This topic describes the CloudMonitor metrics and system events in AnalyticDB for PostgreSQL.
Metrics
The following tables describe the metrics and recommended settings for AnalyticDB for PostgreSQL instances in elastic storage mode and reserved storage mode.
| Metric | Unit | Recommended cycle | Recommended threshold |
|---|---|---|---|
| adbpg_conn_count | N/A | 1Minute cycle
Continue for 5 periods |
Average >= 90%.
Assume that you have an instance that has 2 cores and 16 GB memory, and the maximum number of connections allowed for the instance is 550. You can set the metric to 495, which is obtained by using the following formula: 550 × 90%. For more information about the maximum number of user connections of instances, see Limits. |
| node_cpu_used_percent | % | 1Minute cycle
Continue for 5 periods |
Average >= 90%. |
| node_disk_iops | N/A | None | |
| node_disk_used_percent | % | 1Minute cycle
Continue for 5 periods |
Max. value >= 80%. |
| node_mem_used_percent | % | 1Minute cycle
Continue for 5 periods |
Average >= 90%. |
| Metric | Unit | Recommended cycle | Recommended threshold |
|---|---|---|---|
| ConnectionUsage | % | 1Minute cycle
Continue for 5 periods |
Average >= 90%. |
| CpuUsage | % | 1Minute cycle
Continue for 5 periods |
Average >= 90%. |
| DiskUsage | % | 1Minute cycle
Continue for 5 periods |
Average >= 80%. |
| IOPSUsage | % | 1Minute cycle
Continue for 5 periods |
Average >= 90%. |
| MemoryUsage | % | 1Minute cycle
Continue for 5 periods |
Average >= 90%. |
System events
The following table describes the system events of AnalyticDB for PostgreSQL instances.
| Type | Event | Description | Status | Level |
|---|---|---|---|---|
| Exception | ComputeGroupCpuHigh | The CPU usage of a compute group exceeds 90%. | Failed | Critical |
| Exception | ComputeGroupMemoryHigh | The memory usage of a compute group exceeds 85%. | Failed | Critical |
| Exception | HighCoumputeGroupDiskUsed | The disk usage of a compute group exceeds 80%. | Failed | Critical |
| Exception | InstanceDown | The instance is unavailable. | Failed | Critical |
| Exception | LongTransaction | The number of long-running transactions is greater than or equal to 5. | Failed | Critical |
| Exception | MasterCpuHigh | The CPU usage of the master node exceeds 90%. | Failed | Critical |
| Exception | MasterMemoryHigh | The memory usage of the master node exceeds 85%. | Failed | Critical |
| Exception | NonSuperConnection | The proportion of connections with an instance exceeds 80%. | Failed | Critical |