我的书签
添加书签
移除书签Metrics Schema
METRICS_SCHEMA 是基于 Prometheus 中 TiDB 监控指标的一组视图。每个表的 PromQL（Prometheus 查询语言）的源均可在 INFORMATION_SCHEMA.METRICS_TABLES 表中找到。
1. use metrics_schema;
2. SELECT * FROM uptime;
3. SELECT * FROM information_schema.metrics_tables WHERE table_name='uptime'\G
1. +----------------------------+-----------------+------------+--------------------+
2. | time | instance | job | value |
3. +----------------------------+-----------------+------------+--------------------+
4. | 2020-07-06 15:26:26.203000 | 127.0.0.1:10080 | tidb | 123.60300016403198 |
5. | 2020-07-06 15:27:26.203000 | 127.0.0.1:10080 | tidb | 183.60300016403198 |
6. | 2020-07-06 15:26:26.203000 | 127.0.0.1:20180 | tikv | 123.60300016403198 |
7. | 2020-07-06 15:27:26.203000 | 127.0.0.1:20180 | tikv | 183.60300016403198 |
8. | 2020-07-06 15:26:26.203000 | 127.0.0.1:2379 | pd | 123.60300016403198 |
9. | 2020-07-06 15:27:26.203000 | 127.0.0.1:2379 | pd | 183.60300016403198 |
10. | 2020-07-06 15:26:26.203000 | 127.0.0.1:9090 | prometheus | 123.72300004959106 |
11. | 2020-07-06 15:27:26.203000 | 127.0.0.1:9090 | prometheus | 183.72300004959106 |
12. +----------------------------+-----------------+------------+--------------------+
13. 8 rows in set (0.00 sec)
14. *************************** 1. row ***************************
15. TABLE_NAME: uptime
16. PROMQL: (time() - process_start_time_seconds{$LABEL_CONDITIONS})
17. LABELS: instance,job
18. QUANTILE: 0
19. COMMENT: TiDB uptime since last restart(second)
20. 1 row in set (0.00 sec)
1. show tables;
1. +---------------------------------------------------+
2. | Tables_in_metrics_schema |
3. +---------------------------------------------------+
4. | abnormal_stores |
5. | etcd_disk_wal_fsync_rate |
6. | etcd_wal_fsync_duration |
7. | etcd_wal_fsync_total_count |
8. | etcd_wal_fsync_total_time |
9. | go_gc_count |
10. | go_gc_cpu_usage |
11. | go_gc_duration |
12. | go_heap_mem_usage |
13. | go_threads |
14. | goroutines_count |
15. | node_cpu_usage |
16. | node_disk_available_size |
17. | node_disk_io_util |
18. | node_disk_iops |
19. | node_disk_read_latency |
20. | node_disk_size |
21. ..
22. | tikv_storage_async_request_total_time |
23. | tikv_storage_async_requests |
24. | tikv_storage_async_requests_total_count |
25. | tikv_storage_command_ops |
26. | tikv_store_size |
27. | tikv_thread_cpu |
28. | tikv_thread_nonvoluntary_context_switches |
29. | tikv_thread_voluntary_context_switches |
30. | tikv_threads_io |
31. | tikv_threads_state |
32. | tikv_total_keys |
33. | tikv_wal_sync_duration |
34. | tikv_wal_sync_max_duration |
35. | tikv_worker_handled_tasks |
36. | tikv_worker_handled_tasks_total_num |
37. | tikv_worker_pending_tasks |
38. | tikv_worker_pending_tasks_total_num |
39. | tikv_write_stall_avg_duration |
40. | tikv_write_stall_max_duration |
41. | tikv_write_stall_reason |
42. | up |
43. | uptime |
44. +---------------------------------------------------+
45. 626 rows in set (0.00 sec)
METRICS_SCHEMA 是监控相关的 summary 表的数据源，例如 metrics_summary、metrics_summary_by_label 和 inspection_summary。
更多例子
下面以 metrics_schema 中的 tidb_query_duration 监控表为例，介绍监控表相关的使用和原理，其他的监控表原理均类似。
查询 information_schema.metrics_tables 中关于 tidb_query_duration 表相关的信息如下：
1. select * from information_schema.metrics_tables where table_name='tidb_query_duration';
1. +---------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------+-------------------+----------+----------------------------------------------+
2. | TABLE_NAME | PROMQL | LABELS | QUANTILE | COMMENT |
3. +---------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------+-------------------+----------+----------------------------------------------+
4. | tidb_query_duration | histogram_quantile($QUANTILE, sum(rate(tidb_server_handle_query_duration_seconds_bucket{$LABEL_CONDITIONS}[$RANGE_DURATION])) by (le,sql_type,instance)) | instance,sql_type | 0.9 | The quantile of TiDB query durations(second) |
