database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-08-08 11:22:35 +03:00
Code Activity

MDEV-32265: seconds_behind_master is inaccurate for Delayed replication

Browse Source 
If a replica is actively delaying a transaction when restarted (STOP
SLAVE/START SLAVE), when the sql thread is back up,
Seconds_Behind_Master will present as 0 until the configured
MASTER_DELAY has passed. That is, before the restart,
last_master_timestamp is updated to the timestamp of the delayed
event. Then after the restart, the negation of sql_thread_caught_up
is skipped because the timestamp of the event has already been used
for the last_master_timestamp, and their update is grouped together
in the same conditional block.

This patch fixes this by separating the negation of
sql_thread_caught_up out of the timestamp-dependent block, so it is
called any time an idle parallel slave queues an event to a worker.

Note that sql_thread_caught_up is still left in the check for internal
events, as SBM should remain idle in such case to not "magically" begin
incrementing.

Reviewed By:
============
Andrei Elkin <andrei.elkin@mariadb.com>
This commit is contained in:
Brandon Nesterenko
2023-09-27 14:39:03 -06:00
parent 9517755165
commit c5f776e9fa
3 changed files with 34 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
mysql-test/suite/rpl/r/rpl_parallel_sbm.result
View File

@@ -15,9 +15,6 @@ create table t2 (a int);
include/sync_slave_sql_with_master.inc
#
# Pt 1) Ensure SBM is updated immediately upon arrival of the next event
# Lock t1 on slave so the first received transaction does not complete/commit
connection slave;
LOCK TABLES t1 WRITE;
connection master;
# Sleep 2 to allow a buffer between events for SBM check
insert into t1 values (0);
@@ -26,8 +23,16 @@ connection slave;
# Waiting for transaction to arrive on slave and begin SQL Delay..
# Validating SBM is updated on event arrival..
# ..done
# MDEV-32265. At time of STOP SLAVE, if the SQL Thread is currently
# delaying a transaction; then when the reciprocal START SLAVE occurs,
# if the event is still to be delayed, SBM should resume accordingly
include/stop_slave.inc
include/start_slave.inc
connection slave;
UNLOCK TABLES;
# Waiting for replica to resume the delay for the transaction
# Sleeping 1s to increment SBM
# Ensuring Seconds_Behind_Master increases after sleeping..
# ..done
include/sync_with_master_gtid.inc
#
# Pt 2) If the worker threads have not entered an idle state, ensure