knielsen-hq.org>
Andrei Elkin <andrei.elkin@mariadb.com>
The problem is that these tests run optimistic parallel replication with
non-transactional mysql.gtid_slave_pos table. Very occasionally InnoDB stats
update may race with one another and cause a parallel replication deadlock
kill after the mysql.gtid_slave_pos table has been updated. Since
mysql.gtid_slave_pos is non-transactional, the update cannot be rolled back,
and transaction retry will fail with a duplicate key error in
mysql.gtid_slave_pos.
Fixed by altering the storage engine to InnoDB for the table.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
Add a test case that demonstrates a working setup as described in MDEV-26632.
This requires --gtid-ignore-duplicates=1 and --gtid-strict-mode=0.
In A->B->C, B filters some (but not all) events from A. C is promoted to
create A->C->B, and the current GTID position in B contains a GTID from A that
is not present in C (due to filtering). Demonstrate that B can still connect
with GTID to C, starting at the "hole" in the binlog stream on C originating
from A.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
Fix some random test failures following MDEV-32168 push.
Don't blindly set $rpl_only_running_threads in many places. Instead explicit
stop only the IO or SQL thread, as appropriate. Setting it interfered with
rpl_end.inc in some cases. Rather than clearing it afterwards, better to
not set it at all when it is not needed, removing ambiguity in the test
about the state of the replication threads.
Don't fail the test if include/stop_slave_io.inc finds an error in the IO
thread after stop. Such errors can be simply because slave stop happened in
the middle of the IO thread's initial communication with the master.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
This patch provides a new mode FULL_NODUP to binlog_row_image system
variable. With FULL_NODUP mode, all columns are included in before
image, but only updated columns are included in after image for UPDATE.
While all columns are included in the after image for INSERT.
FULL_NODUP is for replacing FULL mode. It includes all data of
the before and after image as FULL mode, but it uses less storage
especially in the case that only a few columns are updated.
Note: It will binlog full before and after image for all modes if the
table has no primary key. FULL_NODUP follows the behavior.
Wait for the disconnect of the other connection to complete, before running
SHOW BINLOG EVENTS. Otherwise the DROP TEMPORARY TABLE that is binlogged
during disconnect may not have appeared yet depending on thread scheduling.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
Fix sporadic test failure in rpl.rpl_ssl1. The test incorrectly did a STOP
SLAVE too early, which could race with the expected 'Access denied' error.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
Fix sporadic test failures in rpl.rpl_set_statement_default_master and
rpl.rpl_slave_load_tmpdir_not_exist. A race between START and STOP
SLAVE could leave an error condition that causes test failure after
MDEV-32168.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
Test rpl.show_status_stop_slave_race-7126 now fails sporadically
because it is expected to sometimes (but not always) leave an error
condition after slave stop. Fix by explicitly allowing the error
condition in this case.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
Fix a start/stop race that causes occasional test failure after more the more
strict error check of MDEV-32168.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
