This is a server for providing a Virtual IP and load-balancer for the Kubernetes control-plane Usage: kube-vip [command] Available Commands: completion Generate the autocompletion script for the specified shell help Help about any command kubeadm Kubeadm functions manager Start the kube-vip manager manifest Manifest functions sample Generate a Sample configuration service Start the Virtual IP / Load balancer as a service within a Kubernetes cluster version Version and Release information about the Kubernetes Virtual IP Server Flags: --address string an address (IP or DNS name) to use as a VIP --allowInterfaceNotUp Allow kube-vip to start even if the interface is not up --annotations string Set Node annotations prefix for parsing --arp Enable Arp for VIP changes --autodetectcp Determine working address for control plane (from loopback) --bgp This will enable BGP support within kube-vip --bgpAttachIPToInterface Assign BGP service VIPs to the configured interface --bgpHoldTimer uint The hold timer for all bgp peers (it defines the time a session is held) (default 30) --bgpKeepAliveInterval uint The keepalive interval for all bgp peers (it defines the heartbeat of keepalive messages) (default 10) --bgpRouterID string The routerID for the bgp server --bgppeers strings Comma separated BGP Peer, format: address:as:password:multihop --cleanRoutingTable Clean routing table of redundant routes on start --config-file string Path to a JSON/YAML configuration file to load settings from --controlPlaneHealthCheckAddress string URL to poll for the control-plane health check when using BGP without leader election --controlPlaneHealthCheckCAPath string Path to CA certificate for TLS verification when the control-plane health check URL is HTTPS --controlPlaneHealthCheckFailureThreshold int Consecutive control-plane health check failures before withdrawing the BGP route (default 3) --controlPlaneHealthCheckPeriodSeconds int Seconds between control-plane health checks (default 5) --controlPlaneHealthCheckTimeoutSeconds int Timeout for each control-plane health check request (default 3) --controlplane Enable HA for control plane --ddns use Dynamic DNS + DHCP to allocate VIP for address --debounceTime string Configures the time that the event debouncer will wait for the events arrival (default 0s - debouncer disabled, enable with min. 200ms) (default "0s") --dhcpBackoffAttempts uint number of times DHCP client will try to obtain an IP address (defaults to: 3, 0 for unlimited retries) (default 3) --dhcpMode string Mode DHCP resolving will use to obtain IP addresses (ipv4, ipv6, dual) (default "ipv4") --disableServiceUpdates If true, kube-vip will process services as usual, but will not update service's Status.LoadBalancer.Ingress slice --dnsMode string Name of the mode that DNS lookup will be performed (first, ipv4, ipv6, dual) (default "first") --egressWithNftables Use nftables-based egress implementation (default true) --enableEndpoints If enabled, kube-vip will only advertise services, but will use the (deprecated since v1.33) endpoints for IP addresses --enableLoadBalancer enable loadbalancing on the VIP with IPVS --enableNodeLabeling Enable leader node labeling with "kube-vip.io/has-ip", defaults to false --enablePprof Expose the pprof profiling endpoints (debug only, they are unauthenticated) --etcdCACert string Verify certificates of TLS-enabled secure servers using this CA bundle file --etcdCert string Identify secure client using this TLS certificate file --etcdEndpoints strings Etcd member endpoints --etcdKey string Identify secure client using this TLS key file --healthCheckPort int If set to non-zero (> 1024), then this is the port that the healthcheck will listen on -h, --help help for kube-vip --instanceName string Unique name for this kube-vip instance (currently used to isolate nftables egress tables) --interface string Name of the interface to bind to --k8sConfigPath string Path to the configuration file used with the Kubernetes client (default "/etc/kubernetes/admin.conf") --lbClassName string Name of load balancer class for kube-VIP, defaults to "kube-vip.io/kube-vip-class" (default "kube-vip.io/kube-vip-class") --lbClassNameLegacyHandling Use legacy LoadBalancer class name handling (e.g. accepting services both with empty and non-empty class) (default true) --lbClassOnly Enable load balancing only for services with LoadBalancerClass "kube-vip.io/kube-vip-class" --lbForwardingMethod string loadbalancer forwarding method (default "local") --lbPort uint16 loadbalancer port for the VIP (default 6443) --leaderElection Use the Kubernetes leader election mechanism for clustering --leaderElectionType string Defines the backend to run the leader election: kubernetes or etcd. Defaults to kubernetes. (default "kubernetes") --leaseDuration int Length of time (in seconds) a Kubernetes leader lease can be held for (default 15) --leaseName string Name of the lease that is used for leader election (default "plndr-cp-lock") --leaseRenewDuration int Length of time (in seconds) a Kubernetes leader can attempt to renew its lease (default 10) --leaseRetry int Length of time (in seconds) the LeaderElector clients should wait between tries of actions (default 2) --loInterfaceGlobalScope If true, kube-vip will set global scope when using the lo interface, otherwise a host scope will be used by default --localAS uint32 The local AS number for the bgp server (default 65000) --log int32 Set the level of logging --loseLeadership Lose leadership when VIP interface goes down --loseLeadershiptTimeoutSeconds int Timeout before re-electing a leader when the VIP interface is down (default 30) --mirrorDestInterface string network interface where all traffic that traverses the service interface will be mirrored to. Source interface will use default interface is servicesInterface is not set. --multihop This will enable BGP multihop support -n, --namespace string The namespace for the configmap defined within the cluster (default "kube-system") --nodeName string Name to be used for lease holder. Must be unique for each node/instance --onlyAllowTrafficServicePorts Only allow traffic to service ports, others will be dropped, defaults to false --peerAS uint32 The AS number for a BGP peer (default 65000) --peerAddress string The address of a BGP peer --peerPass string The md5 password for a BGP peer --perServiceElectionOnDemand Allow kube-vip to use per-service election for annotated services --port uint16 Port for the VIP (default 6443) --pprofHTTPServer string Host and port used to expose the pprof endpoints, only used with --enablePprof (default "127.0.0.1:6060") --preserveVipOnLeadershipLoss Preserve ARP VIP addresses on interface when leadership is lost (default: false for backward compatibility) --prometheusHTTPServer string Host and port used to expose Prometheus metrics via an HTTP server (default ":2112") --routingProtocol int The routing protocol value used to create routes (default 248) --serviceInterface string Name of the interface to bind to (for services) --services Enable Kubernetes services --servicesElection Enable leader election per kubernetes service --servicesLeaseName string Name of the lease that is used for leader election for services (in arp mode) (default "plndr-svcs-lock") --sourceIF string The source interface for bgp peering (not to be used with sourceIP) --sourceIP string The source address for bgp peering (not to be used with sourceIF) --table Enable Routing Table for services VIPs --tableID int The routing table used for all table entries (default 198) --tableType int The type of route that will be added to the routing table (default 1) --vip string The Virtual IP address --vipSubnet string The Virtual IP address subnet e.g. /32 /24 /8 etc.. (Default to 32 for IPv4 and 128 for IPv6) --wireguard Enable Wireguard for services VIPs --zebra This will enable Zebra support within kube-vip --zebraSoftwareName string Software Name for Zebra (default "frr8.3") --zebraUrl string Path to the unix domain socket for connecting to Zebra daemon (default "unix:/var/run/frr/zserv.api") --zebraVersion uint32 Zebra API Version (default 6) Use "kube-vip [command] --help" for more information about a command.