Fix inaccurate MySQL/MariaDB, Redis metrics (#12130)

apache · Apr 22, 2024 · 86155be · 86155be
1 parent e9a2ed2
commit 86155be
Show file tree

Hide file tree

Showing 15 changed files with 294 additions and 185 deletions.
diff --git a/docs/en/changes/changes.md b/docs/en/changes/changes.md
@@ -105,7 +105,8 @@
   - `memory_swap_percentage` -> `memory_virtual_memory_percentage`
 * Fix/Change UI init setting for Windows Swap -> Virtual Memory
 * Fix `Memory Swap Usage`/`Virtual Memory Usage` display with UI init.(Linux/Windows)
-* Fix inaccurate APISIX metrics
+
+* Fix inaccurate APISIX metrics.
 * Fix inaccurate MongoDB Metrics.
 * Support Apache ActiveMQ server monitoring. 
 * Add Service Hierarchy auto matching layer relationships (upper -> lower) as following:
@@ -114,6 +115,7 @@
 * MQE query: make metadata not return `null`.
 * MQE labeled metrics Binary Operation: return empty value if the labels not match rather than report error.
 * Fix inaccurate Hierarchy of RabbitMQ Server monitoring metrics.
+* Fix inaccurate MySQL/MariaDB, Redis, PostgreSQL metrics.
 
 #### UI
 

diff --git a/docs/en/setup/backend/backend-mysql-monitoring.md b/docs/en/setup/backend/backend-mysql-monitoring.md
@@ -17,19 +17,19 @@ SkyWalking leverages prometheus/mysqld_exporter for collecting metrics data. It
 MySQL/MariaDB monitoring provides monitoring of the status and resources of the MySQL/MariaDB server. MySQL/MariaDB cluster is cataloged as a `Layer: MYSQL` `Service` in OAP.
 Each MySQL/MariaDB server is cataloged as an `Instance` in OAP.
 #### Supported Metrics 
-| Monitoring Panel | Unit | Metric Name | Description | Data Source |
-|-----|------|-----|-----|-----|
-| MySQL Uptime |   day   | meter_mysql_uptime | The MySQL startup time | mysqld_exporter|
-| Max Connections |      | meter_mysql_max_connections | The max number of connections. | mysqld_exporter|
-| Innodb Buffer Pool Size |  MB    | meter_mysql_innodb_buffer_pool_size | The buffer pool size in Innodb engine | mysqld_exporter|
-| Thread Cache Size |      | meter_mysql_thread_cache_size | The size of thread cache | mysqld_exporter|
-| Current QPS|      | meter_mysql_qps | Queries Per Second | mysqld_exporter|
-| Current TPS |      | meter_mysql_tps | Transactions Per Second | mysqld_exporter|
+| Monitoring Panel | Unit | Metric Name                                                                                                                                      | Description | Data Source |
+|-----|------|--------------------------------------------------------------------------------------------------------------------------------------------------|-----|-----|
+| MySQL Uptime |   day   | meter_mysql_uptime                                                                                                                               | The MySQL startup time | mysqld_exporter|
+| Max Connections |      | meter_mysql_max_connections                                                                                                                      | The max number of connections. | mysqld_exporter|
+| Innodb Buffer Pool Size |  MB    | meter_mysql_innodb_buffer_pool_size                                                                                                              | The buffer pool size in Innodb engine | mysqld_exporter|
+| Thread Cache Size |      | meter_mysql_thread_cache_size                                                                                                                    | The size of thread cache | mysqld_exporter|
+| Current QPS|      | meter_mysql_qps                                                                                                                                  | Queries Per Second | mysqld_exporter|
+| Current TPS |      | meter_mysql_tps                                                                                                                                  | Transactions Per Second | mysqld_exporter|
 | Commands Rate |     | meter_mysql_commands_insert_rate <br/>meter_mysql_commands_select_rate<br />meter_mysql_commands_delete_rate<br />meter_mysql_commands_update_rate | The rate of total number of insert/select/delete/update executed by the current server | mysqld_exporter|
-| Threads |    | meter_mysql_threads_connected<br />meter_mysql_threads_created<br />meter_mysql_threads_cached<br />meter_mysql_threads_running | The number of currently open connections(threads_connected) <br/> The number of threads created(threads_created) <br/> The number of threads in the thread cache(threads_cached) <br/> The number of threads that are not sleeping(threads_running) | mysqld_exporter|
-| Connects |    | meter_mysql_connects_available<br />meter_mysql_connects_aborted | The number of available connections(connects_available)<br/>The number of MySQL instance connection rejections(connects_aborted)| mysqld_exporter|
-| Connection Errors |      | meter_mysql_connection_errors_internal </br> meter_mysql_connection_errors_max_connections | Errors due to exceeding the max_connections(connection_errors_max_connections) </br>Error caused by internal system(connection_errors_internal) | mysqld_exporter|
-| Slow Queries Rate |      | meter_mysql_slow_queries_rate | The rate of slow queries  | mysqld_exporter|
+| Threads |    | meter_mysql_threads_connected<br />meter_mysql_threads_created<br />meter_mysql_threads_cached<br />meter_mysql_threads_running                  | The number of currently open connections(threads_connected) <br/> The number of threads created(threads_created) <br/> The number of threads in the thread cache(threads_cached) <br/> The number of threads that are not sleeping(threads_running) | mysqld_exporter|
+| Connects |    | meter_mysql_max_connections<br />meter_mysql_status_thread_connected<br />meter_mysql_connects_aborted                                                                              | The number of available connections(connects_available)<br/>The number of MySQL instance connection rejections(connects_aborted)| mysqld_exporter|
+| Connection Errors |      | meter_mysql_connection_errors_internal </br> meter_mysql_connection_errors_max_connections                                                       | Errors due to exceeding the max_connections(connection_errors_max_connections) </br>Error caused by internal system(connection_errors_internal) | mysqld_exporter|
+| Slow Queries Rate |      | meter_mysql_slow_queries_rate                                                                                                                    | The rate of slow queries  | mysqld_exporter|
 
