mirror of https://github.com/hashicorp/consul
1236 lines
40 KiB
Go
1236 lines
40 KiB
Go
// Copyright (c) HashiCorp, Inc.
|
|
// SPDX-License-Identifier: BUSL-1.1
|
|
|
|
package xds
|
|
|
|
import (
|
|
"errors"
|
|
"fmt"
|
|
"net"
|
|
"sort"
|
|
"strings"
|
|
"time"
|
|
|
|
envoy_core_v3 "github.com/envoyproxy/go-control-plane/envoy/config/core/v3"
|
|
envoy_route_v3 "github.com/envoyproxy/go-control-plane/envoy/config/route/v3"
|
|
envoy_matcher_v3 "github.com/envoyproxy/go-control-plane/envoy/type/matcher/v3"
|
|
"golang.org/x/exp/maps"
|
|
"golang.org/x/exp/slices"
|
|
|
|
"github.com/hashicorp/consul/agent/connect"
|
|
"github.com/hashicorp/consul/agent/consul/discoverychain"
|
|
"github.com/hashicorp/consul/agent/proxycfg"
|
|
"github.com/hashicorp/consul/agent/structs"
|
|
"github.com/hashicorp/consul/agent/xds/config"
|
|
"github.com/hashicorp/consul/agent/xds/response"
|
|
"google.golang.org/protobuf/proto"
|
|
"google.golang.org/protobuf/types/known/durationpb"
|
|
"google.golang.org/protobuf/types/known/wrapperspb"
|
|
)
|
|
|
|
// routesFromSnapshot returns the xDS API representation of the "routes" in the
|
|
// snapshot.
|
|
func (s *ResourceGenerator) routesFromSnapshot(cfgSnap *proxycfg.ConfigSnapshot) ([]proto.Message, error) {
|
|
if cfgSnap == nil {
|
|
return nil, errors.New("nil config given")
|
|
}
|
|
|
|
switch cfgSnap.Kind {
|
|
case structs.ServiceKindConnectProxy:
|
|
return s.routesForConnectProxy(cfgSnap)
|
|
case structs.ServiceKindIngressGateway:
|
|
return s.routesForIngressGateway(cfgSnap)
|
|
case structs.ServiceKindAPIGateway:
|
|
return s.routesForAPIGateway(cfgSnap)
|
|
case structs.ServiceKindTerminatingGateway:
|
|
return s.routesForTerminatingGateway(cfgSnap)
|
|
case structs.ServiceKindMeshGateway:
|
|
return s.routesForMeshGateway(cfgSnap)
|
|
default:
|
|
return nil, fmt.Errorf("Invalid service kind: %v", cfgSnap.Kind)
|
|
}
|
|
}
|
|
|
|
// routesFromSnapshotConnectProxy returns the xDS API representation of the
|
|
// "routes" in the snapshot.
|
|
func (s *ResourceGenerator) routesForConnectProxy(cfgSnap *proxycfg.ConfigSnapshot) ([]proto.Message, error) {
|
|
var resources []proto.Message
|
|
for uid, chain := range cfgSnap.ConnectProxy.DiscoveryChain {
|
|
if chain.Default {
|
|
continue
|
|
}
|
|
|
|
if !structs.IsProtocolHTTPLike(chain.Protocol) {
|
|
// Routes can only be defined for HTTP services
|
|
continue
|
|
}
|
|
|
|
virtualHost, err := s.makeUpstreamRouteForDiscoveryChain(cfgSnap, uid, chain, []string{"*"}, false, perRouteFilterBuilder{})
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if virtualHost == nil {
|
|
continue
|
|
}
|
|
|
|
route := &envoy_route_v3.RouteConfiguration{
|
|
Name: uid.EnvoyID(),
|
|
VirtualHosts: []*envoy_route_v3.VirtualHost{virtualHost},
|
|
// ValidateClusters defaults to true when defined statically and false
|
|
// when done via RDS. Re-set the reasonable value of true to prevent
|
|
// null-routing traffic.
|
|
ValidateClusters: response.MakeBoolValue(true),
|
|
}
|
|
resources = append(resources, route)
|
|
}
|
|
addressesMap := make(map[string]map[string]string)
|
|
err := cfgSnap.ConnectProxy.DestinationsUpstream.ForEachKeyE(func(uid proxycfg.UpstreamID) error {
|
|
svcConfig, ok := cfgSnap.ConnectProxy.DestinationsUpstream.Get(uid)
|
|
if !ok || svcConfig == nil {
|
|
return nil
|
|
}
|
|
if !structs.IsProtocolHTTPLike(svcConfig.Protocol) {
|
|
// Routes can only be defined for HTTP services
|
|
return nil
|
|
}
|
|
|
|
for _, address := range svcConfig.Destination.Addresses {
|
|
|
|
routeName := clusterNameForDestination(cfgSnap, "~http", fmt.Sprintf("%d", svcConfig.Destination.Port), svcConfig.NamespaceOrDefault(), svcConfig.PartitionOrDefault())
|
|
if _, ok := addressesMap[routeName]; !ok {
|
|
addressesMap[routeName] = make(map[string]string)
|
|
}
|
|
// cluster name is unique per address/port so we should not be doing any override here
|
|
|
|
clusterName := clusterNameForDestination(cfgSnap, svcConfig.Name, address, svcConfig.NamespaceOrDefault(), svcConfig.PartitionOrDefault())
|
|
addressesMap[routeName][clusterName] = address
|
|
}
|
|
return nil
|
|
})
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
for routeName, clusters := range addressesMap {
|
|
routes, err := s.makeRoutesForAddresses(routeName, clusters)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if routes != nil {
|
|
resources = append(resources, routes...)
|
|
}
|
|
}
|
|
|
|
// TODO(rb): make sure we don't generate an empty result
|
|
return resources, nil
|
|
}
|
|
|
|
func (s *ResourceGenerator) makeRoutesForAddresses(routeName string, addresses map[string]string) ([]proto.Message, error) {
|
|
var resources []proto.Message
|
|
|
|
route, err := makeNamedAddressesRoute(routeName, addresses)
|
|
if err != nil {
|
|
s.Logger.Error("failed to make route", "cluster", "error", err)
|
|
return nil, err
|
|
}
|
|
resources = append(resources, route)
|
|
|
|
return resources, nil
|
|
}
|
|
|
|
// routesFromSnapshotTerminatingGateway returns the xDS API representation of the "routes" in the snapshot.
|
|
// For any HTTP service we will return a default route.
|
|
func (s *ResourceGenerator) routesForTerminatingGateway(cfgSnap *proxycfg.ConfigSnapshot) ([]proto.Message, error) {
|
|
if cfgSnap == nil {
|
|
return nil, errors.New("nil config given")
|
|
}
|
|
|
|
var resources []proto.Message
|
|
for _, svc := range cfgSnap.TerminatingGateway.ValidServices() {
|
|
clusterName := connect.ServiceSNI(svc.Name, "", svc.NamespaceOrDefault(), svc.PartitionOrDefault(), cfgSnap.Datacenter, cfgSnap.Roots.TrustDomain)
|
|
cfg, err := config.ParseProxyConfig(cfgSnap.TerminatingGateway.ServiceConfigs[svc].ProxyConfig)
|
|
if err != nil {
|
|
// Don't hard fail on a config typo, just warn. The parse func returns
|
|
// default config if there is an error so it's safe to continue.
