Basic MRS Monitoring Metrics
Function Description
You can use Cloud Eye to monitor MRS cluster metrics in real time and quickly handle exceptions. This section describes the namespaces, metric list, and dimensions of cluster and node metrics reported by MRS to Cloud Eye. You can use the Cloud Eye console or APIs to retrieve the monitoring metrics and alarms generated for MRS.
Cloud Eye can monitor dimensions nested to a maximum depth of four levels (levels 0 to 3). Level 3 is the deepest level. For example, if the monitored dimension of a metric is cluster_id,node_id, cluster_id indicates level 0 and node_id indicates level 1.
Namespaces
SYS.MRS
Basic Monitoring Metrics
The components included in MRS clusters and supported monitoring metrics may vary depending on the version. You can also view the metrics of each cloud service in the Cloud Eye Metrics and Events Center.
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| ck_concurrency_query | ClickHouseServer concurrency | Number of tasks concurrently executed on a node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ck_disks_local | ClickHouse local disk usage | ClickHouse local disk usage | ≥ 0 | Byte/s | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ck_disks_obs | ClickHouse OBS usage | ClickHouse OBS usage | ≥ 0 | Byte/s | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ck_system_detached_parts | Size of data separated from a ClickHouse MergeTree table | Size of data separated from a MergeTree table | ≥ 0 | GiB/s | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ck_system_detached_parts_allhost | Size of data separated from a MergeTree table in a ClickHouse cluster | Size of data separated from a MergeTree table in a cluster | ≥ 0 | GiB/s | 1024 | cluster_id,service_name | 5 minutes |
| ck_system_metrics_mysqlConnection | Number of ClickHouse MySQL connections | Number of MySQL connections | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ck_system_metrics_mysqlConnection_allhost | Number of MySQL connections in a ClickHouse cluster | Number of MySQL connections in a cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| ck_system_replication_queue_get_part | Number of GET_PART tasks in a ClickHouse replication queue | Number of GET_PART tasks in a replication queue | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ck_system_replication_queue_get_part_allhost | Number of GET_PART tasks in a replication queue of a ClickHouse cluster | Number of GET_PART tasks in a replication queue of a cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| ck_system_replication_queue_merge_parts | Number of MERGE_PARTS tasks in a replication queue of ClickHouseServer | Number of MERGE_PARTS tasks in a replication queue | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ck_system_replication_queue_merge_parts_allhost | Number of MERGE_PARTS tasks in a replication queue of a ClickHouse cluster | Number of MERGE_PARTS tasks in a replication queue of a cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| ck_system_replication_queue_mutate_part | Number of MUTATE_PART tasks in a replication queue of ClickHouseServer | Number of MUTATE_PART tasks in a replication queue | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ck_system_replication_queue_mutate_part_allhost | Number of MUTATE_PART tasks in a replication queue of a ClickHouse cluster | Number of MUTATE_PART tasks in a replication queue of a cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| doris_fe_cpu_usage | Doris FE CPU usage | Doris FE CPU usage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_mem | Doris FE memory usage | Memory used by Doris FE | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_image_write_failed | Doris FE failures to generate metadata image files | The count of failures for the Doris FE to generate metadata image files. This operation should not fail. If it fails, manual intervention is required. No monitoring data is reported if no failure occurs. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_max_journal_id | Maximum metadata log ID of the Doris FE node | Maximum metadata log ID of the Doris FE node. For the master FE node, this value indicates the maximum ID of the metadata log that is being written. For other FE nodes, the value indicates the maximum ID of the metadata log that is being replayed. This metric is used to monitor ID gap among multiple FE nodes. A significant gap indicates a metadata synchronization error. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_qps | QPS (including query requests only) of the Doris FE node | QPS of the Doris FE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_report_queue_size | Queue length of Doris BE periodic report tasks on the FE node | Queue length of Doris BE periodic report tasks on the FE node. This value indicates the number of report tasks waiting on the master FE node. A large value indicates a poor FE processing capability. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_rps | RPS (including query and other statements) of the Doris FE node | RPS (including query and other statements) of the Doris FE node, used together with QPS to monitor the number of requests processed by the cluster. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_scheduled_tablet_num_allhost | Number of tablets that are being scheduled by all Doris FE nodes | Number of tablets that are being scheduled by all Doris FE nodes. The value indicates the number of tablets that are being migrated in the current cluster. If there always are such tablets, the cluster is unstable. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| doris_fe_tablet_num | Total number of tablets on each Doris BE node | Total number of tablets on each Doris BE node. This parameter can help you check whether tablets are evenly distributed and whether the absolute value is proper. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_jvm_thread | Number of JVM threads on the Doris FE | Number of JVM threads on the Doris FE node. This parameter can help you check whether the JVM thread count is proper. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cpu_usage | CPU usage on the Doris BE node | CPU usage on the Doris BE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_mem | Memory usage on the Doris BE node | Memory usage on the Doris BE node | ≥ 0 | KiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cache_usage | Usage of a specified LRU cache on the Doris BE node | Usage of a specified LRU cache on the Doris BE node. This is used to monitor the memory usage. | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_engine_requests_failed_total | Total number of failed tasks of each type on the Doris BE node | Total number of failed tasks of each type on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_engine_requests_total | Total number of tasks of each type on the Doris BE node | Total number of tasks of each type on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_process_thread_num | Threads used by the Doris BE process | The threads used by Doris BE process are displayed in the /proc/pid/task directory. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_fragment_thread_pool_queue_size | Length of the queue in the current query thread pool on the Doris BE node | Length of the queue in the current query thread pool on the Doris BE node. If the value is greater than 0, the query threads are used up and queries are queued. If the value is 0, no monitoring data is reported. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_image_clean_failed | Doris FE failures to clear historical metadata image files | The count of failures for the Doris FE to clear historical metadata image files. This operation should not fail. If it fails, manual intervention is required. No monitoring data is reported if no failure occurs. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_mem_per | Doris FE memory usage | Doris FE memory usage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_mem_per | Doris BE memory usage | Doris BE memory usage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_mysql_thread_pool | Thread pools for processing tasks on the Doris FE node | Thread pools for processing tasks on the Doris FE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_thread_pool | Thread pools for interaction between the Doris FE and BE nodes | Thread pools for interaction between the Doris FE and BE nodes | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_max_tablet_compaction_score | Maximum compaction score among all Doris BE nodes | Maximum compaction score among all Doris BE nodes | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| doris_fe_thrift_rpc_latency_ms | Latency of RPC requests received by each method of the Doris FE thrift interface | Latency of RPC requests received by each method of the Doris FE thrift interface | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| system_meminfo | Host memory of the Doris FE node | Host memory of the Doris FE node | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_query_err | Error queries on Doris FE | Total number of error queries on the Doris FE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_query_err_allhost | Error queries on all Doris FE nodes | Total number of error queries on all Doris FE nodes | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| doris_fe_query_err_rate | Error queries per second on Doris FE | Number of error queries per second on Doris FE | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_query_err_rate_allhost | Error queries per second on all Doris FE nodes | Number of error queries per second on all Doris FE nodes | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| doris_fe_txn_status | Transactions in each state on Doris FE | Number of transactions in each state on the Doris FE node | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| doris_be_snmp_tcp | TCP packet status on Doris BE | TCP packet sending and receiving status on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_disks_state | Disk status of a specified data directory on Doris BE | Disk status of a specified data directory on the Doris BE node 1: Normal 0: Abnormal | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cache_usage_ratio | LRU cache usage on Doris BE | Memory usage on the Doris BE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cache_lookup_count | Lookups performed on a specified LRU cache on Doris BE | Number of lookups performed on a specified LRU cache on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cache_hit_count | Hit count of a specified LRU cache on Doris BE | Number of hits in a specified LRU cache on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cache_hit_ratio | Hit ratio of a specified LRU cache on Doris BE | Hit ratio of a specified LRU cache on the Doris BE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cache_capacity | Capacity of a specified LRU cache on Doris BE | Capacity of a specified LRU cache on the Doris BE node | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_load_channel_count | Load channels opened on Doris BE | Number of load channels currently opened on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_disks_compaction_num | Compaction tasks being executed in a specified data directory on Doris BE | Number of compaction tasks that are being executed in a specified data directory on the Doris BE node. For example, {path="/path1/"} indicates the number of tasks that are being executed in the /path1 directory. