mirror of
https://github.com/MariaDB/server.git
synced 2025-08-01 03:47:19 +03:00
Fixes for Bug#12429: Replication tests fail: "Slave_IO_Running" (?) differs related to
MySQL 4.1 and Bug#16920 rpl_deadlock_innodb fails in show slave status (reported for MySQL 5.1) - backport of several fixes done in MySQL 5.0 to 4.1 - fix for new discovered instability (see comment on Bug#12429 + Bug#16920) - reenabling of testcases mysql-test/r/rpl_deadlock.result: Updated results mysql-test/r/rpl_relayrotate.result: Updated results mysql-test/t/disabled.def: Reenabling of tests mysql-test/t/rpl_deadlock.test: - replace sleep with real_sleep (backport fix for Bug#15624 MySQL 5.0) - egalized value for Slave_IO_Running - line 105 (backport fix for Bug#12429 MySQL 5.0) - line 85 (see comment in Bug#12429 + Bug#16920 rpl_deadlock_innodb fails in show slave status) - improve readability of show slave status output (--vertical_results) mysql-test/t/rpl_relayrotate.test: - Replace select ... with select max(a) - add sync_with_master (backport fix done by Kristian in MySQL 5.0 because of timing problems similar to Bug#15624) mysql-test/t/rpl_until.test: Add wait_for_slave_to_stop like in the current MySQL 5.0 version of this test. I assume this makes the test results more predictable.
This commit is contained in:
@ -10,10 +10,7 @@ reset slave;
|
||||
start slave;
|
||||
stop slave;
|
||||
start slave;
|
||||
select master_pos_wait('master-bin.001',3000)>=0;
|
||||
master_pos_wait('master-bin.001',3000)>=0
|
||||
1
|
||||
select * from t1 where a=8000;
|
||||
a
|
||||
select max(a) from t1;
|
||||
max(a)
|
||||
8000
|
||||
drop table t1;
|
||||
|
Reference in New Issue
Block a user