|
|
s.Logger.Warn(
|
|
"failed to parse Proxy.Config",
|
|
"service", svc.String(),
|
|
"error", err,
|
|
)
|
|
}
|
|
if !structs.IsProtocolHTTPLike(cfg.Protocol) {
|
|
// Routes can only be defined for HTTP services
|
|
continue
|
|
}
|
|
routes, err := s.makeRoutes(cfgSnap, svc, clusterName, true)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if routes != nil {
|
|
resources = append(resources, routes...)
|
|
}
|
|
}
|
|
|
|
for _, svc := range cfgSnap.TerminatingGateway.ValidDestinations() {
|
|
svcConfig := cfgSnap.TerminatingGateway.ServiceConfigs[svc]
|
|
|
|
for _, address := range svcConfig.Destination.Addresses {
|
|
clusterName := clusterNameForDestination(cfgSnap, svc.Name, address, svc.NamespaceOrDefault(), svc.PartitionOrDefault())
|
|
cfg, err := config.ParseProxyConfig(cfgSnap.TerminatingGateway.ServiceConfigs[svc].ProxyConfig)
|
|
if err != nil {
|
|
// Don't hard fail on a config typo, just warn. The parse func returns
|
|
// default config if there is an error so it's safe to continue.
|
|
s.Logger.Warn(
|
|
"failed to parse Proxy.Config",
|
|
"service", svc.String(),
|
|
"error", err,
|
|
)
|
|
}
|
|
if !structs.IsProtocolHTTPLike(cfg.Protocol) {
|
|
// Routes can only be defined for HTTP services
|
|
continue
|
|
}
|
|
routes, err := s.makeRoutes(cfgSnap, svc, clusterName, false)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if routes != nil {
|
|
resources = append(resources, routes...)
|
|
}
|
|
}
|
|
}
|
|
|
|
return resources, nil
|
|
}
|
|
|
|
func (s *ResourceGenerator) makeRoutes(
|
|
cfgSnap *proxycfg.ConfigSnapshot,
|
|
svc structs.ServiceName,
|
|
clusterName string,
|
|
autoHostRewrite bool,
|
|
) ([]proto.Message, error) {
|
|
resolver, hasResolver := cfgSnap.TerminatingGateway.ServiceResolvers[svc]
|
|
|
|
if !hasResolver {
|
|
// Use a zero value resolver with no timeout and no subsets
|
|
resolver = &structs.ServiceResolverConfigEntry{}
|
|
}
|
|
|
|
var resources []proto.Message
|
|
var lb *structs.LoadBalancer
|
|
if resolver.LoadBalancer != nil {
|
|
lb = resolver.LoadBalancer
|
|
}
|
|
route, err := makeNamedDefaultRouteWithLB(clusterName, lb, resolver.RequestTimeout, autoHostRewrite)
|
|
if err != nil {
|
|
s.Logger.Error("failed to make route", "cluster", clusterName, "error", err)
|
|
return nil, err
|
|
}
|
|
resources = append(resources, route)
|
|
|
|
// If there is a service-resolver for this service then also setup routes for each subset
|
|
for name := range resolver.Subsets {
|
|
clusterName = connect.ServiceSNI(svc.Name, name, svc.NamespaceOrDefault(), svc.PartitionOrDefault(), cfgSnap.Datacenter, cfgSnap.Roots.TrustDomain)
|
|
route, err := makeNamedDefaultRouteWithLB(clusterName, lb, resolver.RequestTimeout, true)
|
|
if err != nil {
|
|
s.Logger.Error("failed to make route", "cluster", clusterName, "error", err)
|
|
return nil, err
|
|
}
|
|
resources = append(resources, route)
|
|
}
|
|
return resources, nil
|
|
}
|
|
|
|
func (s *ResourceGenerator) routesForMeshGateway(cfgSnap *proxycfg.ConfigSnapshot) ([]proto.Message, error) {
|
|
if cfgSnap == nil {
|
|
return nil, errors.New("nil config given")
|
|
}
|
|
|
|
var resources []proto.Message
|
|
for _, svc := range cfgSnap.MeshGatewayValidExportedServices() {
|
|
chain := cfgSnap.MeshGateway.DiscoveryChain[svc]
|
|
|
|
if !structs.IsProtocolHTTPLike(chain.Protocol) {
|
|
continue // ignore; not relevant
|
|
}
|
|
|
|
uid := proxycfg.NewUpstreamIDFromServiceName(svc)
|
|
|
|
virtualHost, err := s.makeUpstreamRouteForDiscoveryChain(
|
|
cfgSnap,
|
|
uid,
|
|
chain,
|
|
[]string{"*"},
|
|
true,
|
|
perRouteFilterBuilder{},
|
|
)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if virtualHost == nil {
|
|
continue
|
|
}
|
|
|
|
route := &envoy_route_v3.RouteConfiguration{
|
|
Name: uid.EnvoyID(),
|
|
VirtualHosts: []*envoy_route_v3.VirtualHost{virtualHost},
|
|
// ValidateClusters defaults to true when defined statically and false
|
|
// when done via RDS. Re-set the reasonable value of true to prevent
|
|
// null-routing traffic.
|
|
ValidateClusters: response.MakeBoolValue(true),
|
|
}
|
|
resources = append(resources, route)
|
|
}
|
|
|
|
return resources, nil
|
|
}
|
|
|
|
func makeNamedDefaultRouteWithLB(clusterName string, lb *structs.LoadBalancer, timeout time.Duration, autoHostRewrite bool) (*envoy_route_v3.RouteConfiguration, error) {
|
|
action := makeRouteActionFromName(clusterName)
|
|
|
|
if err := injectLBToRouteAction(lb, action.Route); err != nil {
|
|
return nil, fmt.Errorf("failed to apply load balancer configuration to route action: %v", err)
|
|
}
|
|
|
|
// Configure Envoy to rewrite Host header
|
|
if autoHostRewrite {
|
|
action.Route.HostRewriteSpecifier = &envoy_route_v3.RouteAction_AutoHostRewrite{
|
|
AutoHostRewrite: response.MakeBoolValue(true),
|
|
}
|
|
}
|
|
|
|
if timeout != 0 {
|
|
action.Route.Timeout = durationpb.New(timeout)
|
|
}
|
|
|
|
return &envoy_route_v3.RouteConfiguration{
|
|
Name: clusterName,
|
|
VirtualHosts: []*envoy_route_v3.VirtualHost{
|
|
{
|
|
Name: clusterName,
|
|
Domains: []string{"*"},
|
|
Routes: []*envoy_route_v3.Route{
|
|
{
|
|
Match: makeDefaultRouteMatch(),
|
|
Action: action,
|
|
},
|
|
},
|
|
},
|
|
},
|
|
// ValidateClusters defaults to true when defined statically and false
|
|
// when done via RDS. Re-set the reasonable value of true to prevent
|
|
// null-routing traffic.