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_disks_local_used_capacity | Used local space of the disk where a specified data directory on the Doris BE node is located | Used local space of the disk where a specified data directory on the Doris BE node is located | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_disks_remote_used_capacity | Used remote directory space of the disk where a specified data directory on the Doris BE is located | Used remote directory space of the disk where a specified data directory on the Doris BE is located | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cpu | Doris BE CPU resource statistics | Doris BE CPU resource statistics, which are collected from the /proc/stat directory. | ≥ 0 | jiffies | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_disks_total_capacity | Total capacity of the disk where a specified data directory on Doris BE is located | Total capacity of the disk where a specified data directory on Doris BE is located | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_disks_avail_capacity | Remaining space of the disk where a specified data directory on Doris BE is located | Remaining space of the disk where a specified data directory on Doris BE is located | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_thread_pool_queue | Number of tasks queuing in the Doris FE thread pool for interacting with BE nodes | Number of tasks queuing in the Doris FE thread pool for interacting with BE nodes | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_mysql_thread_pool_queue | Number of tasks queuing in the Doris FE thread pool for processing tasks | Number of tasks queuing in the Doris FE thread pool for processing tasks | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_used_memory_ratio | Memory usage of the Doris FE node | Memory usage of the Doris FE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_heap_mem_used_ratio | Heap memory usage of the Doris FE node | Heap memory usage of the Doris FE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_fragment_queue_size_alarm | Length of the queue in the current query thread pool on the Doris BE node | Length of the queue in the current query thread pool on the Doris BE node. If the value is greater than 0, the query threads are used up and queries are queued. If the value is 0, no monitoring data is reported. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_snmp_tcp_in_errs_ratio | Error rate of TCP packet receiving on the Doris BE node | Error rate of TCP packet receiving on the Doris BE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_disks_used_capacity_ratio | Data disk usage on the Doris BE node | Data disk usage on the Doris BE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_max_mem_relationship_available_mem | Relationship between the maximum memory configured for Doris BE node and the available memory on the host | Relationship between the maximum memory configured for Doris BE node and the available memory on the host | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_failed_tasks_total | Total number of failed tasks of a certain type on Doris BE | Whether the number of failed tasks of a certain type on Doris BE is increasing | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| fe_status | Doris FE instance status | Doris FE instance status | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| be_status | Doris BE instance status | Doris BE instance status | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_connection_total_ratio | Percentage of MySQL port connections on Doris FE | Percentage of MySQL port connections to the maximum allowed connections of the Doris FE. This parameter is used to monitor the percentage of query connections. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_tenant_concurrency_query_num_sum | Total concurrent queries across all Doris tenants | Total concurrent queries on all FE nodes of each Doris tenant | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| doris_be_tenant_memory_usage_avg | Average memory usage per tenant in Doris | Average memory usage on all BE nodes of each Doris tenant | ≥ 0 | MiB | 1024 | cluster_id,service_name | 5 minutes |
| doris_be_tenant_cpu_usage_avg | Average CPU usage per tenant in Doris | Average CPU usage on all BE nodes of each Doris tenant | 0.00-100.00% | % | N/A | cluster_id,service_name | 5 minutes |
| doris_fe_tenant_concurrency_query_num_alarm | An alarm that is generated when the concurrent queries of a Doris tenant exceed the threshold | An alarm that is generated when the concurrent queries of a Doris tenant exceed the threshold | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_tenant_memory_usage_alarm | An alarm that is generated when the memory usage of a Doris tenant exceeds the threshold | An alarm that is generated when the memory usage of a Doris tenant exceeds the threshold | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_hot_cold_table_num | Number of cold and hot data tables on Doris FE | The parameter is used to check whether the cold and hot data tables are evenly distributed and whether the absolute values are proper. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_hot_cold_table_percent | Percentages of cold and hot data tables on Doris FE | Percentages of cold and hot data tables on Doris FE | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cold_data_volume | Cold data size on Doris BE | The parameter is used to check whether the cold data is evenly distributed and whether the absolute values are proper. | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_cold_data_volume_percent | Cold data ratio on Doris BE | Cold data ratio on Doris BE | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_cold_data_query_percent | Cold data query ratio on Doris FE | Cold data query ratio on Doris FE | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_data_get_cold_per_second_volume | Size of data cooled down per second by Doris BE | Size of data that Doris BE transfers to cold storage per second | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_obs_availability_alarm | An alarm that monitors the connection between the Doris FE node and OBS in the tiered storage scenario | An alarm that monitors the connection between the Doris FE node and OBS in the tiered storage scenario | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_tenant_concurrency_query_num | Concurrent queries of all Doris FE tenants | Number of concurrent queries of all tenants on a single FE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_tenant_memory_usage | Memory usage of all Doris BE tenants | Memory usage of all tenants on a single Doris BE node | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| doris_be_tenant_cpu_usage | CPU usage of all Doris BE tenants | CPU usage of all tenants on a single Doris BE node | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_stream_load_txn_request | Stream Load tasks on Doris BE | Stream Load tasks on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_light_work_threads_pools | bRPC light thread pools on Doris BE | bRPC light thread pools on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_be_heavy_work_threads_pools | bRPC heavy thread pools on Doris BE | bRPC heavy thread pools on the Doris BE node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_sql_defender_metric | Large SQL statement interception on Doris FE | Number of large SQL statements intercepted by the Doris FE | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| doris_fe_publish_version_queue_size | Size of the queue for executing publishVersion tasks on Doris FE | Size of the queue for executing publishVersion tasks on Doris FE | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| token_server_directmemory_calculate | Guardian TokenServer direct memory usage | Direct memory usage of Guardian TokenServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| token_server_directmemory_max | Maximum direct memory size of Guardian TokenServer | Maximum direct memory size set for Guardian TokenServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| token_server_directmemory_used | Direct memory used by Guardian TokenServer | Direct memory used by Guardian TokenServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| token_server_heapmemory_calculate | Guardian TokenServer heap memory usage | Heap memory usage of the Guardian TokenServer process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| token_server_heapmemory_max | Maximum heap memory size of Guardian TokenServer | Maximum heap memory available for the Guardian TokenServer process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| token_server_heapmemory_used | Heap memory used by Guardian TokenServer | Heap memory used by the Guardian TokenServer process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| token_server_nonheapmemory_calculate | Guardian TokenServer non-heap memory usage | Percentage of non-heap memory used by the Guardian TokenServer process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| token_server_nonheapmemory_max | Maximum non-heap memory size of Guardian TokenServer | Maximum non-heap memory available for the Guardian TokenServer process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| token_server_nonheapmemory_used | Non-heap memory used by Guardian TokenServer | Non-heap memory used by the Guardian TokenServer process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| token_server_request_iam_interface_count | Requests initiated by Guardian TokenServer to IAM | Number of requests initiated by Guardian TokenServer to IAM | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| token_server_sts_interface_received_request_count | Requests received by the Guardian TokenServer STS interface | Number of requests received by the Guardian TokenServer STS interface | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| pqs_connection_used_rate | HBase PhoenixQueryServer connection usage | Connection usage of HBase PhoenixQueryServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_cpu_core_num | CPU cores of HBase PhoenixQueryServer | Number of CPU cores of HBase PhoenixQueryServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_cpu_usage_calculate | HBase PhoenixQueryServer CPU usage statistics | CPU usage statistics of HBase PhoenixQueryServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_cpu_used | HBase PhoenixQueryServer CPU usage | Percentage of CPU resources used by the HBase PhoenixQueryServer process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_direct_memory_used_calculate | HBase PhoenixQueryServer direct memory usage | Direct memory usage of HBase PhoenixQueryServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_directmemMax | Direct memory allocated to HBase PhoenixQueryServer | Size of the direct memory allocated to HBase PhoenixQueryServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| pqs_directmemUsed | Direct memory used by HBase PhoenixQueryServer | Size of the direct memory used by HBase PhoenixQueryServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| pqs_gcCount | GC count of HBase PhoenixQueryServer | Total number of GCs occurred on HBase PhoenixQueryServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_gcTimeMillis | Total GC time of HBase PhoenixQueryServer | Total GC time of HBase PhoenixQueryServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_gcTimeMillisParNew | GC duration in the young area of HBase PhoenixQueryServer | Time required for GC in the young area of HBase PhoenixQueryServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_gcTimeMillisParOld | GC duration in the old area of HBase PhoenixQueryServer | Time required for GC in the old area of HBase PhoenixQueryServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_heap_memory_used_calculate | HBase PhoenixQueryServer heap memory usage | Heap memory usage of HBase PhoenixQueryServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_mem_usage | HBase PhoenixQueryServer memory usage | HBase PhoenixQueryServer memory usage | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| pqs_memheapcommittedm | Heap memory allocated to HBase PhoenixQueryServer | Heap memory allocated to HBase PhoenixQueryServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| pqs_memheapusedm | Heap memory used by HBase PhoenixQueryServer | Heap memory used by HBase PhoenixQueryServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| pqs_memnonheapcommittedm | Non-heap memory allocated to HBase PhoenixQueryServer | Non-heap memory allocated to HBase PhoenixQueryServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| pqs_memnonheapusedm | Non-heap memory used by HBase PhoenixQueryServer | Non-heap memory used by HBase PhoenixQueryServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| pqs_numOpenConnections | Opened HBase PhoenixQueryServer connections | Number of opened HBase PhoenixQueryServer connections | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_threadsblocked | HBase PhoenixQueryServer threads in the blocked state | Number of HBase PhoenixQueryServer threads in the blocked state | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_threadsnew | HBase PhoenixQueryServer threads in the new state | Number of HBase PhoenixQueryServer threads in the new state | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_threadsrunnable | HBase PhoenixQueryServer threads in the runnable state | Number of HBase PhoenixQueryServer threads in the runnable state | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_threadsterminated | HBase PhoenixQueryServer threads in the terminated state | Number of HBase PhoenixQueryServer threads in the terminated state | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| pqs_threadswaiting | HBase PhoenixQueryServer threads in the waiting state | Number of HBase PhoenixQueryServer threads in the waiting state | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_pread_time_999th_percentile | P999 Pread latency in HBase RegionServer | 999th percentile of the Pread latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_pread_time_99th_percentile | P99 Pread latency in HBase RegionServer | 99th percentile of the Pread latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_pread_time_max | Maximum Pread latency in HBase RegionServer | Maximum Pread latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_pread_time_max_allhost | Maximum Pread latency in HBase RegionServer | Maximum Pread latency across all hosts in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_pread_time_max_allhost_calculate | Maximum Pread latency in HBase RegionServer | Maximum Pread latency across all hosts in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_read_time_999th_percentile | P999 read latency in HBase RegionServer | 999th percentile of the read latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_read_time_99th_percentile | P99 read latency in HBase RegionServer | 99th percentile of the read latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_read_time_max | Maximum read latency in HBase RegionServer | Maximum read latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_read_time_max_allhost | Maximum read latency in HBase RegionServer | Maximum read latency across all hosts in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_read_time_max_allhost_calculate | Maximum read latency in HBase RegionServer | Maximum read latency across all hosts in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_write_time_999th_percentile | P999 write latency in HBase RegionServer | 999th percentile of the write latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_write_time_99th_percentile | P99 write latency in HBase RegionServer | 99th percentile of the write latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_write_time_max | Maximum write latency in HBase RegionServer | Maximum write latency in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_write_time_max_allhost | Maximum write latency in HBase RegionServer | Maximum write latency across all hosts in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_fs_write_time_max_allhost_calculate | Maximum write latency in HBase RegionServer | Maximum write latency across all hosts in HBase RegionServer | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_numCallsInReadQueue | RPC read queues in HBase RegionServer | Number of RPC read queues in HBase RegionServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_numCallsInWriteQueue | RPC write queues in HBase RegionServer | Number of RPC write queues in HBase RegionServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| rs_slowWalSyncCount | Timed-out WAL writes in HBase RegionServer | Number of timed-out WAL writes in HBase RegionServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| dn_readblockopavgtime | Average ReadBlock time of HDFS DataNode | Average time for HDFS DataNode to perform the ReadBlock operation | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| dn_slowacktoupstream_speed | Slow AckToUpstream occurrences per second on HDFS DataNode | Number of slow AckToUpstream operations per second on HDFS DataNode | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| dn_slowflushorsync_speed | Slow Flush or Sync occurrences per second on HDFS DataNode | Number of slow Flush or Sync operations per second on HDFS DataNode | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| dn_slowsyncwriteroscache_speed | Slow SyncWriterOsCache occurrences per second on HDFS DataNode | Number of slow SyncWriterOsCache operations per second on HDFS DataNode | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| dn_slowwritedatatodisk_speed | Slow WriteDataToDisk occurrences per second on HDFS DataNode | Number of slow WriteDataToDisk operations per second on HDFS DataNode | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| dn_slowwritepackettodownstream_speed | Slow WritePacketToDownStream occurrences per second on HDFS DataNode | Number of slow WritePacketToDownStream operations per second on HDFS DataNode | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| dn_writeblockopavgtime | Average WriteBlock time of HDFS DataNode | Average time for HDFS DataNode to perform the WriteBlock operation | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| dn_xceivercount | Connections used by HDFS DataNode to receive data | Number of connections used by HDFS DataNode to receive data | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| nn_rpc_queuetime_numops_default | Total HDFS RPCs | Total number of RPCs per second in the HDFS | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| qas_cpuusage | HetuEngine QAS CPU usage | CPU usage of the HetuEngine QAS process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| qas_disk_usage | HetuEngine QAS disk usage | Disk usage of HetuEngine QAS | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| qas_gc_time | GC time of the HetuEngine QAS process | GC time of the HetuEngine QAS process | ≥ 0 | s | N/A | cluster_id,node_id,role_name | 5 minutes |
| qas_memusage | Memory usage of the HetuEngine QAS process | Memory used by the HetuEngine QAS process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| hs_close_connect_duration | HiveServer connection closing duration | Time required for HiveServer to close a connection | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| hs_establish_connect_duration | HiveServer connection setup duration | Time required for HiveServer to set up a connection | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| hs_failedhqlnum | HQL statements that fail to be executed by HiveServer | Number of HQL statements that fail to be executed by Hive | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| hs_mean_compile_time | Average HQL compilation time | Average HQL compilation time | ≥ 0 | s | N/A | cluster_id,node_id,role_name | 5 minutes |
| hs_mean_execute_time | Average HQL execution time | Average HQL execution time | ≥ 0 | s | N/A | cluster_id,node_id,role_name | 5 minutes |
| hs_mean_sql_time | Average HQL completion time | Average HQL completion time | ≥ 0 | s | N/A | cluster_id,node_id,role_name | 5 minutes |