5. +---------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------+-------------------+----------+----------------------------------------------+
TABLE_NAME：对应于metrics_schema中的表名，这里表名是tidb_query_duration。PROMQL：因为监控表的原理是将 SQL 映射成PromQL后向 Prometheus 请求数据，并将 Prometheus 返回的结果转换成 SQL 查询结果。该字段是PromQL的表达式模板，查询监控表数据时使用查询条件改写模板中的变量，生成最终的查询表达式。LABELS：监控项定义的 label，tidb_query_duration有两个 label，分别是instance和sql_type。QUANTILE：百分位。直方图类型的监控数据会指定一个默认百分位。如果值为0，表示该监控表对应的监控不是直方图。tidb_query_duration默认查询 0.9，也就是 P90 的监控值。COMMENT：对这个监控表的解释。可以看出tidb_query_duration表是用来查询 TiDB query 执行的百分位时间，如 P999/P99/P90 的查询耗时，单位是秒。
再来看 tidb_query_duration 的表结构：
1. show create table metrics_schema.tidb_query_duration;
1. +---------------------+--------------------------------------------------------------------------------------------------------------------+
2. | Table | Create Table |
3. +---------------------+--------------------------------------------------------------------------------------------------------------------+
4. | tidb_query_duration | CREATE TABLE `tidb_query_duration` ( |
5. | | `time` datetime unsigned DEFAULT CURRENT_TIMESTAMP, |
6. | | `instance` varchar(512) DEFAULT NULL, |
7. | | `sql_type` varchar(512) DEFAULT NULL, |
8. | | `quantile` double unsigned DEFAULT '0.9', |
9. | | `value` double unsigned DEFAULT NULL |
10. | | ) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_bin COMMENT='The quantile of TiDB query durations(second)' |
11. +---------------------+--------------------------------------------------------------------------------------------------------------------+
time：监控项的时间。instance和sql_type：是tidb_query_duration这个监控项的 label。instance表示监控的地址，sql_type表示执行 SQL 的类似。quantile，百分位，直方图类型的监控都会有该列，表示查询的百分位时间，如quantile=0.9就是查询 P90 的时间。value：监控项的值。
下面是查询时间 [2020-03-25 23:40:00, 2020-03-25 23:42:00] 范围内的 P99 的 TiDB Query 耗时：
1. select * from metrics_schema.tidb_query_duration where value is not null and time>='2020-03-25 23:40:00' and time <= '2020-03-25 23:42:00' and quantile=0.99;
1. +---------------------+-------------------+----------+----------+----------------+
2. | time | instance | sql_type | quantile | value |
3. +---------------------+-------------------+----------+----------+----------------+
4. | 2020-03-25 23:40:00 | 172.16.5.40:10089 | Insert | 0.99 | 0.509929485256 |
5. | 2020-03-25 23:41:00 | 172.16.5.40:10089 | Insert | 0.99 | 0.494690793986 |
6. | 2020-03-25 23:42:00 | 172.16.5.40:10089 | Insert | 0.99 | 0.493460506934 |
7. | 2020-03-25 23:40:00 | 172.16.5.40:10089 | Select | 0.99 | 0.152058493415 |
8. | 2020-03-25 23:41:00 | 172.16.5.40:10089 | Select | 0.99 | 0.152193879678 |
9. | 2020-03-25 23:42:00 | 172.16.5.40:10089 | Select | 0.99 | 0.140498483232 |
10. | 2020-03-25 23:40:00 | 172.16.5.40:10089 | internal | 0.99 | 0.47104 |
11. | 2020-03-25 23:41:00 | 172.16.5.40:10089 | internal | 0.99 | 0.11776 |
12. | 2020-03-25 23:42:00 | 172.16.5.40:10089 | internal | 0.99 | 0.11776 |
13. +---------------------+-------------------+----------+----------+----------------+
以上查询结果的第一行意思是，在 2020-03-25 23:40:00 时，在 TiDB 实例 172.16.5.40:10089 上，Insert 类型的语句的 P99 执行时间是 0.509929485256 秒。其他各行的含义类似，sql_type 列的其他值含义如下：
Select：表示执行的select类型的语句。internal：表示 TiDB 的内部 SQL 语句，一般是统计信息更新，获取全局变量相关的内部语句。
进一步再查看上面语句的执行计划如下：
1. desc select * from metrics_schema.tidb_query_duration where value is not null and time>='2020-03-25 23:40:00' and time <= '2020-03-25 23:42:00' and quantile=0.99;
1. +------------------+----------+------+---------------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
2. | id | estRows | task | access object | operator info |
3. +------------------+----------+------+---------------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
