[Backport/1.9.x] Backport #10073 (#10104)

* Merge pull request #10094 from hashicorp/update-fingerprint

updating fingerprint

* Add replication metrics (#10073)

Co-authored-by: Daniel Nephin <dnephin@hashicorp.com>

.circleci/config.yml

@@ -655,7 +655,7 @@ jobs:
       - checkout
       - add_ssh_keys: # needs a key to push updated static asset commit back to github
           fingerprints:
-            - "3d:6b:98:55:78:4e:52:17:4e:17:ba:f3:bf:0b:96:2a"
+            - "fc:55:84:15:0a:1d:c8:e9:06:d0:e8:9c:7b:a9:b7:31"
       - attach_workspace:
           at: .
       - run:
@@ -665,8 +665,8 @@ jobs:
             # if there are, we commit the ui static asset file
             # HEAD^! is shorthand for HEAD^..HEAD (parent of HEAD and HEAD)
             if ! git diff --quiet --exit-code HEAD^! ui/; then
-              git config --local user.email "hashicorp-ci@users.noreply.github.com"
-              git config --local user.name "hashicorp-ci"
+              git config --local user.email "github-team-consul-core@hashicorp.com"
+              git config --local user.name "hc-github-team-consul-core"
 
               short_sha=$(git rev-parse --short HEAD)
               git add agent/uiserver/bindata_assetfs.go
@@ -844,7 +844,7 @@ jobs:
       - checkout
       - add_ssh_keys: # needs a key to push cherry-picked commits back to github
           fingerprints:
-            - "3d:6b:98:55:78:4e:52:17:4e:17:ba:f3:bf:0b:96:2a"
+            - "fc:55:84:15:0a:1d:c8:e9:06:d0:e8:9c:7b:a9:b7:31"
       - run: .circleci/scripts/cherry-picker.sh
       - run: *notify-slack-failure
 

.circleci/scripts/cherry-picker.sh

@@ -160,8 +160,8 @@ fi
 backport_failures=0
 # loop through all labels on the PR
 for label in $labels; do
-    git config --local user.email "hashicorp-ci@users.noreply.github.com"
-    git config --local user.name "hashicorp-ci"
+    git config --local user.email "github-team-consul-core@hashicorp.com"
+    git config --local user.name "hc-github-team-consul-core"
     status "checking label: $label"
     # if the label matches docs-cherrypick, it will attempt to cherry-pick to stable-website
     if [[ $label =~ docs-cherrypick ]]; then

agent/consul/leader.go

@@ -812,10 +812,19 @@ func (s *Server) runLegacyACLReplication(ctx context.Context) error {
 		}
 
 		if err != nil {
+			metrics.SetGauge([]string{"leader", "replication", "acl-legacy", "status"},
+				0,
+			)
 			lastRemoteIndex = 0
 			s.updateACLReplicationStatusError()
 			legacyACLLogger.Warn("Legacy ACL replication error (will retry if still leader)", "error", err)
 		} else {
+			metrics.SetGauge([]string{"leader", "replication", "acl-legacy", "status"},
+				1,
+			)
+			metrics.SetGauge([]string{"leader", "replication", "acl-legacy", "index"},
+				float32(index),
+			)
 			lastRemoteIndex = index
 			s.updateACLReplicationStatusIndex(structs.ACLReplicateLegacy, index)
 			legacyACLLogger.Debug("Legacy ACL replication completed through remote index", "index", index)
@@ -873,7 +882,7 @@ type replicateFunc func(ctx context.Context, logger hclog.Logger, lastRemoteInde
 func (s *Server) runACLPolicyReplicator(ctx context.Context) error {
 	policyLogger := s.aclReplicationLogger(structs.ACLReplicatePolicies.SingularNoun())
 	policyLogger.Info("started ACL Policy replication")
-	return s.runACLReplicator(ctx, policyLogger, structs.ACLReplicatePolicies, s.replicateACLPolicies)
+	return s.runACLReplicator(ctx, policyLogger, structs.ACLReplicatePolicies, s.replicateACLPolicies, "acl-policies")
 }
 
 // This function is only intended to be run as a managed go routine, it will block until
@@ -881,7 +890,7 @@ func (s *Server) runACLPolicyReplicator(ctx context.Context) error {
 func (s *Server) runACLRoleReplicator(ctx context.Context) error {
 	roleLogger := s.aclReplicationLogger(structs.ACLReplicateRoles.SingularNoun())
 	roleLogger.Info("started ACL Role replication")
-	return s.runACLReplicator(ctx, roleLogger, structs.ACLReplicateRoles, s.replicateACLRoles)
+	return s.runACLReplicator(ctx, roleLogger, structs.ACLReplicateRoles, s.replicateACLRoles, "acl-roles")
 }
 
