cmd/{k8s-operator,containerboot},kube: ensure egress ProxyGroup proxies don't terminate while cluster traffic is still routed to them (#14436)

cmd/{containerboot,k8s-operator},kube: add preshutdown hook for egress PG proxies

This change is part of work towards minimizing downtime during update
rollouts of egress ProxyGroup replicas.
This change:
- updates the containerboot health check logic to return Pod IP in headers,
if set
- always runs the health check for egress PG proxies
- updates ClusterIP Services created for PG egress endpoints to include
the health check endpoint
- implements preshutdown endpoint in proxies. The preshutdown endpoint
logic waits till, for all currently configured egress services, the ClusterIP
Service health check endpoint is no longer returned by the shutting-down Pod
(by looking at the new Pod IP header).
- ensures that kubelet is configured to call the preshutdown endpoint

This reduces the possibility that, as replicas are terminated during an update,
a replica gets terminated to which cluster traffic is still being routed via
the ClusterIP Service because kube proxy has not yet updated routig rules.
This is not a perfect check as in practice, it only checks that the kube
proxy on the node on which the proxy runs has updated rules. However, overall
this might be good enough.

The preshutdown logic is disabled if users have configured a custom health check
port via TS_LOCAL_ADDR_PORT env var. This change throws a warnign if so and in
future setting of that env var for operator proxies might be disallowed (as users
shouldn't need to configure this for a Pod directly).
This is backwards compatible with earlier proxy versions.

Updates tailscale/tailscale#14326


Signed-off-by: Irbe Krumina <irbe@tailscale.com>

cmd/containerboot/settings.go

@@ -64,16 +64,16 @@ type settings struct {
 	// when setting up rules to proxy cluster traffic to cluster ingress
 	// target.
 	// Deprecated: use PodIPv4, PodIPv6 instead to support dual stack clusters
-	PodIP               string
-	PodIPv4             string
-	PodIPv6             string
-	PodUID              string
-	HealthCheckAddrPort string
-	LocalAddrPort       string
-	MetricsEnabled      bool
-	HealthCheckEnabled  bool
-	DebugAddrPort       string
-	EgressSvcsCfgPath   string
+	PodIP                string
+	PodIPv4              string
+	PodIPv6              string
+	PodUID               string
+	HealthCheckAddrPort  string
+	LocalAddrPort        string
+	MetricsEnabled       bool
+	HealthCheckEnabled   bool
+	DebugAddrPort        string
+	EgressProxiesCfgPath string
 }
 
 func configFromEnv() (*settings, error) {
@@ -107,7 +107,7 @@ func configFromEnv() (*settings, error) {
 		MetricsEnabled:                        defaultBool("TS_ENABLE_METRICS", false),
 		HealthCheckEnabled:                    defaultBool("TS_ENABLE_HEALTH_CHECK", false),
 		DebugAddrPort:                         defaultEnv("TS_DEBUG_ADDR_PORT", ""),
-		EgressSvcsCfgPath:                     defaultEnv("TS_EGRESS_SERVICES_CONFIG_PATH", ""),
+		EgressProxiesCfgPath:                  defaultEnv("TS_EGRESS_PROXIES_CONFIG_PATH", ""),
 		PodUID:                                defaultEnv("POD_UID", ""),
 	}
 	podIPs, ok := os.LookupEnv("POD_IPS")
@@ -186,7 +186,7 @@ func (s *settings) validate() error {
 			return fmt.Errorf("error parsing TS_HEALTHCHECK_ADDR_PORT value %q: %w", s.HealthCheckAddrPort, err)
 		}
 	}
-	if s.localMetricsEnabled() || s.localHealthEnabled() {
+	if s.localMetricsEnabled() || s.localHealthEnabled() || s.EgressProxiesCfgPath != "" {
 		if _, err := netip.ParseAddrPort(s.LocalAddrPort); err != nil {
 			return fmt.Errorf("error parsing TS_LOCAL_ADDR_PORT value %q: %w", s.LocalAddrPort, err)
 		}
@@ -199,8 +199,8 @@ func (s *settings) validate() error {
 	if s.HealthCheckEnabled && s.HealthCheckAddrPort != "" {
 		return errors.New("TS_HEALTHCHECK_ADDR_PORT is deprecated and will be removed in 1.82.0, use TS_ENABLE_HEALTH_CHECK and optionally TS_LOCAL_ADDR_PORT")
 	}
-	if s.EgressSvcsCfgPath != "" && !(s.InKubernetes && s.KubeSecret != "") {
-		return errors.New("TS_EGRESS_SERVICES_CONFIG_PATH is only supported for Tailscale running on Kubernetes")
+	if s.EgressProxiesCfgPath != "" && !(s.InKubernetes && s.KubeSecret != "") {
+		return errors.New("TS_EGRESS_PROXIES_CONFIG_PATH is only supported for Tailscale running on Kubernetes")
 	}
 	return nil
 }
@@ -291,7 +291,7 @@ func isOneStepConfig(cfg *settings) bool {
 // as an L3 proxy, proxying to an endpoint provided via one of the config env
 // vars.
 func isL3Proxy(cfg *settings) bool {
-	return cfg.ProxyTargetIP != "" || cfg.ProxyTargetDNSName != "" || cfg.TailnetTargetIP != "" || cfg.TailnetTargetFQDN != "" || cfg.AllowProxyingClusterTrafficViaIngress || cfg.EgressSvcsCfgPath != ""
+	return cfg.ProxyTargetIP != "" || cfg.ProxyTargetDNSName != "" || cfg.TailnetTargetIP != "" || cfg.TailnetTargetFQDN != "" || cfg.AllowProxyingClusterTrafficViaIngress || cfg.EgressProxiesCfgPath != ""
 }
 
 // hasKubeStateStore returns true if the state must be stored in a Kubernetes
@@ -308,6 +308,10 @@ func (cfg *settings) localHealthEnabled() bool {
 	return cfg.LocalAddrPort != "" && cfg.HealthCheckEnabled
 }
 
+func (cfg *settings) egressSvcsTerminateEPEnabled() bool {
+	return cfg.LocalAddrPort != "" && cfg.EgressProxiesCfgPath != ""
+}
+
 // defaultEnv returns the value of the given envvar name, or defVal if
 // unset.
 func defaultEnv(name, defVal string) string {