| hs_scan_parts_num | Number of partitions scanned by HiveServer | Number of partitions scanned by HiveServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_create_table_db_75th | 75th percentile of the execution time of the metadata database during Hive MetaStore table creation | Time in which 75% of the execution times for MetaStore calls to underlying database APIs during table creation is complete | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_create_table_db_99th | 99th percentile of the execution time of the metadata database during Hive MetaStore table creation | Time in which 99% of the execution times for MetaStore calls to underlying database APIs during table creation is complete | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_create_table_db_max | Maximum execution time of the metadata database during Hive MetaStore table creation | Maximum execution time for MetaStore calls to underlying database APIs during table creation | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_create_table_db_mean | Average execution time of the metadata database during Hive MetaStore table creation | Average execution time for MetaStore calls to underlying database APIs during table creation | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_create_table_db_min | Minimum execution time of the metadata database during Hive MetaStore table creation | Minimum execution time for MetaStore calls to underlying database APIs during table creation | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_get_database_75th | 75th percentile of the execution time for Hive MetaStore database query | Time in which 75% of get_database API execution in MetaStore is complete | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_get_database_99th | 99th percentile of the execution time for Hive MetaStore database query | Time in which 99% of get_database API execution in MetaStore is complete | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_get_database_max | Maximum execution time of the Hive MetaStore get_database API | Maximum duration of executing the get_database API by MetaStore | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_get_database_mean | Average execution time of the Hive MetaStore get_database API | Average duration of executing the get_database API by MetaStore | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_api_get_database_min | Minimum execution time of the Hive MetaStore get_database API | Minimum duration of executing the get_database API by MetaStore | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_close_connect_duration | Hive MetaStore connection closing duration | Time required for MetaStore to close a connection | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_establish_connect_duration | Hive MetaStore connection setup duration | Time required for MetaStore to set up a connection | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_host_session_map | Client sessions connected to MetaStore from Hive | Number of client sessions connected to MetaStore | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_last_5_minute_mean_create_table_time | Average execution time of the metadata database during Hive MetaStore table creation in the last 5 minutes | Average execution duration of all create_table API metadata operations performed by MetaStore in the last five minutes | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ms_user_session_group | User sessions connected to MetaStore from Hive | Number of user sessions connected to MetaStore | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| catalog_cpuusage | CPU usage of the Impala Catalog process | CPU usage of the Impala Catalog process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| catalog_gc_times | Impala Catalog GC times in a period | GC times of different durations in the Impala process in a period | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| catalog_gc_times_exceeds_1s | Total Impala Catalog GC times | GC times of different durations in the Impala process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| catalog_jvm_sleep_time | Impala Catalog GC Time | JVM pause time caused by GC in the Impala process | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| catalog_memusage | Memory usage of the Impala Catalog process | Memory resource usage of the Impala Catalog process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| impala_thrift_server_beeswax_frontend_total_connections | Connections to Impala through the Beeswax API | Total number of Beeswax API connections made to the Impala Daemon over its lifetime | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| impala_thrift_server_hiveserver2_frontend_connection_setup_queue_size | HiveServer2 API connections to Impala that have been accepted and are waiting to be set (across all instances) | Number of HiveServer2 API connections to the Impala Daemon that have been accepted and are waiting to be set | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| impala_thrift_server_hiveserver2_frontend_connections_in_use | Active HiveServer2 API connections to Impala (across all instances) | Number of active HiveServer2 API connections to the Impala Daemon | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| impala_thrift_server_hiveserver2_frontend_total_connections | Total connections to Impala through the HiveServer2 API (across all instances) | Total number of HiveServer2 API connections made to the Impala Daemon over its lifetime | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| impala_thrift_server_hiveserver2_http_frontend_connection_setup_queue_size | HiveServer2 HTTP API connections to Impala that have been accepted and are waiting to be set (across all instances) | Number of HiveServer2 HTTP API connections to the Impala Daemon that have been accepted and are waiting to be set | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| impala_thrift_server_hiveserver2_http_frontend_connections_in_use | Active HiveServer2 HTTP API connections to Impala (across all instances) | Number of active HiveServer2 HTTP API connections to the Impala Daemon | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| impala_thrift_server_hiveserver2_http_frontend_total_connections | HiveServer2 connections to Impala | Number of HiveServer2 connections to Impala | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| impalad_cpuusage | CPU usage of the Impalad process | CPU resource usage of the Impalad process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| impalad_gc_times | Impalad GC times in a period | GC times of different durations in the Impala process in a period | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| impalad_gc_times_exceeds_1s | Total Impalad GC times | GC times of different durations in the Impala process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| impalad_jdbc_number | Number of JDBC connections to Impalad | Number of JDBC connections to the Impalad process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| impalad_jvm_sleep_time | Impalad GC time | JVM pause time caused by GC in the Impala process | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| impalad_memusage | Memory usage of the Impalad process | Memory resource usage of the Impalad process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| impalad_odbc_number | Number of ODBC connections to Impalad | Number of ODBC connections to the Impalad process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| memory_mapped_bytes | Virtual memory size of the Impalad process (across all instances) | Total bytes of memory mappings in the Impalad process (the virtual memory size) | ≥ 0 | MiB | 1024 | cluster_id,service_name | 5 minutes |
| memory_rss | Physical memory size of the Impalad process | Resident set size (RSS) of the Impalad process, including TCMalloc, buffer pool, and JVM | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| num_executing_queries | Queries being executed by Impala | Number of queries being executed by Impala | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| num_in_flight_queries | Queries being submitted by Impala across all instances | Number of queries being submitted by Impala | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| num_in_flight_queries_allhost | Query tasks in the Impala process | Total number of query tasks in the Impala process | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| num_waiting_queries | Queries waiting in the Impala process across all instances | Number of queries waiting in the Impala process | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| js_cpu_usage | JobServer CPU usage | CPU resource usage of the JobServer process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_gc_fgc_usage | JobServer full GC times | Number of JobServer full GC times | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_gc_old_usage | Memory usage of the old-generation JobServer | Memory usage of the old-generation JobServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_memory_per_usage | Memory usage of the JobServer process | Memory resource usage of the JobServer process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_memory_usage | Direct memory usage of JobServer | Memory usage of JobServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_active_job | JobServer tasks being executed | Number of tasks being executed by JobServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_active_job_allhost | Tasks being executed in the JobServer cluster | Number of tasks being executed in the JobServer cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| js_metrics_events_active_nginx_connect | JobBalancer connections | Number of connections being processed by the JobBalancer process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_active_nginx_connect_allhost | Connections being processed by the JobBalancer cluster | Number of JobBalancer cluster connections | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| js_metrics_events_kill_delay | JobServer job termination API latency | Latency of the JobServer job termination API | ≥ 0 | s | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_kill_failure_rate | JobServer job termination API failure rate | Failure rate of the JobServer job termination API | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_query_delay | JobServer job query API latency | Latency of the JobServer job query API | ≥ 0 | s | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_query_failure_rate | JobServer job query API failure rate | Failure rate of the JobServer job query API | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_submit_delay | JobServer job submission API latency | Latency of the JobServer job submission API | ≥ 0 | s | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_submit_failure_rate | JobServer job submission API failure rate | Failure rate of the JobServer job submission API | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_wait_job | JobServer tasks to be executed | Number of JobServer tasks waiting to be executed | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_metrics_events_wait_job_allhost | Tasks to be executed in the JobServer cluster | Number of tasks waiting to be executed in the JobServer cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| bk_underreplicatedpartitions_allhost_percentage_calculate | Percentage of Kafka partitions that are not completely synchronized | Percentage of the total number of partitions that are not completely synchronized on all brokers in the Kafka cluster to the total number of partitions | 0.00-100.00% | % | N/A | cluster_id,service_name | 5 minutes |