|
|
ValidateClusters: response.MakeBoolValue(true),
|
|
}, nil
|
|
}
|
|
|
|
func makeNamedAddressesRoute(routeName string, addresses map[string]string) (*envoy_route_v3.RouteConfiguration, error) {
|
|
route := &envoy_route_v3.RouteConfiguration{
|
|
Name: routeName,
|
|
// ValidateClusters defaults to true when defined statically and false
|
|
// when done via RDS. Re-set the reasonable value of true to prevent
|
|
// null-routing traffic.
|
|
ValidateClusters: response.MakeBoolValue(true),
|
|
}
|
|
for clusterName, address := range addresses {
|
|
action := makeRouteActionFromName(clusterName)
|
|
virtualHost := &envoy_route_v3.VirtualHost{
|
|
Name: clusterName,
|
|
Domains: []string{address},
|
|
Routes: []*envoy_route_v3.Route{
|
|
{
|
|
Match: makeDefaultRouteMatch(),
|
|
Action: action,
|
|
},
|
|
},
|
|
}
|
|
route.VirtualHosts = append(route.VirtualHosts, virtualHost)
|
|
}
|
|
|
|
// sort virtual hosts to have a stable order
|
|
sort.SliceStable(route.VirtualHosts, func(i, j int) bool {
|
|
return route.VirtualHosts[i].Name > route.VirtualHosts[j].Name
|
|
})
|
|
return route, nil
|
|
}
|
|
|
|
// routesForIngressGateway returns the xDS API representation of the
|
|
// "routes" in the snapshot.
|
|
func (s *ResourceGenerator) routesForIngressGateway(cfgSnap *proxycfg.ConfigSnapshot) ([]proto.Message, error) {
|
|
var result []proto.Message
|
|
for listenerKey, upstreams := range cfgSnap.IngressGateway.Upstreams {
|
|
// Do not create any route configuration for TCP listeners
|
|
if listenerKey.Protocol == "tcp" {
|
|
continue
|
|
}
|
|
|
|
// Depending on their TLS config, upstreams are either attached to the
|
|
// default route or have their own routes. We'll add any upstreams that
|
|
// don't have custom filter chains and routes to this.
|
|
defaultRoute := &envoy_route_v3.RouteConfiguration{
|
|
Name: listenerKey.RouteName(),
|
|
// ValidateClusters defaults to true when defined statically and false
|
|
// when done via RDS. Re-set the reasonable value of true to prevent
|
|
// null-routing traffic.
|
|
ValidateClusters: response.MakeBoolValue(true),
|
|
}
|
|
|
|
for _, u := range upstreams {
|
|
uid := proxycfg.NewUpstreamID(&u)
|
|
chain := cfgSnap.IngressGateway.DiscoveryChain[uid]
|
|
if chain == nil {
|
|
// Note that if we continue here we must also do this in the cluster generation
|
|
s.Logger.Warn("could not find discovery chain for ingress upstream",
|
|
"listener", listenerKey, "upstream", uid)
|
|
continue
|
|
}
|
|
|
|
domains := generateUpstreamIngressDomains(listenerKey, u)
|
|
virtualHost, err := s.makeUpstreamRouteForDiscoveryChain(cfgSnap, uid, chain, domains, false, perRouteFilterBuilder{})
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if virtualHost == nil {
|
|
continue
|
|
}
|
|
|
|
// Lookup listener and service config details from ingress gateway
|
|
// definition.
|
|
lCfg, ok := cfgSnap.IngressGateway.Listeners[listenerKey]
|
|
if !ok {
|
|
return nil, fmt.Errorf("missing ingress listener config (service %q listener on proto/port %s/%d)",
|
|
u.DestinationID(), listenerKey.Protocol, listenerKey.Port)
|
|
}
|
|
svc := findIngressServiceMatchingUpstream(lCfg, u)
|
|
if svc == nil {
|
|
return nil, fmt.Errorf("missing service in listener config (service %q listener on proto/port %s/%d)",
|
|
u.DestinationID(), listenerKey.Protocol, listenerKey.Port)
|
|
}
|
|
|
|
if err := injectHeaderManipToVirtualHost(svc, virtualHost); err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
// See if this upstream has its own route/filter chain
|
|
svcRouteName := routeNameForUpstream(lCfg, *svc)
|
|
|
|
// If the routeName is the same as the default one, merge the virtual host
|
|
// to the default route
|
|
if svcRouteName == defaultRoute.Name {
|
|
defaultRoute.VirtualHosts = append(defaultRoute.VirtualHosts, virtualHost)
|
|
} else {
|
|
svcRoute := &envoy_route_v3.RouteConfiguration{
|
|
Name: svcRouteName,
|
|
ValidateClusters: response.MakeBoolValue(true),
|
|
VirtualHosts: []*envoy_route_v3.VirtualHost{virtualHost},
|
|
}
|
|
result = append(result, svcRoute)
|
|
}
|
|
}
|
|
|
|
if len(defaultRoute.VirtualHosts) > 0 {
|
|
result = append(result, defaultRoute)
|
|
}
|
|
}
|
|
|
|
return result, nil
|
|
}
|
|
|
|
// routesForAPIGateway returns the xDS API representation of the "routes" in the snapshot.
|
|
func (s *ResourceGenerator) routesForAPIGateway(cfgSnap *proxycfg.ConfigSnapshot) ([]proto.Message, error) {
|
|
var result []proto.Message
|
|
|
|
// Build up the routes in a deterministic way
|
|
readyListeners := getReadyListeners(cfgSnap)
|
|
listenerNames := maps.Keys(readyListeners)
|
|
sort.Strings(listenerNames)
|
|
|
|
// Iterate over all listeners that are ready and configure their routes.
|
|
for _, listenerName := range listenerNames {
|
|
readyListener, ok := readyListeners[listenerName]
|
|
if !ok {
|
|
continue
|
|
}
|
|
// Do not create any route configuration for TCP listeners
|
|
if readyListener.listenerCfg.Protocol != structs.ListenerProtocolHTTP {
|
|
continue
|
|
}
|
|
|
|
listenerRoute := &envoy_route_v3.RouteConfiguration{
|
|
Name: readyListener.listenerKey.RouteName(),
|
|
// ValidateClusters defaults to true when defined statically and false
|
|
// when done via RDS. Re-set the reasonable value of true to prevent
|
|
// null-routing traffic.
|
|
ValidateClusters: response.MakeBoolValue(true),
|
|
}
|
|
|
|
// Consolidate all routes for this listener into the minimum possible set based on hostname matching.
