Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Fixed
-
10.5, 10.6, 10.7
Description
The test innodb.page_cleaner would occasionally time out like this:
@@ -2,6 +2,11 @@
|
SET @save_pct_lwm= @@GLOBAL.innodb_max_dirty_pages_pct_lwm;
|
SET GLOBAL innodb_max_dirty_pages_pct_lwm=0.0;
|
SET GLOBAL innodb_max_dirty_pages_pct=0.0;
|
+Timeout in wait_condition.inc for SELECT variable_value = 0
|
+FROM information_schema.global_status
|
+WHERE variable_name = 'INNODB_BUFFER_POOL_PAGES_DIRTY'
|
+Id User Host db Command Time State Info Progress
|
+28 root localhost test Query 0 starting show full processlist 0.000
|
SET GLOBAL innodb_max_dirty_pages_pct=90.0;
|
CREATE TABLE t ENGINE=InnoDB SELECT * FROM seq_1_to_10000;
|
SELECT variable_value>0 FROM information_schema.global_status |
The reason is that we fail to wake up the buf_pool_page_cleaner from pthread_cond_wait() when innodb_max_dirty_pages_pct is assigned. Invoking pthread_cond_signal() on the condition variable is not enough, not even while holding buf_pool.flush_list_mutex.
Invoking buf_pool.page_cleaner_wakeup() seems to do the trick. With that fix, the test will execute several hundred times without failures.
Attachments
Issue Links
- is caused by
-
-
- Closed
-