Harthelm Docs

Routing

Chart values: routing, trustedProxies.

Generated from charts/harthelm/values.yaml at v1.5.0

routing

ValueTypeDefaultDescription
routing.edgeauto, ingress, gatewayAPI, none"auto"auto: Gateway API when the HTTPRoute CRD exists and gatewayAPI.parentRefs is set, else Ingress. ingress | gatewayAPI | none (none renders no edge objects; route the two hosts yourself).
routing.ingress.classNamestring""
routing.ingress.annotationsobject{}
routing.gatewayAPI.parentRefslist of object[][{name: gateway, namespace: gateway-system, sectionName: https}]
routing.gatewayAPI.timeouts.requeststring""e.g. "0s" to disable; only where the implementation honours them
routing.gatewayAPI.timeouts.backendRequeststring""
routing.tls.consoleobjectSecret names default to <fullname>-console-tls and <fullname>-workspaces-tls. With certManager.enabled the chart creates the Certificates that fill them.
routing.tls.console.secretNamestring""
routing.tls.workspaces.secretNamestring""
routing.tls.certManager.enabledbooleanfalse
routing.tls.certManager.issuerRef.namestring""
routing.tls.certManager.issuerRef.kindstring"ClusterIssuer"
routing.tls.certManager.issuerRef.groupstring"cert-manager.io"

trustedProxies

Proxies in front of core and the router (the ingress controller's or the Gateway's pods, a load balancer), as CIDRs or addresses. Only from these peers do core and the router believe X-Forwarded-For, so per-address limits (30 CLI sign-ins per address in 10 minutes), audit rows and the sign-in approval page see the real client instead of the ingress pod. Empty: nobody is trusted and the peer is the client. Never 0.0.0.0/0. Workspace pods reach core's plain API port only while internal.tls is off; then do not list the pod range here.

ValueTypeDefaultDescription
trustedProxieslist of string[]CIDRs (or addresses) of the proxies in front of core and the router whose X-Forwarded-For counts.

On this page