|
|
allRoutesForListener := []*structs.HTTPRouteConfigEntry{}
|
|
for _, routeRef := range maps.Keys(readyListener.routeReferences) {
|
|
route, ok := cfgSnap.APIGateway.HTTPRoutes.Get(routeRef)
|
|
if !ok {
|
|
return nil, fmt.Errorf("missing route for route routeRef %s:%s", routeRef.Name, routeRef.Kind)
|
|
}
|
|
allRoutesForListener = append(allRoutesForListener, route)
|
|
}
|
|
consolidatedRoutes := discoverychain.ConsolidateHTTPRoutes(cfgSnap.APIGateway.GatewayConfig, &readyListener.listenerCfg, allRoutesForListener...)
|
|
|
|
// Produce one virtual host per hostname. If no hostname is specified for a set of
|
|
// Gateway + HTTPRoutes, then the virtual host will be "*".
|
|
for _, consolidatedRoute := range consolidatedRoutes {
|
|
upstream := buildHTTPRouteUpstream(consolidatedRoute, readyListener.listenerCfg)
|
|
// Consolidate all routes for this listener into the minimum possible set based on hostname matching.
|
|
uid := proxycfg.NewUpstreamID(&upstream)
|
|
chain := cfgSnap.APIGateway.DiscoveryChain[uid]
|
|
if chain == nil {
|
|
s.Logger.Debug("Discovery chain not found for flattened route", "discovery chain ID", uid)
|
|
continue
|
|
}
|
|
|
|
consolidatedRoute := consolidatedRoute // Reassignment to avoid closure issues with the loop variable.
|
|
domains := generateUpstreamAPIsDomains(readyListener.listenerKey, upstream, consolidatedRoute.Hostnames)
|
|
|
|
filterBuilder := perRouteFilterBuilder{providerMap: cfgSnap.JWTProviders, listener: &readyListener.listenerCfg, route: &consolidatedRoute}
|
|
virtualHost, err := s.makeUpstreamRouteForDiscoveryChain(cfgSnap, uid, chain, domains, false, filterBuilder)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
listenerRoute.VirtualHosts = append(listenerRoute.VirtualHosts, virtualHost)
|
|
}
|
|
|
|
if len(listenerRoute.VirtualHosts) > 0 {
|
|
// Build up the virtual hosts in a deterministic way
|
|
slices.SortStableFunc(listenerRoute.VirtualHosts, func(a, b *envoy_route_v3.VirtualHost) int {
|
|
if a.Name < b.Name {
|
|
return -1
|
|
}
|
|
return 1
|
|
})
|
|
result = append(result, listenerRoute)
|
|
}
|
|
}
|
|
|
|
return result, nil
|
|
}
|
|
|
|
func buildHTTPRouteUpstream(route structs.HTTPRouteConfigEntry, listener structs.APIGatewayListener) structs.Upstream {
|
|
return structs.Upstream{
|
|
DestinationName: route.GetName(),
|
|
DestinationNamespace: route.NamespaceOrDefault(),
|
|
DestinationPartition: route.PartitionOrDefault(),
|
|
IngressHosts: route.Hostnames,
|
|
LocalBindPort: listener.Port,
|
|
Config: map[string]interface{}{
|
|
"protocol": string(listener.Protocol),
|
|
},
|
|
}
|
|
}
|
|
|
|
func makeHeadersValueOptions(vals map[string]string, add bool) []*envoy_core_v3.HeaderValueOption {
|
|
opts := make([]*envoy_core_v3.HeaderValueOption, 0, len(vals))
|
|
for k, v := range vals {
|
|
o := &envoy_core_v3.HeaderValueOption{
|
|
Header: &envoy_core_v3.HeaderValue{
|
|
Key: k,
|
|
Value: v,
|
|
},
|
|
}
|
|
if !add {
|
|
// default is APPEND_IF_EXISTS_OR_ADD
|
|
o.AppendAction = envoy_core_v3.HeaderValueOption_OVERWRITE_IF_EXISTS_OR_ADD
|
|
}
|
|
opts = append(opts, o)
|
|
}
|
|
return opts
|
|
}
|
|
|
|
func findIngressServiceMatchingUpstream(l structs.IngressListener, u structs.Upstream) *structs.IngressService {
|
|
// Hunt through for the matching service. We validate now that there is
|
|
// only one IngressService for each unique name although originally that
|
|
// wasn't checked as it didn't matter. Assume there is only one now
|
|
// though!
|
|
wantSID := u.DestinationID().ServiceName.ToServiceID()
|
|
var foundSameNSWildcard *structs.IngressService
|
|
for _, s := range l.Services {
|
|
sid := structs.NewServiceID(s.Name, &s.EnterpriseMeta)
|
|
if wantSID.Matches(sid) {
|
|
return &s
|
|
}
|
|
if s.Name == structs.WildcardSpecifier &&
|
|
s.NamespaceOrDefault() == wantSID.NamespaceOrDefault() &&
|
|
s.PartitionOrDefault() == wantSID.PartitionOrDefault() {
|
|
// Make a copy so we don't take a reference to the loop variable
|
|
found := s
|
|
foundSameNSWildcard = &found
|
|
}
|
|
}
|
|
// Didn't find an exact match. Return the wildcard from same service if we
|
|
// found one.
|
|
return foundSameNSWildcard
|
|
}
|
|
|
|
func generateUpstreamIngressDomains(listenerKey proxycfg.IngressListenerKey, u structs.Upstream) []string {
|
|
var domains []string
|
|
domainsSet := make(map[string]bool)
|
|
|
|
namespace := u.GetEnterpriseMeta().NamespaceOrDefault()
|
|
switch {
|
|
case len(u.IngressHosts) > 0:
|
|
// If a user has specified hosts, do not add the default
|
|
// "<service-name>.ingress.*" prefixes
|
|
domains = u.IngressHosts
|
|
case namespace != structs.IntentionDefaultNamespace:
|
|
domains = []string{fmt.Sprintf("%s.ingress.%s.*", u.DestinationName, namespace)}
|
|
default:
|
|
domains = []string{fmt.Sprintf("%s.ingress.*", u.DestinationName)}
|
|
}
|
|
|
|
for _, h := range domains {
|
|
domainsSet[h] = true
|
|
}
|
|
|
|
// Host headers may contain port numbers in them, so we need to make sure
|
|
// we match on the host with and without the port number. Well-known
|
|
// ports like HTTP/HTTPS are stripped from Host headers, but other ports
|
|
// will be in the header.
|
|
for _, h := range domains {
|
|
_, _, err := net.SplitHostPort(h)
|
|
// Error message from Go's net/ipsock.go
|
|
// We check to see if a port is not missing, and ignore the
|
|
// error from SplitHostPort otherwise, since we have previously
|
|
// validated the Host values and should trust the user's input.
