Infrastructure Monitoring
Once an agent is installed, Applicare automatically discovers and charts CPU, memory, disk, network, and process metrics — no manual configuration required. This guide covers what is collected and how to build views on top of it.
What is collected automatically#
- Host: CPU (per-core), memory, swap, load, disk I/O and capacity, network throughput and errors.
- Processes: top consumers by CPU/RAM, restarts, and open file descriptors.
- Containers: per-container resource use mapped back to the pod/service.
Metrics arrive at 10-second resolution by default and are visible on the host page within seconds of enrollment.
Building host groups#
Group hosts by tag to monitor fleets instead of individual machines. Tags can come from cloud metadata, the agent config, or Kubernetes labels.
applicare-agent config set tags="env:prod,team:payments,region:us-east-1"In the console, create a dynamic group with a filter like env:prod AND team:payments; membership updates as hosts come and go.
Alerting on infrastructure#
Attach thresholds or baselines to any metric. Applicare's baselines learn normal behaviour per host, which reduces noise from expected daily cycles. See Alert Configuration and let ArcIn correlate host anomalies with the services they affect.