| mm_cpuusage | MirrorMaker CPU usage | CPU usage of MirrorMaker | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| mm_memusage | MirrorMaker memory usage | Memory used by MirrorMaker | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| kudu_data_dirs_full | Used-up KuduTserver disks | Number of KuduTserver disks that are fully occupied | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| tserver_mem_percentage_consumed | KuduTServer memory usage | KuduTServer memory usage, in percentage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| tserver_tablet_leader_num | Running tablet leaders on KuduTServer | Number of running tablet leaders on KuduTServer, which is used to monitor the TServer balancing status of each node | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| tserver_tablet_leader_num_allhost | Running tablet leaders on all KuduTServers in the cluster | Number of running tablet leaders on all KuduTServers in the cluster, which is used to monitor the load status of the Kudu cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| tserver_tablet_num | Running tablets on KuduTServer | Number of running tablets on KuduTServer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| tserver_tablet_num_allhost | Running tablets on all KuduTServers in the cluster | Number of running tablets on all KuduTServers in the cluster, which is used to monitor the average load status of the Kudu cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| ldms_cpuusage | LDMS LDMSServer CPU usage | CPU usage of the LDMS LDMSServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| ldms_memusage | LDMS LDMSServer memory usage | Memory usage of the LDMS LDMSServer process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| dev_cpu_ni | CPU usage of processes whose priorities have been changed on the MRS Manager host | The percentage of CPU time consumed by processes whose priorities have been changed on the MRS Manager host in a monitoring interval | 0.00-100.00% | % | N/A | cluster_id,node_id | 5 minutes |
| dev_disk_r_await_per_second | Average disk read time on MRS Manager host | Average time required for each read operation on the MRS Manager host disk | ≥ 0 | ms | N/A | cluster_id,node_id,device_name | 5 minutes |
| dev_disk_w_await_per_second | Average disk write time on MRS Manager host | Average time required for each write operation on the MRS Manager host disk | ≥ 0 | ms | N/A | cluster_id,node_id,device_name | 5 minutes |
| dev_instance_process_memory_usage | Memory usage of the MRS Manager host process | Memory used by the component instance process and MRS Manager host process | ≥ 0 | MiB | 1024 | cluster_id,node_id | 5 minutes |
| dev_is_dstatus_process_ok | Processes in D and Z states on the MRS Manager host | Total number of processes in the D and Z states on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_load_value | Average load per CPU core on the MRS Manager host | Average load per CPU core on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_mem_buffers | Kernel cache memory size on the MRS Manager host | Total memory used for the kernel cache on the MRS Manager host in a monitoring interval | ≥ 0 | MiB | 1024 | cluster_id,node_id | 5 minutes |
| dev_mem_cached | Cache memory size on the MRS Manager host | Cache memory size on the MRS Manager host in a monitoring interval | ≥ 0 | MiB | 1024 | cluster_id,node_id | 5 minutes |
| dev_mem_shared | Shared memory size on the MRS Manager host | Shared memory size on the MRS Manager host in a monitoring interval | ≥ 0 | MiB | 1024 | cluster_id,node_id | 5 minutes |
| dev_net_host_read_errors | MRS Manager host network read packet errors | Number of read packet errors on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_host_read_frames | MRS Manager host network data packet frame errors | Number of data packet frame errors on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_host_write_collisions | MRS Manager host network packet collisions | Number of data packet collisions on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_host_write_dropped | MRS Manager host network write packet drops | Number of write packets discarded by the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_host_write_errors | MRS Manager host network write packet errors | Number of write packet errors on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_read_bs_speed | MRS Manager host network port read rate | Read rate of network port on the MRS Manager host | ≥ 0 | KiB/s | 1024 | cluster_id,node_id | 5 minutes |
| dev_net_read_errors | MRS Manager host read packet errors | Number of read packet errors on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_read_frames | MRS Manager host data packet frame errors | Number of data packet frame errors on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_read_throughput | MRS Manager host read throughput | Read throughput of the MRS Manager host | 0.00-100.00% | % | N/A | cluster_id,node_id | 5 minutes |
| dev_net_write_collisions | MRS Manager host data packet collisions | Number of data packet collisions on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_write_dropped | MRS Manager host write packet drops | Number of write packets discarded by the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_write_dropped_ratio | MRS Manager host write packet drop rate | Write packet drop rate on the MRS Manager host | 0.00-100.00% | % | N/A | cluster_id,node_id | 5 minutes |
| dev_net_write_errors | MRS Manager host write packet errors | Number of write packet errors on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_net_write_errors_ratio | MRS Manager host write packet error rate | Write packet error rate on the MRS Manager host | 0.00-100.00% | % | N/A | cluster_id,node_id | 5 minutes |
| dev_net_write_throughput | MRS Manager host write throughput | Write throughput of the MRS Manager host | 0.00-100.00% | % | N/A | cluster_id,node_id | 5 minutes |
| dev_node_agent_fgc_count | NodeAgent Full GC count on the MRS Manager host | Number of NodeAgent Full GCs on the MRS Manager host | ≥ 0 | N/A | N/A | cluster_id,node_id | 5 minutes |
| dev_semaphore_array_usage_rate | MRS Manager Host semaphore array usage | Semaphore array usage of the MRS Manager host in a monitoring interval | 0.00-100.00% | % | N/A | cluster_id,node_id | 5 minutes |
| dev_temp_port_usage | MRS Manager host TCP ephemeral port usage | TCP ephemeral port usage on the MRS Manager host | 0.00-100.00% | % | N/A | cluster_id,node_id | 5 minutes |
| kadmind_connected | KerberosAdmin connections | Number of KerberosAdmin process connections | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| kadmind_cpuusage | KerberosAdmin CPU usage | CPU usage of KerberosAdmin | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| kadmind_memusage | KerberosAdmin memory usage | Memory used by KerberosAdmin | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| krb5kdc_allreqs_avg | KerberosServer requests | Total number of KerberosServer requests | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| krb5kdc_asreqs_avg | KerberosServer AS requests | Number of KerberosServer AS requests | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| krb5kdc_connected | KerberosServer connections | Number of KerberosServer connections | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| krb5kdc_cpuusage | KerberosServer CPU usage | CPU usage of KerberosServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| krb5kdc_memusage | KerberosServer memory usage | Memory used by KerberosServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| krb5kdc_tgsreqs_avg | KerberosServer TGS requests | Number of KerberosServer TGS requests | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ldapserver_connected | SlapdServer connections | Number of SlapdServer connections | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| ldapserver_cpuusage | SlapdServer CPU usage | CPU usage of SlapdServer | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| ldapserver_memusage | SlapdServer memory usage | Memory used by SlapdServer | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| worker_cpuusage | MemArtsCC Worker CPU usage | CPU resource usage of the MemArtsCC Worker process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_evict_shard_num_history_usage | Historically evicted shards of MemArtsCC Worker | Number of shards that have been evicted from the MemArtsCC Worker process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_evict_shard_size_history_usage | Historically evicted shard size of MemArtsCC Worker | Size of shards that have been evicted from the MemArtsCC Worker process | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| worker_latency_avg_usage | Average read request latency of MemArtsCC Worker | Average latency of the MemArtsCC Worker process | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_latency_max_usage | Maximum read request latency of MemArtsCC Worker | Maximum latency of the MemArtsCC Worker process | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_latency_min_usage | Minimum read request latency of MemArtsCC Worker | Minimum latency of the MemArtsCC Worker process | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_mem_percent | MemArtsCC Worker memory usage | Memory resource usage of the MemArtsCC Worker process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_process_completed_load_tasks | Historical MemArtsCC load tasks | Number of historical load tasks processed by MemArtsCC | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_process_completed_load_tasks_allhost | Historical load tasks in a MemArtsCC cluster | Number of historical load tasks processed by a MemArtsCC cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| worker_process_hit_count_distribution_huge_usage_allhost | Cache shard hit distribution in a MemArtsCC cluster | Distribution of cache shard hits in a MemArtsCC cluster | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| worker_process_hit_count_distribution_large_usage | Cache shard hit distribution in MemArtsCC | Distribution of cache shard hits in MemArtsCC | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_process_obsfetch_effected_traffic_usage_calculate | OBS bandwidth reduction rate of the MemArtsCC worker process | OBS bandwidth reduction rate of the MemArtsCC worker process. It is calculated using the following formula: (Traffic read by OBSA from OBS when MemArtsCC is enabled + Traffic asynchronously cached by MemArtsCC from OBS when MemArtsCC is enabled)/Traffic read by OBSA from OBS when MemArtsCC is disabled. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_process_obsfetch_miss_traffic_usage | Traffic details when MemArtsCC is enabled | Traffic details when MemArtsCC is enabled | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| worker_process_obsfetch_original_traffic_usage_allhost | Periodic OBS read traffic before and after MemArtsCC is enabled | Periodic OBS read traffic before and after MemArtsCC is enabled | ≥ 0 | MiB | 1024 | cluster_id,service_name | 5 minutes |