 ### Customizations
 You can customize your own metrics/expression/dashboard panel.

diff --git a/docs/en/setup/backend/backend-redis-monitoring.md b/docs/en/setup/backend/backend-redis-monitoring.md
@@ -17,19 +17,19 @@ SkyWalking leverages redis-exporter for collecting metrics data from Redis. It l
 Redis monitoring provides monitoring of the status and resources of the Redis server. Redis cluster is cataloged as a `Layer: REDIS` `Service` in OAP.
 Each Redis server is cataloged as an `Instance` in OAP.
 #### Supported Metrics
-| Monitoring Panel                  | Unit   | Metric Name                                                                                       | Description                                        | Data Source    |
-|-----------------------------------|--------|---------------------------------------------------------------------------------------------------|----------------------------------------------------|----------------|
-| Uptime                            | day    | meter_redis_uptime                                                                                | The uptime of Redis.                               | redis-exporter |
-| Connected Clients                 |        | meter_redis_connected_clients                                                                     | The number of connected clients.                   | redis-exporter |
-| Blocked Clients                   |        | meter_redis_blocked_clients                                                                       | The number of blocked clients.                     | redis-exporter |
-| Memory Max Bytes                  | MB     | meter_redis_memory_max_bytes                                                                      | The max bytes of memory.                           | redis-exporter |
-| Hits Rate                         | %      | meter_redis_hit_rate                                                                              | Hit rate of redis when used as a cache.            | redis-exporter |
-| Average Time Spend By Command     | second | meter_redis_average_time_spent_by_command                                                         | Average time to execute various types of commands. | redis-exporter |
-| Total Commands Trend              |        | meter_redis_total_commands_rate                                                                   | The Trend of total commands.                       | redis-exporter |
+| Monitoring Panel                  | Unit   | Metric Name                                                                                      | Description                                        | Data Source    |
+|-----------------------------------|--------|--------------------------------------------------------------------------------------------------|----------------------------------------------------|----------------|
+| Uptime                            | day    | meter_redis_uptime                                                                               | The uptime of Redis.                               | redis-exporter |
+| Connected Clients                 |        | meter_redis_connected_clients                                                                    | The number of connected clients.                   | redis-exporter |
+| Blocked Clients                   |        | meter_redis_blocked_clients                                                                      | The number of blocked clients.                     | redis-exporter |
+| Memory Max Bytes                  | MB     | meter_redis_memory_max_bytes                                                                     | The max bytes of memory.                           | redis-exporter |
+| Hits Rate                         | %      | meter_redis_hit_rate                                                                             | Hit rate of redis when used as a cache.            | redis-exporter |
+| Average Time Spend By Command     | second | meter_redis_average_time_spent_by_command                                                        | Average time to execute various types of commands. | redis-exporter |
+| Total Commands Trend              |        | meter_redis_total_commands_rate                                                                  | The Trend of total commands.                       | redis-exporter |
 | DB keys                           |        | meter_redis_evicted_keys_total  </br>   meter_redis_expired_keys_total  </br> meter_redis_db_keys | The number of Expired / Evicted / total keys.      | redis-exporter |
-| Net Input/Output Bytes            | KB     | meter_redis_net_input_bytes  </br>  meter_redis_net_output_bytes                                  | Total bytes of input / output of redis net.        | redis-exporter |
-| Memory Usage                      | %      | meter_redis_memory_usage                                                                          | Percentage of used memory.                         | redis-exporter |
-| Total Time Spend By Command Trend |        | meter_redis_commands_duration_seconds_total_rate                                                  | The trend of total time spend by command           | redis-exporter |
+| Net Input/Output Bytes            | KB     | meter_redis_net_input_bytes  </br>  meter_redis_net_output_bytes                                 | Total bytes of input / output of redis net.        | redis-exporter |
+| Memory Usage                      | %      | meter_redis_memory_used_bytes </br> meter_redis_memory_max_bytes                                 | Percentage of used memory.                         | redis-exporter |
+| Total Time Spend By Command Trend |        | meter_redis_commands_duration    </br>     meter_redis_commands_total                                                           | The trend of total time spend by command           | redis-exporter |
 