|
|
if err == nil || !strings.Contains(err.Error(), "missing port in address") {
|
|
continue
|
|
}
|
|
|
|
domainWithPort := fmt.Sprintf("%s:%d", h, listenerKey.Port)
|
|
|
|
// Do not add a duplicate domain if a hostname with port is already in the
|
|
// set
|
|
if !domainsSet[domainWithPort] {
|
|
domains = append(domains, domainWithPort)
|
|
}
|
|
}
|
|
|
|
return domains
|
|
}
|
|
|
|
func generateUpstreamAPIsDomains(listenerKey proxycfg.APIGatewayListenerKey, u structs.Upstream, hosts []string) []string {
|
|
u.IngressHosts = hosts
|
|
return generateUpstreamIngressDomains(listenerKey, u)
|
|
}
|
|
|
|
func (s *ResourceGenerator) makeUpstreamRouteForDiscoveryChain(
|
|
cfgSnap *proxycfg.ConfigSnapshot,
|
|
uid proxycfg.UpstreamID,
|
|
chain *structs.CompiledDiscoveryChain,
|
|
serviceDomains []string,
|
|
forMeshGateway bool,
|
|
filterBuilder perRouteFilterBuilder,
|
|
) (*envoy_route_v3.VirtualHost, error) {
|
|
routeName := uid.EnvoyID()
|
|
var routes []*envoy_route_v3.Route
|
|
|
|
startNode := chain.Nodes[chain.StartNode]
|
|
if startNode == nil {
|
|
return nil, fmt.Errorf("missing first node in compiled discovery chain for: %s", chain.ServiceName)
|
|
}
|
|
|
|
upstreamsSnapshot, err := cfgSnap.ToConfigSnapshotUpstreams()
|
|
if err != nil && !forMeshGateway {
|
|
return nil, err
|
|
}
|
|
|
|
switch startNode.Type {
|
|
case structs.DiscoveryGraphNodeTypeRouter:
|
|
routes = make([]*envoy_route_v3.Route, 0, len(startNode.Routes))
|
|
|
|
for _, discoveryRoute := range startNode.Routes {
|
|
discoveryRoute := discoveryRoute
|
|
routeMatch := makeRouteMatchForDiscoveryRoute(discoveryRoute)
|
|
|
|
var (
|
|
routeAction *envoy_route_v3.Route_Route
|
|
err error
|
|
)
|
|
|
|
nextNode := chain.Nodes[discoveryRoute.NextNode]
|
|
|
|
var lb *structs.LoadBalancer
|
|
if nextNode.LoadBalancer != nil {
|
|
lb = nextNode.LoadBalancer
|
|
}
|
|
|
|
switch nextNode.Type {
|
|
case structs.DiscoveryGraphNodeTypeSplitter:
|
|
routeAction, err = s.makeRouteActionForSplitter(upstreamsSnapshot, nextNode.Splits, chain, forMeshGateway)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
case structs.DiscoveryGraphNodeTypeResolver:
|
|
ra, ok := s.makeRouteActionForChainCluster(upstreamsSnapshot, nextNode.Resolver.Target, chain, forMeshGateway)
|
|
if !ok {
|
|
continue
|
|
}
|
|
routeAction = ra
|
|
|
|
default:
|
|
return nil, fmt.Errorf("unexpected graph node after route %q", nextNode.Type)
|
|
}
|
|
|
|
if err := injectLBToRouteAction(lb, routeAction.Route); err != nil {
|
|
return nil, fmt.Errorf("failed to apply load balancer configuration to route action: %v", err)
|
|
}
|
|
|
|
// TODO(rb): Better help handle the envoy case where you need (prefix=/foo/,rewrite=/) and (exact=/foo,rewrite=/) to do a full rewrite
|
|
|
|
destination := discoveryRoute.Definition.Destination
|
|
|
|
route := &envoy_route_v3.Route{}
|
|
|
|
if destination != nil {
|
|
if destination.PrefixRewrite != "" {
|
|
routeAction.Route.PrefixRewrite = destination.PrefixRewrite
|
|
}
|
|
|
|
if destination.RequestTimeout > 0 {
|
|
routeAction.Route.Timeout = durationpb.New(destination.RequestTimeout)
|
|
}
|
|
// Disable the timeout if user specifies negative value. Setting 0 disables the timeout in Envoy.
|
|
if destination.RequestTimeout < 0 {
|
|
routeAction.Route.Timeout = durationpb.New(0 * time.Second)
|
|
}
|
|
|
|
if destination.IdleTimeout > 0 {
|
|
routeAction.Route.IdleTimeout = durationpb.New(destination.IdleTimeout)
|
|
}
|
|
// Disable the timeout if user specifies negative value. Setting 0 disables the timeout in Envoy.
|
|
if destination.IdleTimeout < 0 {
|
|
routeAction.Route.IdleTimeout = durationpb.New(0 * time.Second)
|
|
}
|
|
|
|
if destination.HasRetryFeatures() {
|
|
routeAction.Route.RetryPolicy = getRetryPolicyForDestination(destination)
|
|
}
|
|
|
|
if err := injectHeaderManipToRoute(destination, route); err != nil {
|
|
return nil, fmt.Errorf("failed to apply header manipulation configuration to route: %v", err)
|
|
}
|
|
}
|
|
|
|
filter, err := filterBuilder.buildTypedPerFilterConfig(routeMatch, routeAction)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
route.Match = routeMatch
|
|
route.Action = routeAction
|
|
route.TypedPerFilterConfig = filter
|
|
|
|
routes = append(routes, route)
|
|
}
|
|
|
|
case structs.DiscoveryGraphNodeTypeSplitter:
|
|
routeAction, err := s.makeRouteActionForSplitter(upstreamsSnapshot, startNode.Splits, chain, forMeshGateway)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
var lb *structs.LoadBalancer
|
|
if startNode.LoadBalancer != nil {
|
|
lb = startNode.LoadBalancer
|
|
}
|
|
if err := injectLBToRouteAction(lb, routeAction.Route); err != nil {
|
|
return nil, fmt.Errorf("failed to apply load balancer configuration to route action: %v", err)
|
|
}
|
|
|
|
defaultRoute := &envoy_route_v3.Route{
|
|
Match: makeDefaultRouteMatch(),
|
|
Action: routeAction,
|
|
}
|
|
|
|
routes = []*envoy_route_v3.Route{defaultRoute}
|
|
|
|
case structs.DiscoveryGraphNodeTypeResolver:
|
|
routeAction, ok := s.makeRouteActionForChainCluster(upstreamsSnapshot, startNode.Resolver.Target, chain, forMeshGateway)
|
|
if !ok {
|
|
break
|
|
}
|
|
var lb *structs.LoadBalancer
|
|
if startNode.LoadBalancer != nil {
|
|
lb = startNode.LoadBalancer
|
|
}
|
|
if err := injectLBToRouteAction(lb, routeAction.Route); err != nil {
|
|
return nil, fmt.Errorf("failed to apply load balancer configuration to route action: %v", err)
|
|
}
|
|
|
|
// A request timeout can be configured on a resolver or router. If configured on a resolver, the timeout will
|
|
// only apply if the start node is a resolver. This is because the timeout is attached to an (Envoy
|
|
// RouteAction)[https://www.envoyproxy.io/docs/envoy/latest/api-v3/config/route/v3/route_components.proto#envoy-v3-api-msg-config-route-v3-routeaction]
|
|
// If there is a splitter before this resolver, the branches of the split are configured within the same
|
|
// RouteAction, and the timeout cannot be shared between branches of a split.