| worker_process_obsfetch_wasted_traffic_usage | Periodic OBS read traffic before and after MemArtsCC is enabled | Periodic OBS read traffic before and after MemArtsCC is enabled | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| worker_read_bandwidth_usage_allhost | MemArtsCC cluster read bandwidth | Read bandwidth of a MemArtsCC cluster | ≥ 0 | Byte/s | 1024 | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| store_client_proxy_handle_flush_message_avg_latency | Average latency for MemArtsStore StoreClientProxy to process flush messages | Average latency for MemArtsStore StoreClientProxy to process flush messages. This metric is used to monitor the running status of the StoreClientProxy process. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_client_proxy_handle_write_message_avg_latency | Average latency for MemArtsStore StoreClientProxy to process write messages | Average latency for MemArtsStore StoreClientProxy to process write messages. This metric is used to monitor the running status of the StoreClientProxy process. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_client_proxy_process_connections_to_worker | Connections between MemArtsStore StoreClientProxy and StoreWorker | Number of connections between MemArtsStore StoreClientProxy and StoreWorker. This metric is used to monitor the server network resources occupied by the StoreClientProxy process. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_client_proxy_process_memory_used | MemArtsStore StoreClientProxy memory usage | Memory usage of MemArtsStore StoreClientProxy. This metric is used to monitor the server memory resources occupied by the StoreClientProxy process. | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| store_client_proxy_socket_cpu_usage | MemArtsStore StoreClientProxy CPU usage | CPU usage of MemArtsStore StoreClientProxy. This metric is used to monitor the server CPU resources occupied by the StoreClientProxy process. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_process_connections | MemArtsStore StoreMaster connections | Number of connections to the MemArtsStore StoreMaster process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_process_fds | File descriptors in the MemArtsStore StoreMaster process | Number of files loaded by the MemArtsStore StoreMaster process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_process_memory | MemArtsStore StoreMaster memory usage | Memory resource usage of the MemArtsStore StoreMaster process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| store_master_process_store_handle_cpu | MemArtsStore StoreMaster CPU usage | CPU resource usage of the MemArtsStore StoreMaster process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_process_threads | MemArtsStore StoreMaster process threads | Number of threads used by the MemArtsStore StoreMaster process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_process_worker_sdk_connections_ratio | Percentage of MemArtsStore StoreMaster SDK network connections to the configured upper limit | Percentage of MemArtsStore StoreMaster SDK network connections to the configured upper limit. This metric is used to monitor the resource usage of the StoreMaster process. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_space_ratio_num_allhost | MemArtsStore total capacity utilization | MemArtsStore total capacity utilization. This metric is used to monitor the resource usage of the MemArtsStore cluster. | 0.00-100.00% | % | N/A | cluster_id,service_name | 5 minutes |
| store_master_stop_remote_latency_max | Latency for MemArtsStore Master to instruct Worker to delete data | Latency for MemArtsStore StoreMaster to instruct Worker to delete data. This metric is used to monitor the running status of the StoreMaster process. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_worker_process_fds_ratio | Percentage of used MemArtsStore StoreMaster FDs to the configured upper limit | Percentage of used MemArtsStore StoreMaster FDs to the configured upper limit. This metric is used to monitor the resource usage of the StoreMaster process. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_worker_routine_total_ratio | MemArtsStore StoreMaster process coroutine usage | Goroutine resource usage of the MemArtsStore StoreMaster process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_zk_read_iops | ZooKeeper read IOPS of MemArtsStore StoreMaster | Number of ZooKeeper read operations performed by MemArtsStore StoreMaster per second. This metric is used to observe and analyze the load and latency of StoreMaster accessing ZooKeeper. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_zk_read_latency | Average latency for MemArtsStore StoreMaster to read data from ZooKeeper | Average latency for MemArtsStore StoreMaster to read data from ZooKeeper. This metric is used to observe and analyze the load and latency of StoreMaster accessing ZooKeeper. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_zk_write_iops | ZooKeeper write IOPS of MemArtsStore StoreMaster | Number of ZooKeeper write operations performed by MemArtsStore StoreMaster per second. This metric is used to observe and analyze the load and latency of StoreMaster accessing ZooKeeper. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_master_zk_write_latency | Average latency for MemArtsStore StoreMaster to write data to ZooKeeper | Average latency for MemArtsStore StoreMaster to write data to ZooKeeper. This metric is used to observe and analyze the load and latency of StoreMaster accessing ZooKeeper. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_node_pool_disk_total | MemArtsStore StoreWorker disk capacity | Total disk capacity of a MemArtsStore StoreWorker node. This metric is used to monitor the used capacity of the StoreWorker node. | ≥ 0 | GiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| store_node_pool_disk_usage_calculate | MemArtsStore StoreWorker disk usage | Disk usage of a MemArtsStore StoreWorker node. This metric is used to monitor the used disk capacity of the StoreWorker node. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_sidecar_cpu_usage | MemArtsStore StoreSideCar CPU usage | CPU resource usage of the MemArtsStore StoreSideCar process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_sidecar_mem_used | MemArtsStore StoreSideCar memory usage | Memory resource usage of the MemArtsStore StoreSideCar process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| store_worker_pool_current_mplog_water_level | MPLog usage of the MemArtsStore StoreWorker Pool | MPLog usage of the MemArtsStore StoreWorker Pool, which is used to monitor and analyze the pool usage. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_pool_file_write_latency | Average file operation latency of the MemArtsStore StoreWorker Pool | Average file operation latency of the MemArtsStore StoreWorker Pool, which is used to monitor and analyze the pool load and processing latency. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_pool_mplog_delete_latency | Average latency of MPLog operations in MemArtsStore StoreWorker Pool | Average latency of MPLog operations in the MemArtsStore StoreWorker Pool, which is used to monitor and analyze the pool load and processing latency. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_pool_mplog_relocate_num | MPLog eviction allocations in the MemArtsStore StoreWorker Pool during a monitoring interval | Number of MPLog eviction allocations in the MemArtsStore StoreWorker Pool during a monitoring interval, which is used to monitor and analyze the pool load and processing latency. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_pool_mplog_timeout_free_num | Number of MPLog timeout releases in the MemArtsStore StoreWorker Pool during a monitoring interval | Number of MPLog files released due to timeout in the MemArtsStore StoreWorker Pool during a monitoring interval. This metric is used to monitor and analyze the Pool load and processing latency. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_pool_task_schedule_latency | Average task scheduling latency of MemArtsStore StoreWorker Pool | Average task scheduling latency of MemArtsStore StoreWorker Pool, which is used to monitor and analyze the pool load and processing latency. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_process_connections | MemArtsStore StoreWorker connections | Number of connections to the MemArtsStore StoreWorker process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_process_cpu_usage_allhost | MemArtsStore StoreWorker cluster CPU usage | CPU resource usage of the MemArtsStore StoreWorker process | 0.00-100.00% | % | N/A | cluster_id,service_name | 5 minutes |
| store_worker_process_fds | File descriptors in the MemArtsStore StoreWorker process | Number of files loaded by the MemArtsStore StoreWorker process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_process_memory_used | MemArtsStore StoreWorker memory usage | Memory resource usage of the MemArtsStore StoreWorker process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| store_worker_process_memory_used_allhost | MemArtsStore StoreWorker cluster memory usage | Memory resource usage of the MemArtsStore StoreWorker process | ≥ 0 | MiB | 1024 | cluster_id,service_name | 5 minutes |
