Linux Server / Performance, Capacity & Containers
Use Linux pressure-stall metrics to identify resource contention
CPU, memory or I/O contention can harm latency before a simple utilization threshold triggers. Inspect available PSI metrics.
Environment and inputs: Linux server baseline · Ubuntu 24.04 LTS / RHEL 9-style systems. Use the specified distribution, init system and installed tool; service names, privileges and supported options can differ.
Start with this check
Confirm the host, shell and access role. Replace sample names, paths, domains and interfaces with approved targets. These are diagnostic commands; assess policy and workload impact before making changes.
cat /proc/pressure/cpu
cat /proc/pressure/memory
cat /proc/pressure/ioRead the result
On kernels exposing PSI, the files report pressure averages and cumulative stall time. Availability and exact interpretation depend on kernel and resource context.
Resolve and verify
Compare interval changes with service latency, cgroup limits and host activity. Address the resource where sustained pressure correlates with the incident.
Before changing production
Do not compare cumulative totals from hosts with different uptimes or assume missing PSI files prove no contention exists.