|
|
if startNode.Resolver.RequestTimeout > 0 {
|
|
routeAction.Route.Timeout = durationpb.New(startNode.Resolver.RequestTimeout)
|
|
}
|
|
// Disable the timeout if user specifies negative value. Setting 0 disables the timeout in Envoy.
|
|
if startNode.Resolver.RequestTimeout < 0 {
|
|
routeAction.Route.Timeout = durationpb.New(0 * time.Second)
|
|
}
|
|
defaultRoute := &envoy_route_v3.Route{
|
|
Match: makeDefaultRouteMatch(),
|
|
Action: routeAction,
|
|
}
|
|
|
|
routes = []*envoy_route_v3.Route{defaultRoute}
|
|
|
|
default:
|
|
return nil, fmt.Errorf("unknown first node in discovery chain of type: %s", startNode.Type)
|
|
}
|
|
|
|
host := &envoy_route_v3.VirtualHost{
|
|
Name: routeName,
|
|
Domains: serviceDomains,
|
|
Routes: routes,
|
|
}
|
|
|
|
return host, nil
|
|
}
|
|
|
|
func getRetryPolicyForDestination(destination *structs.ServiceRouteDestination) *envoy_route_v3.RetryPolicy {
|
|
retryPolicy := &envoy_route_v3.RetryPolicy{}
|
|
if destination.NumRetries > 0 {
|
|
retryPolicy.NumRetries = response.MakeUint32Value(int(destination.NumRetries))
|
|
}
|
|
|
|
// The RetryOn magic values come from: https://www.envoyproxy.io/docs/envoy/v1.10.0/configuration/http_filters/router_filter#config-http-filters-router-x-envoy-retry-on
|
|
var retryStrings []string
|
|
|
|
if len(destination.RetryOn) > 0 {
|
|
retryStrings = append(retryStrings, destination.RetryOn...)
|
|
}
|
|
|
|
if destination.RetryOnConnectFailure {
|
|
// connect-failure can be enabled by either adding connect-failure to the RetryOn list or by using the legacy RetryOnConnectFailure option
|
|
// Check that it's not already in the RetryOn list, so we don't set it twice
|
|
connectFailureExists := false
|
|
for _, r := range retryStrings {
|
|
if r == "connect-failure" {
|
|
connectFailureExists = true
|
|
}
|
|
}
|
|
if !connectFailureExists {
|
|
retryStrings = append(retryStrings, "connect-failure")
|
|
}
|
|
}
|
|
|
|
if len(destination.RetryOnStatusCodes) > 0 {
|
|
retryStrings = append(retryStrings, "retriable-status-codes")
|
|
retryPolicy.RetriableStatusCodes = destination.RetryOnStatusCodes
|
|
}
|
|
|
|
retryPolicy.RetryOn = strings.Join(retryStrings, ",")
|
|
|
|
return retryPolicy
|
|
}
|
|
|
|
func makeRouteMatchForDiscoveryRoute(discoveryRoute *structs.DiscoveryRoute) *envoy_route_v3.RouteMatch {
|
|
match := discoveryRoute.Definition.Match
|
|
if match == nil || match.IsEmpty() {
|
|
return makeDefaultRouteMatch()
|
|
}
|
|
|
|
em := &envoy_route_v3.RouteMatch{}
|
|
|
|
switch {
|
|
case match.HTTP.PathExact != "":
|
|
em.PathSpecifier = &envoy_route_v3.RouteMatch_Path{
|
|
Path: match.HTTP.PathExact,
|
|
}
|
|
case match.HTTP.PathPrefix != "":
|
|
em.PathSpecifier = &envoy_route_v3.RouteMatch_Prefix{
|
|
Prefix: match.HTTP.PathPrefix,
|
|
}
|
|
case match.HTTP.PathRegex != "":
|
|
em.PathSpecifier = &envoy_route_v3.RouteMatch_SafeRegex{
|
|
SafeRegex: response.MakeEnvoyRegexMatch(match.HTTP.PathRegex),
|
|
}
|
|
default:
|
|
em.PathSpecifier = &envoy_route_v3.RouteMatch_Prefix{
|
|
Prefix: "/",
|
|
}
|
|
}
|
|
|
|
if match.HTTP.CaseInsensitive {
|
|
em.CaseSensitive = wrapperspb.Bool(false)
|
|
}
|
|
|
|
if len(match.HTTP.Header) > 0 {
|
|
em.Headers = make([]*envoy_route_v3.HeaderMatcher, 0, len(match.HTTP.Header))
|
|
for _, hdr := range match.HTTP.Header {
|
|
eh := &envoy_route_v3.HeaderMatcher{
|
|
Name: hdr.Name,
|
|
}
|
|
|
|
switch {
|
|
case hdr.Exact != "":
|
|
eh.HeaderMatchSpecifier = &envoy_route_v3.HeaderMatcher_StringMatch{
|
|
StringMatch: &envoy_matcher_v3.StringMatcher{
|
|
MatchPattern: &envoy_matcher_v3.StringMatcher_Exact{
|
|
Exact: hdr.Exact,
|
|
},
|
|
IgnoreCase: false,
|
|
},
|
|
}
|
|
case hdr.Regex != "":
|
|
eh.HeaderMatchSpecifier = &envoy_route_v3.HeaderMatcher_StringMatch{
|
|
StringMatch: &envoy_matcher_v3.StringMatcher{
|
|
MatchPattern: &envoy_matcher_v3.StringMatcher_SafeRegex{
|
|
SafeRegex: response.MakeEnvoyRegexMatch(hdr.Regex),
|
|
},
|
|
IgnoreCase: false,
|
|
},
|
|
}
|
|
|
|
case hdr.Prefix != "":
|
|
eh.HeaderMatchSpecifier = &envoy_route_v3.HeaderMatcher_StringMatch{
|
|
StringMatch: &envoy_matcher_v3.StringMatcher{
|
|
MatchPattern: &envoy_matcher_v3.StringMatcher_Prefix{
|
|
Prefix: hdr.Prefix,
|
|
},
|
|
IgnoreCase: false,
|
|
},
|
|
}
|
|
|
|
case hdr.Suffix != "":
|
|
eh.HeaderMatchSpecifier = &envoy_route_v3.HeaderMatcher_StringMatch{
|
|
StringMatch: &envoy_matcher_v3.StringMatcher{
|
|
MatchPattern: &envoy_matcher_v3.StringMatcher_Suffix{
|
|
Suffix: hdr.Suffix,
|
|
},
|
|
IgnoreCase: false,
|
|
},
|
|
}
|
|
|
|
case hdr.Present:
|
|
eh.HeaderMatchSpecifier = &envoy_route_v3.HeaderMatcher_PresentMatch{
|
|
PresentMatch: true,
|
|
}
|
|
default:
|
|
continue // skip this impossible situation
|
|
}
|
|
|
|
if hdr.Invert {
|
|
eh.InvertMatch = true
|
|
}
|
|
|
|
em.Headers = append(em.Headers, eh)
|
|
}
|
|
}
|
|
|
|
if len(match.HTTP.Methods) > 0 {
|
|
methodHeaderRegex := strings.Join(match.HTTP.Methods, "|")
|
|
|
|
eh := &envoy_route_v3.HeaderMatcher{
|
|
Name: ":method",
|
|
HeaderMatchSpecifier: &envoy_route_v3.HeaderMatcher_StringMatch{
|
|
StringMatch: &envoy_matcher_v3.StringMatcher{
|
|
MatchPattern: &envoy_matcher_v3.StringMatcher_SafeRegex{
|
|
SafeRegex: response.MakeEnvoyRegexMatch(methodHeaderRegex),
|
|
},
|
|
},
|
|
},
|
|
}
|
|
|
|
em.Headers = append(em.Headers, eh)
|
|
}
|
|
|
|
if len(match.HTTP.QueryParam) > 0 {
|
|