 // This function is only intended to be run as a managed go routine, it will block until
@@ -889,7 +898,7 @@ func (s *Server) runACLRoleReplicator(ctx context.Context) error {
 func (s *Server) runACLTokenReplicator(ctx context.Context) error {
 	tokenLogger := s.aclReplicationLogger(structs.ACLReplicateTokens.SingularNoun())
 	tokenLogger.Info("started ACL Token replication")
-	return s.runACLReplicator(ctx, tokenLogger, structs.ACLReplicateTokens, s.replicateACLTokens)
+	return s.runACLReplicator(ctx, tokenLogger, structs.ACLReplicateTokens, s.replicateACLTokens, "acl-tokens")
 }
 
 // This function is only intended to be run as a managed go routine, it will block until
@@ -899,6 +908,7 @@ func (s *Server) runACLReplicator(
 	logger hclog.Logger,
 	replicationType structs.ACLReplicationType,
 	replicateFunc replicateFunc,
+	metricName string,
 ) error {
 	var failedAttempts uint
 	limiter := rate.NewLimiter(rate.Limit(s.config.ACLReplicationRate), s.config.ACLReplicationBurst)
@@ -919,6 +929,9 @@ func (s *Server) runACLReplicator(
 		}
 
 		if err != nil {
+			metrics.SetGauge([]string{"leader", "replication", metricName, "status"},
+				0,
+			)
 			lastRemoteIndex = 0
 			s.updateACLReplicationStatusError()
 			logger.Warn("ACL replication error (will retry if still leader)",
@@ -935,6 +948,12 @@ func (s *Server) runACLReplicator(
 				// do nothing
 			}
 		} else {
+			metrics.SetGauge([]string{"leader", "replication", metricName, "status"},
+				1,
+			)
+			metrics.SetGauge([]string{"leader", "replication", metricName, "index"},
+				float32(index),
+			)
 			lastRemoteIndex = index
 			s.updateACLReplicationStatusIndex(replicationType, index)
 			logger.Debug("ACL replication completed through remote index",

agent/consul/replication.go

@@ -22,6 +22,7 @@ const (
 
 type ReplicatorDelegate interface {
 	Replicate(ctx context.Context, lastRemoteIndex uint64, logger hclog.Logger) (index uint64, exit bool, err error)
+	MetricName() string
 }
 
 type ReplicatorConfig struct {
@@ -100,6 +101,9 @@ func (r *Replicator) Run(ctx context.Context) error {
 			return nil
 		}
 		if err != nil {
+			metrics.SetGauge([]string{"leader", "replication", r.delegate.MetricName(), "status"},
+				0,
+			)
 			// reset the lastRemoteIndex when there is an RPC failure. This should cause a full sync to be done during
 			// the next round of replication
 			atomic.StoreUint64(&r.lastRemoteIndex, 0)
@@ -114,6 +118,13 @@ func (r *Replicator) Run(ctx context.Context) error {
 			continue
 		}
 
+		metrics.SetGauge([]string{"leader", "replication", r.delegate.MetricName(), "status"},
+			1,
+		)
+		metrics.SetGauge([]string{"leader", "replication", r.delegate.MetricName(), "index"},
+			float32(index),
+		)
+
 		atomic.StoreUint64(&r.lastRemoteIndex, index)
 		r.logger.Debug("replication completed through remote index", "index", index)
 		r.waiter.Reset()
@@ -128,6 +139,11 @@ type ReplicatorFunc func(ctx context.Context, lastRemoteIndex uint64, logger hcl
 
 type FunctionReplicator struct {
 	ReplicateFn ReplicatorFunc
+	Name        string
+}
+
+func (r *FunctionReplicator) MetricName() string {
+	return r.Name
 }
 
 func (r *FunctionReplicator) Replicate(ctx context.Context, lastRemoteIndex uint64, logger hclog.Logger) (uint64, bool, error) {
@@ -171,6 +187,10 @@ type IndexReplicator struct {
 	Logger   hclog.Logger
 }
 
+func (r *IndexReplicator) MetricName() string {
+	return r.Delegate.MetricName()
+}
+
 func (r *IndexReplicator) Replicate(ctx context.Context, lastRemoteIndex uint64, _ hclog.Logger) (uint64, bool, error) {
 	fetchStart := time.Now()
 	lenRemote, remote, remoteIndex, err := r.Delegate.FetchRemote(lastRemoteIndex)

agent/consul/replication_test.go

@@ -20,6 +20,7 @@ func TestReplicationRestart(t *testing.T) {
 			ReplicateFn: func(ctx context.Context, lastRemoteIndex uint64, logger hclog.Logger) (uint64, bool, error) {
 				return 1, false, nil
 			},
+			Name: "foo",
 		},
 
 		Rate:  1,

agent/consul/server.go

@@ -400,7 +400,7 @@ func NewServer(config *Config, flat Deps) (*Server, error) {
 
 	configReplicatorConfig := ReplicatorConfig{
 		Name:     logging.ConfigEntry,
-		Delegate: &FunctionReplicator{ReplicateFn: s.replicateConfig},
+		Delegate: &FunctionReplicator{ReplicateFn: s.replicateConfig, Name: "config-entries"},
 		Rate:     s.config.ConfigReplicationRate,
 		Burst:    s.config.ConfigReplicationBurst,
 		Logger:   s.logger,