Description
This alert is triggered when memory usage of the cluster exceeds the threshold.
Alert rule
For more information about how to add an alert rule, see Add an alert rule
Alert item | Metric type | Metric | Default threshold | Duration (consecutive cycles for triggering an alert) | Detection cycle | Alert level |
|---|---|---|---|---|---|---|
Cluster memory usage | Single metric | machine_mem_used_percent | 80 | 3 | 1 minute | Warning |
Impact on the system
High memory usage may cause the OS to throw an out-of-memory (OOM) error. This error may lead to a leader switchover in the OceanBase cluster and affect the services.
Insufficient memory may even cause connection creation to fail and increase the memory swapping frequency. Requests may fail due to frequent memory swapping.
Possible causes
Your memory resources are insufficient to cope with the increased memory usage arising from your rapid business growth.
The files generated through minor compactions are too large.
Solutions
Increase the memory resources of your OBServers when they become insufficient in your OceanBase cluster.
Generally, if the files generated through minor compactions are too large, Alibaba Cloud engineers will help resolve the problem. If any urgent help is needed, submit a ticket to contact Alibaba Cloud Technical Support.