em.QueryParameters = make([]*envoy_route_v3.QueryParameterMatcher, 0, len(match.HTTP.QueryParam))
|
|
for _, qm := range match.HTTP.QueryParam {
|
|
eq := &envoy_route_v3.QueryParameterMatcher{
|
|
Name: qm.Name,
|
|
}
|
|
|
|
switch {
|
|
case qm.Exact != "":
|
|
eq.QueryParameterMatchSpecifier = &envoy_route_v3.QueryParameterMatcher_StringMatch{
|
|
StringMatch: &envoy_matcher_v3.StringMatcher{
|
|
MatchPattern: &envoy_matcher_v3.StringMatcher_Exact{
|
|
Exact: qm.Exact,
|
|
},
|
|
},
|
|
}
|
|
case qm.Regex != "":
|
|
eq.QueryParameterMatchSpecifier = &envoy_route_v3.QueryParameterMatcher_StringMatch{
|
|
StringMatch: &envoy_matcher_v3.StringMatcher{
|
|
MatchPattern: &envoy_matcher_v3.StringMatcher_SafeRegex{
|
|
SafeRegex: response.MakeEnvoyRegexMatch(qm.Regex),
|
|
},
|
|
},
|
|
}
|
|
case qm.Present:
|
|
eq.QueryParameterMatchSpecifier = &envoy_route_v3.QueryParameterMatcher_PresentMatch{
|
|
PresentMatch: true,
|
|
}
|
|
default:
|
|
continue // skip this impossible situation
|
|
}
|
|
|
|
em.QueryParameters = append(em.QueryParameters, eq)
|
|
}
|
|
}
|
|
|
|
return em
|
|
}
|
|
|
|
func makeDefaultRouteMatch() *envoy_route_v3.RouteMatch {
|
|
return &envoy_route_v3.RouteMatch{
|
|
PathSpecifier: &envoy_route_v3.RouteMatch_Prefix{
|
|
Prefix: "/",
|
|
},
|
|
// TODO(banks) Envoy supports matching only valid GRPC
|
|
// requests which might be nice to add here for gRPC services
|
|
// but it's not supported in our current envoy SDK version
|
|
// although docs say it was supported by 1.8.0. Going to defer
|
|
// that until we've updated the deps.
|
|
}
|
|
}
|
|
|
|
func (s *ResourceGenerator) makeRouteActionForChainCluster(
|
|
upstreamsSnapshot *proxycfg.ConfigSnapshotUpstreams,
|
|
targetID string,
|
|
chain *structs.CompiledDiscoveryChain,
|
|
forMeshGateway bool,
|
|
) (*envoy_route_v3.Route_Route, bool) {
|
|
clusterName := s.getTargetClusterName(upstreamsSnapshot, chain, targetID, forMeshGateway)
|
|
if clusterName == "" {
|
|
return nil, false
|
|
}
|
|
return makeRouteActionFromName(clusterName), true
|
|
}
|
|
|
|
func makeRouteActionFromName(clusterName string) *envoy_route_v3.Route_Route {
|
|
return &envoy_route_v3.Route_Route{
|
|
Route: &envoy_route_v3.RouteAction{
|
|
ClusterSpecifier: &envoy_route_v3.RouteAction_Cluster{
|
|
Cluster: clusterName,
|
|
},
|
|
},
|
|
}
|
|
}
|
|
|
|
func (s *ResourceGenerator) makeRouteActionForSplitter(
|
|
upstreamsSnapshot *proxycfg.ConfigSnapshotUpstreams,
|
|
splits []*structs.DiscoverySplit,
|
|
chain *structs.CompiledDiscoveryChain,
|
|
forMeshGateway bool,
|
|
) (*envoy_route_v3.Route_Route, error) {
|
|
clusters := make([]*envoy_route_v3.WeightedCluster_ClusterWeight, 0, len(splits))
|
|
for _, split := range splits {
|
|
nextNode := chain.Nodes[split.NextNode]
|
|
|
|
if nextNode.Type != structs.DiscoveryGraphNodeTypeResolver {
|
|
return nil, fmt.Errorf("unexpected splitter destination node type: %s", nextNode.Type)
|
|
}
|
|
targetID := nextNode.Resolver.Target
|
|
|
|
clusterName := s.getTargetClusterName(upstreamsSnapshot, chain, targetID, forMeshGateway)
|
|
if clusterName == "" {
|
|
continue
|
|
}
|
|
|
|
// The smallest representable weight is 1/10000 or .01% but envoy
|
|
// deals with integers so scale everything up by 100x.
|
|
weight := int(split.Weight * 100)
|
|
cw := &envoy_route_v3.WeightedCluster_ClusterWeight{
|
|
Weight: response.MakeUint32Value(weight),
|
|
Name: clusterName,
|
|
}
|
|
if err := injectHeaderManipToWeightedCluster(split.Definition, cw); err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
clusters = append(clusters, cw)
|
|
}
|
|
|
|
if len(clusters) <= 0 {
|
|
return nil, fmt.Errorf("number of clusters in splitter must be > 0; got %d", len(clusters))
|
|
}
|
|
|
|
return &envoy_route_v3.Route_Route{
|
|
Route: &envoy_route_v3.RouteAction{
|
|
ClusterSpecifier: &envoy_route_v3.RouteAction_WeightedClusters{
|
|
WeightedClusters: &envoy_route_v3.WeightedCluster{
|
|
Clusters: clusters,
|
|
},
|
|
},
|
|
},
|
|
}, nil
|
|
}
|
|
|
|
func injectLBToRouteAction(lb *structs.LoadBalancer, action *envoy_route_v3.RouteAction) error {
|
|
if lb == nil || !lb.IsHashBased() {
|
|
return nil
|
|
}
|
|
|
|
result := make([]*envoy_route_v3.RouteAction_HashPolicy, 0, len(lb.HashPolicies))
|
|
for _, policy := range lb.HashPolicies {
|
|
if policy.SourceIP {
|
|
result = append(result, &envoy_route_v3.RouteAction_HashPolicy{
|
|
PolicySpecifier: &envoy_route_v3.RouteAction_HashPolicy_ConnectionProperties_{
|
|
ConnectionProperties: &envoy_route_v3.RouteAction_HashPolicy_ConnectionProperties{
|
|
SourceIp: true,
|
|
},
|
|
},
|
|
Terminal: policy.Terminal,
|
|
})
|
|
|
|
continue
|
|
}
|
|
|
|
switch policy.Field {
|
|
case structs.HashPolicyHeader:
|
|
result = append(result, &envoy_route_v3.RouteAction_HashPolicy{
|
|
PolicySpecifier: &envoy_route_v3.RouteAction_HashPolicy_Header_{
|
|
Header: &envoy_route_v3.RouteAction_HashPolicy_Header{
|
|
HeaderName: policy.FieldValue,
|
|
},
|
|
},
|
|
Terminal: policy.Terminal,
|
|
})
|
|
case structs.HashPolicyCookie:
|
|
cookie := envoy_route_v3.RouteAction_HashPolicy_Cookie{
|
|
Name: policy.FieldValue,
|
|
}
|
|
if policy.CookieConfig != nil {
|
|
cookie.Path = policy.CookieConfig.Path
|
|
|
|
if policy.CookieConfig.TTL != 0*time.Second {
|
|
cookie.Ttl = durationpb.New(policy.CookieConfig.TTL)
|
|
}
|
|
|
|
// Envoy will generate a session cookie if the ttl is present and zero.