 ### Customizations
 You can customize your own metrics/expression/dashboard panel.

diff --git a/oap-server/server-starter/src/main/resources/otel-rules/mysql/mysql-service.yaml b/oap-server/server-starter/src/main/resources/otel-rules/mysql/mysql-service.yaml
@@ -34,39 +34,41 @@ metricPrefix: meter_mysql
 metricsRules:
   # database throughput
   - name: commands_insert_rate
-    exp: mysql_global_status_commands_total.tagEqual('command','insert').rate('PT1M')
+    exp: mysql_global_status_commands_total.tagEqual('command','insert').sum(['service_instance_id','host_name']).rate('PT1M')
   - name: commands_select_rate
-    exp: mysql_global_status_commands_total.tagEqual('command','select').rate('PT1M')
+    exp: mysql_global_status_commands_total.tagEqual('command','select').sum(['service_instance_id','host_name']).rate('PT1M')
   - name: commands_delete_rate
-    exp: mysql_global_status_commands_total.tagEqual('command','delete').rate('PT1M')
+    exp: mysql_global_status_commands_total.tagEqual('command','delete').sum(['service_instance_id','host_name']).rate('PT1M')
   - name: commands_update_rate
-    exp: mysql_global_status_commands_total.tagEqual('command','update').rate('PT1M')
+    exp: mysql_global_status_commands_total.tagEqual('command','update').sum(['service_instance_id','host_name']).rate('PT1M')
   - name: qps
-    exp: mysql_global_status_queries.rate('PT1M')
+    exp: mysql_global_status_queries.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: tps
-    exp: mysql_global_status_commands_total.tagMatch('command','rollback|commit').sum(['host_name']).rate('PT1M')
+    exp: mysql_global_status_commands_total.tagMatch('command','rollback|commit').sum(['host_name', 'service_instance_id']).rate('PT1M')
 
   # connections
   ## threads
   - name: threads_connected
-    exp: mysql_global_status_threads_connected
+    exp: mysql_global_status_threads_connected.sum(['service_instance_id','host_name'])
   - name: threads_created
-    exp: mysql_global_status_threads_created
+    exp: mysql_global_status_threads_created.sum(['service_instance_id','host_name'])
   - name: threads_running
-    exp: mysql_global_status_threads_running
+    exp: mysql_global_status_threads_running.sum(['service_instance_id','host_name'])
   - name: threads_cached
-    exp: mysql_global_status_threads_cached
+    exp: mysql_global_status_threads_cached.sum(['service_instance_id','host_name'])
   ## connect
   - name: connects_aborted
-    exp: mysql_global_status_aborted_connects
-  - name: connects_available
-    exp: mysql_global_variables_max_connections.sum(['host_name']) - mysql_global_status_threads_connected.sum(['host_name'])
+    exp: mysql_global_status_aborted_connects.sum(['service_instance_id','host_name'])
+  - name: max_connections
+    exp: mysql_global_variables_max_connections.sum(['host_name','service_instance_id'])
+  - name: status_thread_connected
+    exp: mysql_global_status_threads_connected.sum(['host_name','service_instance_id'])
   - name: connection_errors_max_connections
-    exp: mysql_global_status_connection_errors_total.tagEqual('error','max_connection')
+    exp: mysql_global_status_connection_errors_total.tagEqual('error','max_connection').sum(['service_instance_id','host_name'])
   - name: connection_errors_internal
-    exp: mysql_global_status_connection_errors_total.tagEqual('error','internal')
+    exp: mysql_global_status_connection_errors_total.tagEqual('error','internal').sum(['service_instance_id','host_name'])
 
