Skip to content

UDS Security Overview

UDS maintains a defense-in-depth baseline, providing real security through the entire software delivery process:

  • Secure supply chain with CVE data and SBOMs for transparent software composition analysis and security audits.
  • Airgap ready with Zarf packages for predictable, offline deployments in disconnected environments.
  • Zero‑trust networking with default‑deny Kubernetes NetworkPolicy, Istio STRICT mTLS, and ALLOW‑based AuthorizationPolicy.
  • Identity & SSO via Keycloak and Authservice so apps can be protected consistently, whether they natively support authentication or not.
  • Admission control enforced by Pepr policies (non‑root, drop capabilities, block privileged/host access, etc.).
  • Runtime security with realtime detection and alerting on malicious behavior.
  • Observability & audit: centralized collection and shipping as well as metrics and dashboards.

What you get:

  • Per-release CVE scanning and SBOMs: View CVEs and full SBOMs for every Core release in the UDS Registry
  • Deterministic packaging: Packages include only what is needed for your environment, reducing drift and surprise dependencies

Why it matters:

  • You can verify exactly what ships with each release through SBOMs and CVE scans
  • Transparent software composition analysis helps you understand and mitigate security risks
  • Clear visibility into dependencies helps with compliance and security audits

References:

What you get:

  • Built with Zarf: Full support for offline/airgapped deployment through Zarf
  • No external dependencies: All tools operate seamlessly without internet access
  • Purpose-built for disconnected environments: Designed from the ground up for airgapped systems

Why it matters:

  • You can deploy and operate securely in offline environments without introducing “backdoor” network dependencies
  • Supports mission-critical use cases in highly secure networks where connectivity cannot be assumed
  • Unlike typical industry products that need adaptation for airgapped systems, UDS is built for airgap from the start

References:

What you get:

  • Keycloak for SSO management with opinionated defaults for realms, clients, and group-based access.
  • Authservice integration to protect apps that don’t natively speak OIDC—enforced at the mesh edge.

Why it matters:

  • Consistent login, token handling, and group mapping across apps.
  • Centralized access control that is easy to audit.

References:

What you get:

  • Default‑deny network posture: Per‑namespace NetworkPolicy isolates workloads by default with explicit allow rules generated from your package’s declared needs.
  • Istio STRICT mTLS: All in-mesh traffic is encrypted and identity-authenticated by default.
  • ALLOW‑based authorization: AuthorizationPolicy enforces least privilege at the service layer.
  • Explicit egress: Outbound access is explicitly declared to both in-cluster endpoints and remote hosts.
  • Admin vs Tenant ingress: Administrative UIs are isolated behind a dedicated admin gateway.

Why it matters:

  • Lateral movement is constrained by both the Kubernetes networking layer and Istio.
  • Connectivity to/from your application is explicitly declared and easily reviewable.

References:

What you get:

  • Secure defaults that block overly privileged workloads.
  • Security mutations to force workloads to use more secure configuration.
  • Controlled exemptions for edge cases, keeping RBAC tight and changes auditable.

Why it matters:

  • Misconfigurations and overly-permissive settings are blocked before they reach the cluster.
  • In some cases workloads are automatically “downgraded” to least privilege.
  • Exemptions are explicit and reviewable.

References:

What you get:

  • Runtime/container security from NeuVector, used in alerting/monitor mode by default.
  • Visibility into suspicious process, network, and file activity with alerts routed to your ops tooling.

Why it matters:

  • You get actionable detection without the risk of blocking production traffic.
  • Malicious behavior is instantly detected, allowing for quick triage and resolution.

What you get:

  • Centralized logging: Vector collects and ships all cluster and application logs for searchable audit trails.
  • Metrics & dashboards: Prometheus scrapes metrics and Grafana provides pre‑wired datasources and dashboards.

Why it matters:

  • Unified troubleshooting across logs and metrics.
  • Auditability for change control and incident response.

Last updated: