Increased decision metrics are vital for monitoring dynamically altering environments and quickly altering software metrics. Examples the place excessive decision metrics are vital embrace excessive quantity e-commerce, stay streaming, autoscaling bursty workloads on Kubernetes clusters, and extra. Increased decision custom, Prometheus, and agent metrics at the moment are typically accessible, and will be written at a granularity of 10 seconds. Beforehand these metric varieties might solely be written as soon as each 60 seconds.
How one can write Monitoring agent metrics at 10-second decision
The Cloud Monitoring agent is a collectd-based daemon that collects system and software metrics from digital machine cases and sends them to Cloud Monitoring. The Monitoring agent collects disk, CPU, community, and course of metrics. By default, agent metrics are written at 60-second granularity. You may modify the agent collectd.conf configuration to ship metrics at 10-second granularity by changing the Interval value to ‘10’ within the Monitoring agent’s collectd.conf file.
After making this modification, you will want to restart your agent (this may increasingly differ based mostly in your working system and distro):
sudo service stackdriver-agent restart
Increased decision agent metrics require Monitoring agent model 6.0.1 or better. Yow will discover documentation for figuring out your agent model here.
Now that your Monitoring agent is emitting metrics at 10-second granularity, you may view them in Metrics Explorer by trying to find metrics with the prefix “agent.googleapis.com/agent/”.