database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-07-29 05:21:33 +03:00
Code Activity

Fix for Bug#47745 innodb.innodb-timeout fails sporadically

Browse Source 
- Second scenario checked:
  Ensure via wait routines that the commit comes after the
  processing of the statement which should get finally
  the ER_LOCK_WAIT_TIMEOUT
  --> This should prevent the current bug.
- First scenario checked:
  Ensure via wait routines that the statement is already waiting
  for getting the lock before the commit is given.
  --> No effect on the current bug, but ensure that the right
      scenario is reached.
- Take care that disconnects are finished before the test ends.
  --> Reduce the potential to harm succeeding tests.
- "Mangle" the printout of the current default innodb_lock_wait_timeout value
  --> No need to adjust the test in case the default gets changed in future.
This commit is contained in:
Matthias Leich
2011-01-07 14:16:28 +01:00
parent a79e8fa2d0
commit d09ed4e08c
2 changed files with 68 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
mysql-test/suite/innodb/r/innodb-timeout.result
View File

@ -13,13 +13,14 @@ set global innodb_lock_wait_timeout=347;
select @@innodb_lock_wait_timeout;
@@innodb_lock_wait_timeout
42
set innodb_lock_wait_timeout=1;
set innodb_lock_wait_timeout=10;
select @@innodb_lock_wait_timeout;
@@innodb_lock_wait_timeout
1
10
select @@innodb_lock_wait_timeout;
@@innodb_lock_wait_timeout
347
SET @connection_b_id = <connection_b_id>;
create table t1(a int primary key)engine=innodb;
begin;
insert into t1 values(1),(2),(3);
@ -31,8 +32,9 @@ a
3
begin;
insert into t1 values(4);
set innodb_lock_wait_timeout=3;
select * from t1 for update;
commit;
ERROR HY000: Lock wait timeout exceeded; try restarting transaction
drop table t1;
set global innodb_lock_wait_timeout=50;
set global innodb_lock_wait_timeout=<initial_timeout>;