Description
This alert is triggered when the memory usage of a tenant exceeds the threshold.
Alert rule
Alert item | Metric type | Metric | Default threshold | Duration (consecutive cycles for triggering an alert) | Detection cycle | Alert level |
|---|---|---|---|---|---|---|
Tenant memory usage | Single metric | memory_usage | 85 | 3 | 1 minute | Warning |
Impact on the system
If the tenant memory is insufficient, SQL statement execution may be slowed down, and writes may fail.
Possible causes
The number of write requests in the business tenant abruptly increased.
The minor compaction feature of the cluster is not functioning.
Solutions
Check the monitoring records for write requests in the tenant. If the number of write requests abruptly increased during the alert period, adjust the memory resources of the tenant based on business scenarios. If the memory resources within the cluster are fully allocated, consider scaling out the cluster.
If the abrupt increase in memory usage was caused by the malfunctioning of the minor compaction feature of the cluster, contact Alibaba Cloud Technical Support.