Infrastructure overview

Infrastructure telemetry shows whether host resources contributed to an error, incident, or outage. It is most useful when viewed alongside errors, logs, deploys, uptime checks, and performance data.

What infrastructure adds

The Infrastructure page organizes data from errorgap-agent into hosts, volumes, processes, clusters, and environments. Use it to spot saturation, capacity pressure, and local process behavior that may not appear in application traces.

Filter the dashboard

Start with the time range that covers the failure, then narrow by cluster and environment. Filters help separate one noisy host group from the rest of the project.

Hosts

Host views summarize CPU, memory, load, uptime, and recent trend signals. Open a host when it has unusual load, memory pressure, or a pattern that matches an incident timeline.

• Compare load and CPU before and after a deploy or incident start.
• Look for hosts with much higher resource use than peers in the same cluster.
• Use uptime to identify recent restarts or replacements.

Volumes

Volume views help identify disk pressure. Open a volume to inspect utilization and, when available, directory-level usage.

• Watch for disks approaching full capacity.
• Compare growth across the selected time range.
• Use directory details to identify logs, caches, or uploads consuming space.

Processes

Process views show which commands are consuming resources on a host. Use process detail when a worker, web process, database client, or sidecar appears to be driving saturation.

Investigate saturation

1. Open the incident or error group and note the affected time window.
2. Open Infrastructure with the same time range and environment.
3. Check hosts for CPU, memory, load, and restart anomalies.
4. Check volumes for disk pressure and directories growing unexpectedly.
5. Check processes for outliers that match the incident timeline.
6. Link the infrastructure finding back to the incident summary or issue tracker.