Which CSS Metrics Should I Focus On?
The metrics that you need to focus on include the disk usage and cluster health status. You can log in to Cloud Eye and configure alarm prompts according to actual conditions. If alarms are reported, clear them by taking related measures. For details about how to configure alarms, see Creating Alarm Rules in the Cloud Search Service User Guide.
Configuration examples:
- Alarms are reported if the disk usage is higher than or equal to a specified value (for example, 85%) and has reached this value multiple times (for example, 5 times) within a specified time period (for example, 5 minutes).
- Alarms are reported if the value of the cluster health status metric exceeds 0 for multiple times (for example, 5 times) within a specified time period (for example, 5 minutes).
Measures:
- Upon receiving alarms related to the disk usage, view disk space consumption, check whether data can be deleted from cluster nodes or archived to other systems to release space, or expand the disk capacity.
- If an alarm related to the cluster health status is received, check whether shard allocation is normal, whether shards are lost, and check whether the process is restarted on Cerebro.
Parent topic: General Consulting
Did this article solve your problem?
Thank you for your score！Your feedback would help us improve the website.