| store_worker_process_store_network_cpu_usage | CPU usage of the service thread in the MemArtsStore StoreWorker process | CPU resource usage of the MemArtsStore StoreWorker process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_process_worker_sdk_connections_ratio | Percentage of MemArtsStore StoreWorker SDK network connections to the configured upper limit | Percentage of MemArtsStore StoreWorker SDK network connections to the configured upper limit. This metric is used to monitor the resource usage of the StoreWorker process. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_read_flows | Read resource usage of MemArtsStore StoreWorker | Resource usage of the read operations in the MemArtsStore StoreWorker process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_read_iops | Read IOPS of MemArtsStore StoreWorker | Number of read operations per second in the MemArtsStore StoreWorker process. This metric is used to monitor the processing capability and load characteristics of the StoreWorker process. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_read_iops_allhost | Total read IOPS of the MemArtsStore Store cluster | Total number of read operations per second in the MemArtsStore cluster. This metric is used to monitor the processing capability and load characteristics of the StoreWorker process. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| store_worker_read_latency_max | MemArtsStore StoreWorker read latency | MemArtsStore StoreWorker read latency. This metric is used to monitor the processing capability and load characteristics of the StoreWorker process. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_sdk_read_first_latency_avg | MemArtsStore StoreWorker SDK average end-to-end first read latency | Average end-to-end first read latency of MemArtsStore StoreWorker SDK. This metric is used to monitor and analyze end-to-end load characteristics. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_sdk_read_iops | End-to-end read IOPS of MemArtsStore StoreWorker SDK | Number of end-to-end read operations per second on MemArtsStore StoreWorker SDK. This metric is used to monitor and analyze end-to-end load characteristics. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_sdk_read_iops_allhost | SDK read IOPS of the MemArtsStore Store cluster | Number of SDK read operations per second in the MemArtsStore Store cluster. This metric is used to monitor and analyze end-to-end load characteristics of the entire cluster. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| store_worker_sdk_write_iops | End-to-end write IOPS of MemArtsStore StoreWorker SDK | Number of end-to-end write operations per second on MemArtsStore StoreWorker SDK. This metric is used to monitor and analyze end-to-end load characteristics. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_sdk_write_iops_allhost | SDK write IOPS of the MemArtsStore Store cluster | Number of SDK write operations per second in the MemArtsStore cluster. This metric is used to monitor and analyze end-to-end load characteristics of the entire cluster. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| store_worker_sdk_write_latency_avg | Average end-to-end write latency of the MemArtsStore StoreWorker SDK | Average end-to-end write latency of the MemArtsStore StoreWorker SDK. This metric is used to monitor and analyze end-to-end load characteristics. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_worker_process_fds_ratio | Percentage of used MemArtsStore StoreWorker FDs to the configured upper limit | Percentage of used MemArtsStore StoreWorker FDs to the configured upper limit. This metric is used to monitor the resource usage of the StoreWorker process. | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_worker_routine_total_ration | MemArtsStore StoreWorker coroutine usage | Goroutine resource usage of the MemArtsStore StoreWorker process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_write_flows | Percentage of concurrent MemArtsStore StoreWorker writes | Percentage of concurrent write operations in the MemArtsStore StoreWorker process | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_write_iops | Write IOPS of MemArtsStore StoreWorker | Number of write operations per second in the MemArtsStore StoreWorker process. This metric is used to monitor the processing capability and load characteristics of StoreWorker. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_write_iops_allhost | Total write IOPS of the MemArtsStore Store cluster | Total number of write operations per second in the MemArtsStore cluster. This metric is used to monitor the processing capability and load characteristics of StoreWorker. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| store_worker_write_latency_avg | MemArtsStore StoreWorker write latency | MemArtsStore StoreWorker write latency. This metric is used to monitor the processing capability and load characteristics of StoreWorker. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| store_worker_write_memory_flows | Ratio of MemArtsStore StoreWorker write resource usage to the flow control threshold | Percentage of memory resources used by MemArtsStore StoreWorker write operations | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_client_read_meta_iops | MemArtsStore KV read IOPS | Number of MemArtsStore KV metadata read operations per second. This metric is used to monitor and analyze the KV metadata read load and processing latency. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_client_read_meta_iops_allhost | Total KV metadata read IOPS of the MemArtsStore cluster | Total number of KV metadata read operations per second in the MemArtsStore cluster. This metric is used to monitor and analyze the KV metadata read load and processing latency. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| worker_kv_client_read_meta_latency_max | MemArtsStore KV metadata read latency | Maximum latency of MemArtsStore KV metadata read operations. This metric is used to monitor and analyze the KV metadata read load and processing latency. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_client_write_meta_iops | MemArtsStore KV metadata write IOPS | Number of MemArtsStore KV metadata write operations per second. This metric is used to monitor and analyze the KV metadata write load and processing latency. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_client_write_meta_iops_allhost | Total KV metadata write IOPS of the MemArtsStore cluster | Total number of KV metadata write operations per second in the MemArtsStore cluster. This metric is used to monitor and analyze the KV metadata write load and processing latency. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| worker_kv_client_write_meta_latency_max | MemArtsStore KV metadata write latency | MemArtsStore KV metadata write latency. This metric is used to monitor and analyze the KV metadata write load and processing latency. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_sdk_append_bandwidth | MemArtsStore KV SDK write bandwidth | MemArtsStore KV SDK write bandwidth. This metric is used to monitor and analyze the load and processing latency of the KV SDK node. | ≥ 0 | Byte/s | 1024 | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_sdk_append_iops | MemArtsStore KV SDK write IOPS | Number of MemArtsStore KV SDK write operations per second. This metric is used to monitor and analyze the load and processing latency of the KV SDK node. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_sdk_append_iops_allhost | Total KV SDK write IOPS of the MemArtsStore cluster | Total number of KV SDK write operations per second in the MemArtsStore cluster. This metric is used to monitor and analyze the load and processing latency of the StoreWorker node. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| worker_kv_sdk_append_latency_avg | MemArtsStore KV SDK write latency | MemArtsStore KV SDK write latency. This metric is used to monitor and analyze the load and processing latency of the KV SDK node. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_sdk_read_bandwidth | MemArtsStore KV SDK read bandwidth | MemArtsStore KV SDK read bandwidth. This metric is used to monitor and analyze the load and processing latency of the KV SDK node. | ≥ 0 | Byte/s | 1024 | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_sdk_read_iops | MemArtsStore KV SDK read IOPS | Number of MemArtsStore KV SDK read operations per second. This metric is used to monitor and analyze the load and processing latency of the KV SDK node. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| worker_kv_sdk_read_iops_allhost | Total KV SDK read IOPS in the MemArtsStore cluster | Total number of KV SDK read operations per second in the MemArtsStore cluster. This metric is used to monitor and analyze the load and processing latency of the StoreWorker node. | ≥ 0 | N/A | N/A | cluster_id,service_name | 5 minutes |
| worker_kv_sdk_read_latency_avg | MemArtsStore KV SDK read latency | MemArtsStore KV SDK read latency. This metric is used to monitor and analyze the load and processing latency of the KV SDK node. | ≥ 0 | µs | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| kms_latencyusage | Ranger RangerKMS request latency | Ranger RangerKMS request latency | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_cpuusage | Ranger PolicySync CPU usage | Ranger PolicySync CPU usage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_directmemory_calculate | Ranger PolicySync direct memory usage | Ranger PolicySync direct memory usage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_directmemory_max | Maximum direct memory set for Ranger PolicySync | Maximum direct memory set for Ranger PolicySync | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ps_directmemory_used | Direct memory used by Ranger PolicySync | Direct memory used by Ranger PolicySync | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ps_gc_old_time | Ranger PolicySync GC duration in the old area | Ranger PolicySync GC duration in the old area | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_gc_time_calculate | Ranger PolicySync GC duration | Ranger PolicySync GC duration | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_gc_young_time | Ranger PolicySync GC duration in the young area | GC duration in the young area for TagSync in Ranger PolicySync | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_heapmemory_calculate | Ranger PolicySync heap memory usage | Heap memory usage of the Ranger PolicySync process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_heapmemory_max | Maximum heap memory configured for Ranger PolicySync | Maximum heap memory available for the Ranger PolicySync process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ps_heapmemory_used | Heap memory used by Ranger PolicySync | Heap memory used by the Ranger PolicySync process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ps_memusage | Memory used by Ranger PolicySync | Memory used by Ranger PolicySync | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ps_nonheapmemory_calculate | Ranger PolicySync non-heap memory usage | Non-heap memory usage of the Ranger PolicySync process | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| ps_nonheapmemory_max | Maximum non-heap memory configured for Ranger PolicySync | Maximum non-heap memory available for the TagSync process in Ranger PolicySync | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ps_nonheapmemory_used | Non-heap memory used by Ranger PolicySync | Non-heap memory used by the Ranger PolicySync process | ≥ 0 | MiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| ra_latencyusage | Ranger RangerAdmin request latency | Ranger RangerAdmin request latency | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| jh_gcCount_G1_old_generation | JobHistory full GC times | JobHistory full GC times | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| js_gcCount_G1_old_generation | JDBCServer full GC times | JDBCServer full GC times | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| starrocks_be_async_delta_writer_queue_count | Queue length of StarRocks BE Async Delta Writer | Queue length of StarRocks BE Async Delta Writer | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_cpu | StarRocks CPU statistics | StarRocks CPU statistics, which are collected from the /proc/stat directory | ≥ 0 | jiffies | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_cpu_usage | StarRocks CN CPU usage | StarRocks CN CPU usage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_cpu_usage_allhost | StarRocks CPU usage | StarRocks CPU usage | 0.00-100.00% | % | N/A | cluster_id,service_name | 5 minutes |