|
|
if policy.CookieConfig.Session {
|
|
cookie.Ttl = durationpb.New(0 * time.Second)
|
|
}
|
|
}
|
|
result = append(result, &envoy_route_v3.RouteAction_HashPolicy{
|
|
PolicySpecifier: &envoy_route_v3.RouteAction_HashPolicy_Cookie_{
|
|
Cookie: &cookie,
|
|
},
|
|
Terminal: policy.Terminal,
|
|
})
|
|
case structs.HashPolicyQueryParam:
|
|
result = append(result, &envoy_route_v3.RouteAction_HashPolicy{
|
|
PolicySpecifier: &envoy_route_v3.RouteAction_HashPolicy_QueryParameter_{
|
|
QueryParameter: &envoy_route_v3.RouteAction_HashPolicy_QueryParameter{
|
|
Name: policy.FieldValue,
|
|
},
|
|
},
|
|
Terminal: policy.Terminal,
|
|
})
|
|
default:
|
|
return fmt.Errorf("unsupported load balancer hash policy field: %v", policy.Field)
|
|
}
|
|
}
|
|
action.HashPolicy = result
|
|
return nil
|
|
}
|
|
|
|
func injectHeaderManipToRoute(dest *structs.ServiceRouteDestination, r *envoy_route_v3.Route) error {
|
|
|
|
if !dest.RequestHeaders.IsZero() {
|
|
r.RequestHeadersToAdd = append(
|
|
r.RequestHeadersToAdd,
|
|
makeHeadersValueOptions(dest.RequestHeaders.Add, true)...,
|
|
)
|
|
r.RequestHeadersToAdd = append(
|
|
r.RequestHeadersToAdd,
|
|
makeHeadersValueOptions(dest.RequestHeaders.Set, false)...,
|
|
)
|
|
r.RequestHeadersToRemove = append(
|
|
r.RequestHeadersToRemove,
|
|
dest.RequestHeaders.Remove...,
|
|
)
|
|
}
|
|
if !dest.ResponseHeaders.IsZero() {
|
|
r.ResponseHeadersToAdd = append(
|
|
r.ResponseHeadersToAdd,
|
|
makeHeadersValueOptions(dest.ResponseHeaders.Add, true)...,
|
|
)
|
|
r.ResponseHeadersToAdd = append(
|
|
r.ResponseHeadersToAdd,
|
|
makeHeadersValueOptions(dest.ResponseHeaders.Set, false)...,
|
|
)
|
|
r.ResponseHeadersToRemove = append(
|
|
r.ResponseHeadersToRemove,
|
|
dest.ResponseHeaders.Remove...,
|
|
)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func injectHeaderManipToVirtualHost(dest *structs.IngressService, vh *envoy_route_v3.VirtualHost) error {
|
|
if !dest.RequestHeaders.IsZero() {
|
|
vh.RequestHeadersToAdd = append(
|
|
vh.RequestHeadersToAdd,
|
|
makeHeadersValueOptions(dest.RequestHeaders.Add, true)...,
|
|
)
|
|
vh.RequestHeadersToAdd = append(
|
|
vh.RequestHeadersToAdd,
|
|
makeHeadersValueOptions(dest.RequestHeaders.Set, false)...,
|
|
)
|
|
vh.RequestHeadersToRemove = append(
|
|
vh.RequestHeadersToRemove,
|
|
dest.RequestHeaders.Remove...,
|
|
)
|
|
}
|
|
if !dest.ResponseHeaders.IsZero() {
|
|
vh.ResponseHeadersToAdd = append(
|
|
vh.ResponseHeadersToAdd,
|
|
makeHeadersValueOptions(dest.ResponseHeaders.Add, true)...,
|
|
)
|
|
vh.ResponseHeadersToAdd = append(
|
|
vh.ResponseHeadersToAdd,
|
|
makeHeadersValueOptions(dest.ResponseHeaders.Set, false)...,
|
|
)
|
|
vh.ResponseHeadersToRemove = append(
|
|
vh.ResponseHeadersToRemove,
|
|
dest.ResponseHeaders.Remove...,
|
|
)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func injectHeaderManipToWeightedCluster(split *structs.ServiceSplit, c *envoy_route_v3.WeightedCluster_ClusterWeight) error {
|
|
if !split.RequestHeaders.IsZero() {
|
|
c.RequestHeadersToAdd = append(
|
|
c.RequestHeadersToAdd,
|
|
makeHeadersValueOptions(split.RequestHeaders.Add, true)...,
|
|
)
|
|
c.RequestHeadersToAdd = append(
|
|
c.RequestHeadersToAdd,
|
|
makeHeadersValueOptions(split.RequestHeaders.Set, false)...,
|
|
)
|
|
c.RequestHeadersToRemove = append(
|
|
c.RequestHeadersToRemove,
|
|
split.RequestHeaders.Remove...,
|
|
)
|
|
}
|
|
if !split.ResponseHeaders.IsZero() {
|
|
c.ResponseHeadersToAdd = append(
|
|
c.ResponseHeadersToAdd,
|
|
makeHeadersValueOptions(split.ResponseHeaders.Add, true)...,
|
|
)
|
|
c.ResponseHeadersToAdd = append(
|
|
c.ResponseHeadersToAdd,
|
|
makeHeadersValueOptions(split.ResponseHeaders.Set, false)...,
|
|
)
|
|
c.ResponseHeadersToRemove = append(
|
|
c.ResponseHeadersToRemove,
|
|
split.ResponseHeaders.Remove...,
|
|
)
|
|
}
|
|
return nil
|
|
}
|