- Docs
- Security
Security
- Security architecture The whole security picture on one page: the layers Syself locks down by default, the layers you own, and where the line sits.
- Threat model What each attacker position must achieve to win, which controls block and detect them, and which risks stay open, from internet scanners to supply-chain attacks.
- What Syself can and cannot do in your cluster The automation provisions and reconciles your cluster and touches none of your data or workloads, with concrete limits you can verify yourself.
- Zero trust on public networks Every node runs on the public internet with no private network, and this is how each connection stays authenticated and encrypted anyway.
- Node and OS security How every server running your workloads is protected: a sealed OS that cannot change at runtime, a default-deny firewall, key-only SSH, and a BSI C5 audit trail.
- Verify node integrity How to trust that a Syself Linux node runs exactly the OS image Syself built, how the platform verifies this at boot with dm-verity, and how to read tamper signals from the Kubernetes API.
- Respond to a tampered node What to do when a node raises NodeTampered, SealedOSTampered, or VerityCorruption, preserve evidence first, rotate exposed credentials, then reprovision the node.
- Manage access and tenancy How to give users and teams exactly the Kubernetes access they need, without handing out admin rights. Covers the hardened API server, least-privilege RBAC bound to OIDC, and per-cluster isolation.
- Multi-tenant isolation for agencies How to keep one client's cluster provably unable to reach another's, and how to isolate tenants inside a shared cluster when you must.
- Run a dedicated management cluster in your own account For regulated and high-assurance environments, run the management cluster in your own Hetzner account so no control plane, and no key, ever leaves your boundary.
- KubeGate: restrict API server access Limit which clients reach the Kubernetes API server through the load balancer with a KubeGate source-CIDR policy, and verify a blocked client is denied.
- Restrict SSH access Narrow SSH access on your Syself Autopilot nodes from open to every source on port 100 down to a single bastion host, or block it entirely using a Cilium deny policy.
- Ports and listeners Every port a Syself Linux node opens, the address it binds to, and which hosts can reach it. Covers control-plane, worker, shared, Cilium, and intentionally disabled ports.
- Segment with network policies How Cilium enforces pod-to-pod traffic policy by label identity, and how to set up a default-deny baseline per namespace using standard NetworkPolicy objects.
- Control pod egress (and the FQDN limitation) Restrict what your pods can call out to with Cilium egress policy, and understand that FQDN and DNS rules are rejected at apply time, so you use address, port, and identity rules instead.
- Encrypt pod traffic with WireGuard Enable WireGuard transparent encryption in Cilium so cross-node pod traffic is encrypted at the kernel level with no changes to your applications or workloads.
- Set security contexts How to set pod and container security contexts to limit what running containers can do, covering non-root users, privilege escalation, capabilities, and seccomp.
- Use AppArmor Every non-privileged container on Syself Linux runs under an enforcing AppArmor profile by default. What the profile blocks, how to change it per pod, where SELinux stands, and how it fits the full workload hardening stack.
- Enforce Pod Security Standards Roll out Pod Security Standards on a Kubernetes namespace without breaking running workloads: audit first, then enforce the baseline or restricted level.
- Control admission How the platform's admission protection prevents a broken webhook from locking your cluster, what it restricts tenant webhooks and Cilium network policies from doing, and how to write hard-block rules safely.
- Use Kyverno policies Install Kyverno on a Syself Autopilot workload cluster, configure failurePolicy correctly, and pick the right rule type for mutations, generation, validation, and image signing.
- OPA Gatekeeper as an alternative If your organization standardizes on OPA Gatekeeper and Rego, it fits the same webhook-guard contract as Kyverno, with these trade-offs.
- Supply-chain security How the Syself Linux node image is built from hash-pinned sources with a reproducible seal, what the SPDX SBOM and VEX verdicts let you verify, and exactly what is and is not signed or attested.
- Verify image signatures Require cryptographically signed container images in your Syself Autopilot workload clusters using Kyverno verifyImages or sigstore policy-controller, with safe rollout steps.
- Scan workload images for vulnerabilities Scan the images your workloads pull for known CVEs with Trivy or Grype, in CI and again inside the cluster, and gate on the results.
- Security evidence a release produces The machine-readable evidence every release produces, the SPDX SBOM with VEX verdicts, the stack security report, the CVE feed, the patch SLA ladder, and the audit evidence a running cluster emits.
- Manage application secrets How to handle Kubernetes Secrets safely: encrypt etcd first, keep secret material out of Git with Sealed Secrets or External Secrets, restrict RBAC, mount secrets as files, and rotate on custody change.
- Encrypt data disks (volume encryption) Encrypt the data disks under your local volumes with LUKS so a stolen or decommissioned disk gives up nothing, following the recommended design by hand.
- Data at rest and key custody What is encrypted at rest on Syself Autopilot and what is not: node disks, PersistentVolumes, and etcd. Who holds each key, from the etcd encryption key to the module-signing key and the verity root hash.
- Meet compliance requirements Shared-responsibility model, control-family mapping across BSI C5, CIS Kubernetes, SOC 2, and ISO 27001, and the actionable checklist for your side of the line.
- Run a CIS Kubernetes benchmark with kube-bench Run kube-bench against a cluster to produce CIS benchmark evidence, and read the results against what the platform already hardens.
- Auditd rules How the immutable auditd ruleset on Syself Linux records privileged activity, which compliance frameworks it supports evidence for, and how the audit log leaves the node.
- Retrieve audit logs How to read the host auditd log, the Kubernetes API audit log, and the KubeGate audit log from a Syself Linux node over SSH or kubectl, and how to ship them off-node into write-once storage for compliance.
- Ship audit logs to a SIEM A copy-paste Grafana Alloy DaemonSet that tails the host auditd, Kubernetes API, and KubeGate audit logs on every Syself Linux node and forwards them to Loki, a SIEM, or write-once object storage.
- Runtime threat detection Deploy Falco or an equivalent to catch what the preventive controls miss, then wire it into a detect-and-respond loop for a compromised workload.