| starrocks_cn_disks_avail_capacity | Remaining space of the disk where a specified StarRocks CN's data directory resides | Remaining space of the disk where a specified StarRocks CN's data directory resides | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_disks_state | Status of the disk where a specified StarRocks CN's data directory resides | Status of the disk where a specified StarRocks CN's data directory resides. 1: Normal 0: Abnormal | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_disks_total_capacity | Total capacity of the disk where a specified StarRocks CN's data directory resides | Total capacity of the disk where a specified StarRocks CN's data directory resides | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_engine_requests_failed_total | Failed tasks of each type on StarRocks CN | Failed tasks of each type on StarRocks CN | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_engine_requests_total | Total number of tasks of each type on StarRocks CN | Total number of tasks of each type on StarRocks CN | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_load_channel_count | Load channels enabled on StarRocks CN | Load channels enabled on StarRocks CN | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_max_mem_relationship_available_mem | Relationship between the maximum memory configured for StarRocks CN and the available memory on the host | Relationship between the maximum memory configured for StarRocks CN and the available memory on the host | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_mem | StarRocks CN memory usage | StarRocks CN memory usage | ≥ 0 | KiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_mem_allhost | Memory used by StarRocks | Memory used by StarRocks | ≥ 0 | Byte | 1024 | cluster_id,service_name | 5 minutes |
| starrocks_cn_mem_per | StarRocks CN memory usage, in percentage | StarRocks CN memory usage, in percentage | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_mem_per_allhost | StarRocks memory usage | StarRocks memory usage | 0.00-100.00% | % | N/A | cluster_id,service_name | 5 minutes |
| starrocks_cn_page_cache_capacity | Page Cache capacity of StarRocks CN | Page Cache capacity of StarRocks CN | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_page_cache_hit_count | Page Cache hit count of StarRocks CN | Page Cache hit count of StarRocks CN | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_page_cache_hit_ratio | Page Cache query hit ratio of StarRocks CN | Page Cache query hit ratio of StarRocks CN | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_page_cache_lookup_count | Page Cache query count of StarRocks CN | Page Cache query count of StarRocks CN | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_process_thread_num | Threads in the StarRocks CN process | Number of threads in the StarRocks CN process, which is collected through the /proc/pid/task directory | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_query_cache_capacity | Query Cache capacity of StarRocks CN | Query Cache capacity of StarRocks CN | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_query_cache_hit_count | Query Cache hit count of StarRocks CN | Query Cache hit count of StarRocks CN | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_query_cache_lookup_count | Query Cache lookup count of StarRocks CN | Query Cache lookup count of StarRocks CN | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_query_cache_usage | Memory usage of Query Cache on StarRocks CN | Memory usage of Query Cache on StarRocks CN | ≥ 0 | Byte | 1024 | cluster_id,node_id,role_name | 5 minutes |
| starrocks_cn_query_cache_usage_ratio | Usage ratio of Query Cache on StarRocks CN | Usage ratio of Query Cache on StarRocks CN | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| nm_availablegb | Available memory of NodeManager | Available memory of NodeManager | ≥ 0 | GiB | 1024 | cluster_id,node_id,role_name | 5 minutes |
| nm_containers_failed_rate | NodeManager container running failure rate | Container running failure rate on NodeManager | 0.00-100.00% | % | N/A | cluster_id,node_id,role_name | 5 minutes |
| nm_containerscompleted | Completed containers on NodeManager | Number of completed containers on NodeManager | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| nm_containersfailed | Failed containers on NodeManager | Number of failed containers on NodeManager | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| nm_containerskilled | Killed containers on NodeManager | Number of killed containers on NodeManager | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| nm_containerslaunched | Launched containers on NodeManager | Number of launched containers on NodeManager | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| nm_containersrunning | Running containers on NodeManager | Number of running containers on NodeManager | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| nm_log_aggregation_tasks_active | Active Tasks in the NodeManager log aggregation service | Number of active tasks in the NodeManager log aggregation service | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| nm_log_aggregation_tasks_completed | Completed Tasks in the NodeManager log aggregation service | Number of completed tasks in the NodeManager log aggregation service | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| rm_resourcepool_available_memory | Available memory of YARN ResourceManager | Available memory of YARN ResourceManager | ≥ 0 | MiB | 1024 | cluster_id,service_name,queue_name | 5 minutes |
| rm_resourcepool_available_memory_percent | Percentage of available memory in YARN ResourceManager | Percentage of available memory of YARN ResourceManager | 0.00-100.00% | % | N/A | cluster_id,service_name,queue_name | 5 minutes |
| rm_resourcepool_available_vcore | Available vCores in YARN ResourceManager | Available vCores in YARN ResourceManager | ≥ 0 | N/A | N/A | cluster_id,service_name,queue_name | 5 minutes |
| rm_resourcepool_available_vcore_percent | Percentage of available vCores in YARN ResourceManager | Percentage of available vCores in YARN ResourceManager | 0.00-100.00% | % | N/A | cluster_id,service_name,queue_name | 5 minutes |
| rm_rpcprocessingtimeavgtime_default | Average RPC processing time of the active ResourceManager | Average RPC processing time of the active ResourceManager | ≥ 0 | ms | N/A | cluster_id,service_name | 5 minutes |
| Metric ID | Metric Name | Metric Description | Value Range | Unit | Base | Dimension | Monitoring Interval |
|---|---|---|---|---|---|---|---|
| zk_avg_latency | Average ZooKeeper request processing latency | Average request processing latency of the ZooKeeper service. A high value indicates insufficient performance. | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| zk_fsync_threshold_exceed_count | Number of times the ZooKeeper fsync threshold is exceeded | Number of times the ZooKeeper fsync threshold is exceeded | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| zk_max_latency | Maximum ZooKeeper request processing latency | Maximum request processing latency of the ZooKeeper service. A high value indicates insufficient performance. | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| zk_maxCnxns | Maximum number of ZooKeeper connections | Total number of connections to the ZooKeeper service, which can be configured using the maxCnxns parameter on the ZooKeeper configuration page. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| zk_memory_used_allhost | Total memory used by ZooKeeper | Total memory used by ZooKeeper | ≥ 0 | MiB | 1024 | cluster_id,service_name | 5 minutes |
| zk_min_latency | Minimum ZooKeeper request processing latency | Minimum request processing latency of the ZooKeeper service. A large value indicates insufficient performance. | ≥ 0 | ms | N/A | cluster_id,node_id,role_name | 5 minutes |
| zk_outstanding_requests | Number of outstanding requests in the ZooKeeper service | Number of requests to be processed by the ZooKeeper service. A large value indicates a heavy load. | ≥ 0 | N/A | N/A | cluster_id,node_id,role_name | 5 minutes |
| zk_read_allhost | Total disk I/O read rate of ZooKeeper | Disk I/O rate of the ZooKeeper service | ≥ 0 | KiB/s | 1024 | cluster_id,service_name | 5 minutes |
If an object contains multiple dimension levels, you must specify those levels when using an API to query its metrics.
For example, to query the total number of RPC calls for the HDFS component in an MRS cluster (metric: nn_rpc_queuetime_numops_default), you need to specify the dimensions cluster_id,service_name, where cluster_id is at level 0 and service_name is at level 1.
- To query a single metric by calling an API, specify the dimensions as follows:
dim.0=cluster_id,cfa673a8-28b8-4509-b9f0-132bc5738aa9&dim.1=service_name,HDFS
cfa673a8-28b8-4509-b9f0-132bc5738aa9 and HDFS are the dimension values of cluster_id and service_name, respectively. For details about how to obtain the values, see Dimensions.
- To query multiple metrics by calling an API, specify the dimensions as follows:
"dimensions": [ { "name": "cluster_id", "value": "cfa673a8-28b8-4509-b9f0-132bc5738aa9" }, { "name": "service_name", "value": "HDFS" } ]cfa673a8-28b8-4509-b9f0-132bc5738aa9 and HDFS are the dimension values of cluster_id and service_name, respectively. For details about how to obtain the values, see Dimensions.
Dimensions
| Key | Value |
|---|---|
| cluster_id | MRS cluster ID. To obtain the value, log in to the MRS console, go to the MRS cluster details page, and view the cluster ID on the Dashboard page. |
| node_id | Node ID in the MRS cluster. To obtain the value, log in to the MRS console, go to the MRS cluster details page, and view the node name on the Nodes page. If the node name contains a period (.), replace it with an underscore (_). For example, if the obtained node name is node-master1FwAx.fzr.xxx.com, the corresponding value of node_id is node-master1FwAx_fzr_xxx_com. |
| service_name | Component name in the MRS cluster. To obtain the value, log in to the MRS console, go to the MRS cluster details page, and view the component name on the Components page. |
| role_name | Instance name of a component in the MRS cluster. To obtain the value, log in to the MRS console, go to the MRS cluster details page, click the component name on the Components tab page, and view the instance name on the Instances page. |
