Details
-
Task
-
Status: Open (View Workflow)
-
Minor
-
Resolution: Unresolved
-
None
-
None
-
None
Description
Percona Server 5.5 provides more statistics in the slow query log than 5.1: there are statistics for
- Tmp_table_sizes
- InnoDB reads
- Profiling.
We should port their new patch.
Example of 5.5's statistics:
|
# Time: 120706 13:45:56
|
# User@Host: root[root] @ localhost []
|
# Thread_id: 1 Schema: test Last_errno: 0 Killed: 0
|
# Query_time: 0.154889 Lock_time: 0.000182 Rows_sent: 999 Rows_examined: 2998 Rows_affected: 0 Rows_read: 999
|
# Bytes_sent: 8049 Tmp_tables: 1 Tmp_disk_tables: 0 Tmp_table_sizes: 126992
|
# Profile_starting: 0.000060 Profile_starting_cpu: 0.000123 .... <lots of Profile_xxx omitted >
|
|
|
|
|
# Profile_total: 0.154894 Profile_total_cpu: 0.320510
|
# InnoDB_trx_id: 222510A
|
# QC_Hit: No Full_scan: Yes Full_join: Yes Tmp_table: Yes Tmp_table_on_disk: No
|
# Filesort: Yes Filesort_on_disk: No Merge_passes: 0
|
# InnoDB_IO_r_ops: 0 InnoDB_IO_r_bytes: 0 InnoDB_IO_r_wait: 0.000000
|
# InnoDB_rec_lock_wait: 0.000000 InnoDB_queue_wait: 0.000000
|
# InnoDB_pages_distinct: 23
|
SET timestamp=1341567956;
|
select max(B.a) from one_ki A , one_ki B where A.a < B.a group by A.a;
|