1

我认为我已将 long_query_time 参数设置为 2（这意味着两秒）。但是我可以在我的慢速日志文件中看到许多查询，这些查询低于 2 秒。实际上它必须填充超过2秒的文件才知道吗？

以下是mysql慢日志条目

Time                 Id Command    Argument
# User@Host: rdsadmin[rdsadmin] @ localhost [127.0.0.1]
# Query_time: 0.013783  Lock_time: 0.000000 Rows_sent: 0  Rows_examined: 0
use mysql;
SET timestamp=1445342700;
flush logs;
# Time: 151020 12:05:15
# User@Host: rdsadmin[rdsadmin] @ localhost [127.0.0.1]
# Query_time: 0.000313  Lock_time: 0.000000 Rows_sent: 1  Rows_examined: 0
SET timestamp=1445342715;
SELECT 1;
# User@Host: rdsadmin[rdsadmin] @ localhost [127.0.0.1]
# Query_time: 0.000129  Lock_time: 0.000000 Rows_sent: 1  Rows_examined: 0
SET timestamp=1445342715;
SELECT 1;
# User@Host: rdsadmin[rdsadmin] @ localhost [127.0.0.1]
# Query_time: 0.000037  Lock_time: 0.000000 Rows_sent: 1  Rows_examined: 0
SET timestamp=1445342715;
SELECT 1;
# User@Host: rdsadmin[rdsadmin] @ localhost [127.0.0.1]
# Query_time: 0.000211  Lock_time: 0.000090 Rows_sent: 1  Rows_examined: 1
SET timestamp=1445342715;
SELECT count(*) from information_schema.TABLES WHERE TABLE_SCHEMA = 'mysql' AND TABLE_NAME = 'rds_heartbeat2';
# User@Host: rdsadmin[rdsadmin] @ localhost [127.0.0.1]
# Query_time: 0.000037  Lock_time: 0.000000 Rows_sent: 1  Rows_examined: 0
SET timestamp=1445342715;
SELECT 1;
# User@Host: rdsadmin[rdsadmin] @ localhost [127.0.0.1]
# Query_time: 0.000124  Lock_time: 0.000030 Rows_sent: 1  Rows_examined: 1
SET timestamp=1445342715;
SELECT value FROM mysql.rds_heartbeat2;
# Time: 151020 12:05:24
# User@Host: root[root] @ ip-127.0.0.1]ec2.internal [127.0.0.1
# Query_time: 0.000296  Lock_time: 0.000079 Rows_sent: 0  Rows_examined: 0
use mobilepush;
SET timestamp=1445342724;
SELECT * FROM mobile_notifications WHERE ACCOUNT_ID = 1010002825 AND NOTIFICATION_TYPE = 1 AND USER_ID != '';
# Time: 151020 12:05:25
# User@Host: root[root] @ ip-127.0.0.1].ec2.internal [127.0.0.1]
# Query_time: 0.002527  Lock_time: 0.000111 Rows_sent: 0  Rows_examined: 0
use reports_db;
SET timestamp=1445342725;
SHOW TABLES LIKE 'reports_partition_mapping';
# User@Host: root[root] @ ip-127.0.0.1.ec2.internal [127.0.0.1]
4

1 回答 1

3

可能你已经log_queries_not_using_indexes打开了。我发现这几乎是无用的——如果缺少索引会导致足够的麻烦，它会显示为大于long_query_time.

于 2015-11-03T03:05:53.967 回答