   # slow queries
   - name: slow_queries_rate
-    exp: mysql_global_status_slow_queries.rate('PT1M')
+    exp: mysql_global_status_slow_queries.sum(['service_instance_id','host_name']).rate('PT1M')
 
diff --git a/oap-server/server-starter/src/main/resources/otel-rules/postgresql/postgresql-instance.yaml b/oap-server/server-starter/src/main/resources/otel-rules/postgresql/postgresql-instance.yaml
@@ -104,12 +104,12 @@ metricsRules:
 
   ## buffers
   - name: instance_buffers_checkpoint
-    exp: pg_stat_bgwriter_buffers_checkpoint.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_checkpoint_total.rate('PT1M')
   - name: instance_buffers_clean
-    exp: pg_stat_bgwriter_buffers_clean.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_clean_total.rate('PT1M')
   - name: instance_buffers_backend_fsync
-    exp: pg_stat_bgwriter_buffers_backend_fsync.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_backend_fsync_total.rate('PT1M')
   - name: instance_buffers_alloc
-    exp: pg_stat_bgwriter_buffers_alloc.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_alloc_total.rate('PT1M')
   - name: instance_buffers_backend
-    exp: pg_stat_bgwriter_buffers_backend.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_backend_total.rate('PT1M')
diff --git a/oap-server/server-starter/src/main/resources/otel-rules/postgresql/postgresql-service.yaml b/oap-server/server-starter/src/main/resources/otel-rules/postgresql/postgresql-service.yaml
@@ -46,7 +46,7 @@ metricsRules:
     exp: pg_stat_database_tup_returned.sum(['service_instance_id','host_name']).rate('PT1M')
   ## locks
   - name: locks_count
-    exp: pg_locks_count.tag({tags -> tags.mode = tags.datname + ":" + tags.mode}).sum(['service_instance_id','host_name'])
+    exp: pg_locks_count.sum(['service_instance_id','host_name'])
 
   ## sessions
   - name: active_sessions
@@ -68,13 +68,13 @@ metricsRules:
 
   ## checkpoint
   - name: checkpoint_write_time_rate
-    exp: pg_stat_bgwriter_checkpoint_write_time_total.rate('PT1M')
+    exp: pg_stat_bgwriter_checkpoint_write_time_total.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: checkpoint_sync_time_rate
-    exp: pg_stat_bgwriter_checkpoint_sync_time_total.rate('PT1M')
+    exp: pg_stat_bgwriter_checkpoint_sync_time_total.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: checkpoint_req_rate
-    exp: pg_stat_bgwriter_checkpoints_req_total.rate('PT1M')
+    exp: pg_stat_bgwriter_checkpoints_req_total.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: checkpoints_timed_rate
-    exp: pg_stat_bgwriter_checkpoints_timed_total.rate('PT1M')
+    exp: pg_stat_bgwriter_checkpoints_timed_total.rate('PT1M').sum(['service_instance_id','host_name'])
 
   ## conflicts and deadlocks
   - name: conflicts_rate
@@ -84,12 +84,12 @@ metricsRules:
 
   ## buffers
   - name: buffers_checkpoint
-    exp: pg_stat_bgwriter_buffers_checkpoint.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_checkpoint_total.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: buffers_clean
-    exp: pg_stat_bgwriter_buffers_clean.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_clean_total.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: buffers_backend_fsync
-    exp: pg_stat_bgwriter_buffers_backend_fsync.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_backend_fsync_total.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: buffers_alloc
-    exp: pg_stat_bgwriter_buffers_alloc.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_alloc_total.rate('PT1M').sum(['service_instance_id','host_name'])
   - name: buffers_backend
-    exp: pg_stat_bgwriter_buffers_backend.rate('PT1M')
+    exp: pg_stat_bgwriter_buffers_backend_total.rate('PT1M').sum(['service_instance_id','host_name'])