4. | Selection_5 | 8000.00 | root | | not(isnull(Column#5)) |
5. | └─MemTableScan_6 | 10000.00 | root | table:tidb_query_duration | PromQL:histogram_quantile(0.99, sum(rate(tidb_server_handle_query_duration_seconds_bucket{}[60s])) by (le,sql_type,instance)), start_time:2020-03-25 23:40:00, end_time:2020-03-25 23:42:00, step:1m0s |
6. +------------------+----------+------+---------------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
可以发现执行计划中有一个 PromQL，以及查询监控的 start_time 和 end_time，还有 step 值，在实际执行时，TiDB 会调用 Prometheus 的 query_range HTTP API 接口来查询监控数据。
从以上结果可知，在 [2020-03-25 23:40:00, 2020-03-25 23:42:00] 时间范围内，每个 label 只有三个时间的值，间隔时间是 1 分钟，即执行计划中的 step 值。该间隔时间由以下两个 session 变量决定：
tidb_metric_query_step：查询的分辨率步长。从 Prometheus 的query_range接口查询数据时需要指定start_time，end_time和step，其中step会使用该变量的值。tidb_metric_query_range_duration：查询监控时，会将PROMQL中的$RANGE_DURATION替换成该变量的值，默认值是 60 秒。
如果想要查看不同时间粒度的监控项的值，用户可以修改上面两个 session 变量后查询监控表，示例如下：
首先修改两个 session 变量的值，将时间粒度设置为 30 秒。
注意
Prometheus 支持查询的最小粒度为 30 秒。
1. set @@tidb_metric_query_step=30;
2. set @@tidb_metric_query_range_duration=30;
再查询 tidb_query_duration 监控如下，可以发现在三分钟时间范围内，每个 label 有六个时间的值，每个值时间间隔是 30 秒。
1. select * from metrics_schema.tidb_query_duration where value is not null and time>='2020-03-25 23:40:00' and time <= '2020-03-25 23:42:00' and quantile=0.99;
1. +---------------------+-------------------+----------+----------+-----------------+
2. | time | instance | sql_type | quantile | value |
3. +---------------------+-------------------+----------+----------+-----------------+
4. | 2020-03-25 23:40:00 | 172.16.5.40:10089 | Insert | 0.99 | 0.483285651924 |
5. | 2020-03-25 23:40:30 | 172.16.5.40:10089 | Insert | 0.99 | 0.484151462113 |
6. | 2020-03-25 23:41:00 | 172.16.5.40:10089 | Insert | 0.99 | 0.504576 |
7. | 2020-03-25 23:41:30 | 172.16.5.40:10089 | Insert | 0.99 | 0.493577384561 |
8. | 2020-03-25 23:42:00 | 172.16.5.40:10089 | Insert | 0.99 | 0.49482474311 |
9. | 2020-03-25 23:40:00 | 172.16.5.40:10089 | Select | 0.99 | 0.189253402185 |
10. | 2020-03-25 23:40:30 | 172.16.5.40:10089 | Select | 0.99 | 0.184224951851 |
11. | 2020-03-25 23:41:00 | 172.16.5.40:10089 | Select | 0.99 | 0.151673410553 |
12. | 2020-03-25 23:41:30 | 172.16.5.40:10089 | Select | 0.99 | 0.127953838989 |
13. | 2020-03-25 23:42:00 | 172.16.5.40:10089 | Select | 0.99 | 0.127455434547 |
14. | 2020-03-25 23:40:00 | 172.16.5.40:10089 | internal | 0.99 | 0.0624 |
15. | 2020-03-25 23:40:30 | 172.16.5.40:10089 | internal | 0.99 | 0.12416 |
16. | 2020-03-25 23:41:00 | 172.16.5.40:10089 | internal | 0.99 | 0.0304 |
17. | 2020-03-25 23:41:30 | 172.16.5.40:10089 | internal | 0.99 | 0.06272 |
18. | 2020-03-25 23:42:00 | 172.16.5.40:10089 | internal | 0.99 | 0.0629333333333 |
19. +---------------------+-------------------+----------+----------+-----------------+
最后查看执行计划，也会发现执行计划中的 PromQL 以及 step 的值都已经变成了 30 秒。
1. desc select * from metrics_schema.tidb_query_duration where value is not null and time>='2020-03-25 23:40:00' and time <= '2020-03-25 23:42:00' and quantile=0.99;
1. +------------------+----------+------+---------------------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
2. | id | estRows | task | access object | operator info |
3. +------------------+----------+------+---------------------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
4. | Selection_5 | 8000.00 | root | | not(isnull(Column#5)) |
5. | └─MemTableScan_6 | 10000.00 | root | table:tidb_query_duration | PromQL:histogram_quantile(0.99, sum(rate(tidb_server_handle_query_duration_seconds_bucket{}[30s])) by (le,sql_type,instance)), start_time:2020-03-25 23:40:00, end_time:2020-03-25 23:42:00, step:30s |
6. +------------------+----------+------+---------------------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
