Skip to main content

Troubleshooting

Start with the symptom, then move to the guide that owns that surface. Deployment pages cover bootstrap and host-level failures. Feature pages cover protocol-specific access failures. Operations pages cover health, recovery, and incident response.

Cluster startup and availability​

  • Cluster on Linux: systemd startup failures, storage connectivity, load balancer health checks, public certificate errors, and bootstrap SSO issues on VM clusters.
  • Cluster on Kubernetes: pod startup failures, storage connectivity, ingress health checks, public certificate errors, and bootstrap SSO issues on Kubernetes.
  • Monitoring: health checks, readiness checks, metrics, scrape targets, and alert signals.
  • High availability: failure modes for Auth, Proxy, caches, state backends, audit backends, and public ingress.

Identity and access incidents​

  • OIDC SSO: login failures, connector selection, redirect URI mismatches, and claim mapping problems.
  • Break-glass access: recovery when SSO or RBAC blocks normal operator login.
  • Revoking access: compromised users, over-broad roles, active sessions, locks, and incident evidence.

Resource access​

  • Server Access: SSH login, target registration, host certificate, and session recording failures.
  • Kubernetes Access: cluster registration, kubeconfig, RBAC, impersonation, and API access failures.
  • Application Access: browser TLS, DNS, upstream TLS, upstream network, and agent routing failures.
  • Database Access: database registration, identity-to-role mapping, TLS, and client connection failures.

Infrastructure dependencies​

  • Networking: load balancers, wildcard DNS, PROXY protocol, listener addresses, and advertise addresses.
  • Storage: state, audit, and recording backend selection and backend-specific troubleshooting links.
  • Backups and restore: restore failures and recovery drills for the state backend.