Capacity Planning

Infrastructure headroom. Tracking VPS-level CPU, memory, and disk alongside per-container metrics lets you predict when to scale before users feel it.

Signals

Signal	Source	What it tells you
VPS CPU	node_exporter	Host-level compute load
VPS Memory	node_exporter	Host-level memory pressure
VPS Disk	node_exporter	Storage headroom
Container CPU	dockerstats	Per-service compute load
Container Memory	dockerstats	Per-service memory usage

Rule of thumb

Keep peak utilization below 70% to preserve headroom for:

• Traffic spikes
• Rolling deployments (two versions briefly co-running)
• GC bursts and periodic batch jobs

At 90%+ utilization, any spike becomes an incident.

Config

Capacity signals are optional — they render null if the query key is absent from the active preset. Add them to any preset via:

queries:
  cap_vps_cpu:  'your_cpu_query'
  cap_vps_mem:  'your_mem_query'
  cap_vps_disk: 'your_disk_query'
  cap_container_cpu: 'your_container_cpu_query'
  cap_container_mem: 'your_container_mem_query'