This topic describes the monitoring metrics for which alerts have not been configured in OceanBase Cloud Platform (OCP). You can use the custom alert feature of OCP to configure alerts for these monitoring metrics.
Background information
OCP provides many built-in alerts for the supported monitoring metrics. For more information, see Overview.
You can configure custom alerts for the monitoring metrics without built-in alerts. For more information, see Create an alert rule.
The following tables describe the monitoring metrics for which you can configure custom alerts.
Host metrics
Resource usage metrics of OCP-Agent processes
| Name | Metric |
|---|---|
| CPU utilization of the monagent process | monagent_process_cpu_percent |
| CPU utilization of the mgragent process | mgragent_process_cpu_percent |
Host performance metrics
| Name | Metric |
|---|---|
| Average system load in the last 5 minutes (based on the average number of CPU cores per host) | load5_per_cpu |
| Average number of I/O writes per second | io_write |
| Average number of I/O reads per second | io_read |
| Average time spent on each I/O write | io_write_time |
| Average time spent on each I/O read | io_read_time |
Tenant metrics
Tenant performance metrics
| Name | Metric | Description |
|---|---|---|
| Data size of the tenant disk | ob_tenant_data_size_gb | The data size refers to the actual data amount. |
| Required size of the tenant disk | ob_tenant_disk_used_size_gb | The required size refers to the disk size occupied after data alignment, including the disk space that has not been written but is already occupied. |
| Tenant disk usage | ob_tenant_disk_used_percentage | The percentage of the required disk size to the total disk size allocated to the tenant. |
| Tenant CPU utilization | ob_tenant_cpu_percent | The percentage of the number of CPU cores occupied by the tenant to the number of CPU cores allocated to the tenant. |
| Tenant memory usage | ob_memory_percent | The percentage of the memory space used by the tenant to the memory space allocated to the tenant. The v$sysstat view is queried with stat_id=140003 to get the value of the memory space used by the tenant, and queried with stat_id=140002 to get the value of the memory space allocated to the tenant. |
| Miss rate of the block cache | block_cache_miss_ratio | Miss rate = 100% - Hit rate |
| Miss rate of the Bloom filter cache | bloom_filter_cache_miss_ratio | Miss rate = 100% - Hit rate |
| Miss rate of the plan cache | plan_cache_miss_ratio | Miss rate = 100% - Hit rate |
| Miss rate of the row cache | row_cache_miss_ratio | Miss rate = 100% - Hit rate |
| Memory usage of the block cache | block_cache_size_mb | - |
| Memory usage of the Bloom filter cache | bloom_filter_cache_size_mb | - |
| Memory usage of the clog cache | clog_cache_size_mb | - |
| Memory usage of the location cache | location_cache_size_mb | - |
| Memory usage of the plan cache | plan_cache_size_mb | - |
| Memory usage of the row cache | row_cache_size_mb | - |
| Average time spent on each clog write | clog_write_disk_rt | - |
| Average time spent on each I/O read | io_read_rt | - |
| Average time spent on each I/O write | io_write_rt | - |
| Average time spent on the synchronization of each clog | trans_commit_log_sync_rt | - |
| Number of successful MemStore write locks | memstore_write_lock_fail_count | - |
| Average time spent on waiting for a MemStore write lock | memstore_write_lock_wait_time | - |
OceanBase Database metrics
| Name | Metric |
|---|---|
| Total number of sessions | all_session |
| Existence of block_file in the sstable directory on the data disk of OceanBase Database | ob_sstable_block_file_exists |
| Existence of the log directory of OceanBase Database | ob_log_dir_exists |
| Existence of the data directory of OceanBase Database | ob_data_dir_exists |
| Existence of the install directory of OceanBase Database | ob_install_dir_exists |
| Existence of the run directory of OceanBase Database | ob_run_dir_exists |
observer process metrics
The following table describes the metrics that allow you to view the monitoring data of the resources used by the observer processes in the operating system.
| Name | Metric |
|---|---|
| Number of threads for the observer process | observer_thread_count |
| Number of file descriptors opened by the observer process | observer_fd_count |
| Resident memory size of the observer process | observer_rss_mem_gb |
| CPU utilization of the observer process | observer_process_cpu_percent |
OBProxy metrics
| Name | Metric | Description |
|---|---|---|
| Client connections | current_session_client | |
| Server connections | current_session_server | |
| Percentage of client connections used by the obproxy process | client_connection_used_percent | The percentage of the number of client connections used by the obproxy process to the value of client_max_connections. |
| Percentage of server connections used by the obproxy process | server_connection_used_percent | The percentage of the number of server connections used by the obproxy process to the value of max_connections. |
| Memory usage of the obproxy process | obproxy_rss_mem_used_percent | The percentage of the resident memory size of the observer process to the value of proxy_mem_limited. |
obproxy process metrics
The following table describes the metrics that allow you to view the monitoring data of the resources used by the observer processes in the operating system.
| Name | Metric |
|---|---|
| Number of file descriptors for the obproxy process | obproxy_fd_count |
| Number of threads in the obproxy process | obproxy_thread_count |
| Resident memory size of the obproxy process | obproxy_rss_mem_gb |
| CPU utilization of the obproxy process | obproxy_process_cpu_percent |
