Skip to main content
Platform monitoring gives admins and security teams visibility into who did what and how the cluster is performing. Monitoring in Domino covers two primary areas: audit logs for user and administrative events, and system telemetry for cluster performance and workload execution.

​
How it works

  • Audit logs: Domino records platform events to provide visibility into user activity and resource modifications. Audit logs let security teams conduct forensic investigations and verify compliance controls.
  • System and workload telemetry: Monitoring cluster health and resource consumption ensures platform availability and helps prevent resource starvation during high-compute workloads.

​
Audit logs

Domino automatically captures the following event types in the audit record:
  • Authentication and sessions: user logins, logouts, password changes, failed login attempts, and API key regenerations.
  • Access control: changes to user roles, organization memberships, project permissions, and administrative privilege assignments.
  • Secrets and credentials: creation, modification, or deletion of user environment variables, project environment variables, and Git credentials.
  • Data sources: creation, update, or deletion of data source configurations and data access grants.
  • Execution management: launch, termination, and configuration changes for workspaces, jobs, apps, and Model APIs.
Platform administrators can review audit events directly within the Domino application. To view audit records:
  1. Log in to Domino as a platform administrator.
  2. Go to Admin > Audit Logs.
  3. Filter events by date range, user, event type, or target resource.
For enterprise log retention and automated threat detection, export Domino audit logs to your central Security Information and Event Management (SIEM) provider. Domino runs on Kubernetes and outputs structured log records to standard output streams. Configure your log collector running in your Kubernetes cluster to ship logs from the Domino control plane namespace to your external SIEM endpoint.

​
System and workload telemetry

Domino exposes system metrics in Prometheus format. You can integrate these metrics with your existing monitoring stack or use Domino’s built-in Grafana dashboards. Key monitoring metrics include:
  • Node resource utilization: CPU, memory, and disk usage across control plane and execution nodes.
  • GPU utilization: memory consumption and compute load on GPU-enabled execution nodes.
  • Execution pod state: status, restart counts, and resource limits for workspace and job containers.
  • Storage consumption: persistent volume usage for datasets, project files, and system caches.
Every workspace, job, app, and Model API run creates dedicated execution logs that capture standard output (stdout) and standard error (stderr). To view execution logs:
  1. Open your project.
  2. Select Jobs, Workspaces, or Model APIs.
  3. Click the specific execution run.
  4. Select the Logs tab to view or download the runtime output.

​
Compliance best practices

When configuring platform monitoring for compliance requirements, follow these guidelines:
  • Restrict audit log access: limit access to the Admin > Audit Logs interface to authorized security and compliance personnel.
  • Enforce external log immutability: export audit records to an external, write-once-read-many (WORM) storage target to prevent tampering.
  • Set up threshold alerts: configure alerts in your SIEM or Prometheus monitoring stack for high-risk events, such as repeated authentication failures, privilege escalations, or unusual spikes in data download activity.
  • Monitor API key rotation: regularly review audit logs for API key regeneration events to verify that team members adhere to your organization’s credential rotation schedules.
Last modified on August 5, 2026

Related topics

Monitoring checksModel Monitoring tagsUse monitoringUse the Model Monitoring APIs