refactor: add SlowQueryRecorder to record slow query in system table and refactor slow query options (#6008)

* refactor: add common-slow-query crate

* refactor: refine the naming

* chore: fix clippy

* chore: fix typo

* chore: sperate SlowQueryOptions From Logging

* chore: fix clippy

* chore: fix ci

* chore: refine the code

* chore: update config example

* refactor: use drop() to end the slow query timer

* refactor: move common-slow-query to frontend crate

* chore: polish some code

* refactor: code review

* refactor: add promql_range/promql_step/promql_start/promql_end fields in slow_queries

* refactor: add build_slow_query_logger()

* refactor: turn on slow query on frontend by default

config/config.md

@@ -188,10 +188,11 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
+| `slow_query` | -- | -- | The slow query log options. |
+| `slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
+| `slow_query.record_type` | String | Unset | The record type of slow queries. It can be `system_table` or `log`. |
+| `slow_query.threshold` | String | Unset | The threshold of slow query. |
+| `slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
 | `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
@@ -288,10 +289,12 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
+| `slow_query` | -- | -- | The slow query log options. |
+| `slow_query.enable` | Bool | `true` | Whether to enable slow query log. |
+| `slow_query.record_type` | String | Unset | The record type of slow queries. It can be `system_table` or `log`.<br/>If `system_table` is selected, the slow queries will be recorded in a system table `greptime_private.slow_queries`.<br/>If `log` is selected, the slow queries will be logged in a log file `greptimedb-slow-queries.*`. |
+| `slow_query.threshold` | String | Unset | The threshold of slow query. It can be human readable time string, for example: `10s`, `100ms`, `1s`. |
+| `slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. For example, `0.1` means 10% of the slow queries will be logged and `1.0` means all slow queries will be logged. |
+| `slow_query.ttl` | String | Unset | The TTL of the `slow_queries` system table. Default is `30d` when `record_type` is `system_table`. |
 | `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
@@ -362,10 +365,6 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
 | `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
@@ -529,10 +528,6 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
 | `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
@@ -585,9 +580,5 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
 | `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
 | `tracing.tokio_console_addr` | String | Unset | The tokio console address. |

config/datanode.example.toml

@@ -632,19 +632,6 @@ max_log_files = 720
 [logging.tracing_sample_ratio]
 default_ratio = 1.0
 
-## The slow query log options.
-[logging.slow_query]
-## Whether to enable slow query log.
-enable = false
-
-## The threshold of slow query.
-## @toml2docs:none-default
-threshold = "10s"
-
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
-## @toml2docs:none-default
-sample_ratio = 1.0
-
 ## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
 [export_metrics]

config/flownode.example.toml

@@ -100,19 +100,6 @@ max_log_files = 720
 [logging.tracing_sample_ratio]
 default_ratio = 1.0
 
-## The slow query log options.
-[logging.slow_query]
-## Whether to enable slow query log.
-enable = false
-
-## The threshold of slow query.
-## @toml2docs:none-default
-threshold = "10s"
-
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
-## @toml2docs:none-default
-sample_ratio = 1.0
-
 ## The tracing options. Only effect when compiled with `tokio-console` feature.
 #+ [tracing]
 ## The tokio console address.

config/frontend.example.toml

@@ -223,18 +223,28 @@ max_log_files = 720
 default_ratio = 1.0
 
 ## The slow query log options.
-[logging.slow_query]
+[slow_query]
 ## Whether to enable slow query log.
-enable = false
+enable = true
 
-## The threshold of slow query.
+## The record type of slow queries. It can be `system_table` or `log`.
+## If `system_table` is selected, the slow queries will be recorded in a system table `greptime_private.slow_queries`.
+## If `log` is selected, the slow queries will be logged in a log file `greptimedb-slow-queries.*`.
 ## @toml2docs:none-default
-threshold = "10s"
+record_type = "system_table"
 
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
+## The threshold of slow query. It can be human readable time string, for example: `10s`, `100ms`, `1s`.
+## @toml2docs:none-default
+threshold = "5s"
+
+## The sampling ratio of slow query log. The value should be in the range of (0, 1]. For example, `0.1` means 10% of the slow queries will be logged and `1.0` means all slow queries will be logged.
 ## @toml2docs:none-default
 sample_ratio = 1.0
 
+## The TTL of the `slow_queries` system table. Default is `30d` when `record_type` is `system_table`.
+## @toml2docs:none-default
+ttl = "30d"
+
 ## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
 [export_metrics]

config/metasrv.example.toml

@@ -218,19 +218,6 @@ max_log_files = 720
 [logging.tracing_sample_ratio]
 default_ratio = 1.0
 
-## The slow query log options.
-[logging.slow_query]
-## Whether to enable slow query log.
-enable = false
-
-## The threshold of slow query.
-## @toml2docs:none-default
-threshold = "10s"
-
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
-## @toml2docs:none-default
-sample_ratio = 1.0
-
 ## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
 [export_metrics]

config/standalone.example.toml

@@ -724,17 +724,21 @@ max_log_files = 720
 default_ratio = 1.0
 
 ## The slow query log options.
-[logging.slow_query]
+[slow_query]
 ## Whether to enable slow query log.
-enable = false
+#+ enable = false
+
+## The record type of slow queries. It can be `system_table` or `log`.
+## @toml2docs:none-default
+#+ record_type = "system_table"
 
 ## The threshold of slow query.
 ## @toml2docs:none-default
-threshold = "10s"
+#+ threshold = "10s"
 
 ## The sampling ratio of slow query log. The value should be in the range of (0, 1].
 ## @toml2docs:none-default
-sample_ratio = 1.0
+#+ sample_ratio = 1.0
 
 ## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.