Platform Observability

Deployment metrics

In Astro Private Cloud versions before 1.0, Deployment-level metrics such as CPU usage, memory usage, and network I/O were available by default through built-in node exporter and cAdvisor integrations. These were removed in APC 1.0 to reduce Prometheus data volume and improve performance.

Starting with APC 2.0, you can re-enable these metrics with two Helm flags. After you enable them, four additional panels appear in the Metrics tab for each Deployment in the Astro Private Cloud UI:

  • CPU Usage: Percentage of CPU consumed by the Deployment’s containers.
  • Memory Usage: Percentage of memory consumed by the Deployment’s containers.
  • Network Rx: Inbound network traffic received by the Deployment’s Pods.
  • Network Tx: Outbound network traffic transmitted by the Deployment’s Pods.

Deployment metrics panels showing CPU Usage, Memory Usage, Network Rx, and Network Tx

Both features are disabled by default.

Prerequisites

  • Astro Private Cloud 2.0 or later with Helm chart version 2.0.0 or later.
  • System Admin access to update the Astro Private Cloud Helm configuration.

Enable Deployment metrics

1

Update your Helm configuration

Add the following values to your values.yaml file:

1global:
2  nodeExporter:
3    enabled: true
4  cadvisor:
5    enabled: true
2

Apply the configuration

Run a Helm upgrade to apply the changes:

$helm upgrade -f values.yaml -n <your-namespace> astronomer astronomer/astronomer
3

Verify the metrics

Open the Astro Private Cloud UI, navigate to a Deployment, and click the Metrics tab. The CPU Usage, Memory Usage, Network Rx, and Network Tx panels appear alongside the existing Database Connections and Waiting Clients panels.