diff --git a/cmd/proxy/cmd/metrics.go b/cmd/proxy/cmd/metrics.go
new file mode 100644
index 000000000..206b6bb7b
--- /dev/null
+++ b/cmd/proxy/cmd/metrics.go
@@ -0,0 +1,64 @@
+// Copyright 2021 Sorint.lab
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package cmd
+
+import (
+	"github.com/prometheus/client_golang/prometheus"
+)
+
+var (
+	proxyHealthGauge = prometheus.NewGauge(
+		prometheus.GaugeOpts{
+			Name: "stolon_proxy_health",
+			Help: "Set to 1 if proxy healthy and accepting connections",
+		},
+	)
+
+	clusterdataLastValidUpdateSeconds = prometheus.NewGauge(
+		prometheus.GaugeOpts{
+			Name: "stolon_proxy_clusterdata_last_valid_update_seconds",
+			Help: "Last time we received a valid clusterdata from our store as seconds since unix epoch",
+		},
+	)
+
+	proxyListenerStartedSeconds = prometheus.NewGauge(
+		prometheus.GaugeOpts{
+			Name: "stolon_proxy_listener_started_seconds",
+			Help: "Last time we started the proxy listener as seconds since unix epoch",
+		},
+	)
+
+	getClusterInfoErrors = prometheus.NewCounter(
+		prometheus.CounterOpts{
+			Name: "stolon_proxy_get_cluster_info_errors",
+			Help: "Count of failed getting and parsing cluster info operationss",
+		},
+	)
+
+	updateProxyInfoErrors = prometheus.NewCounter(
+		prometheus.CounterOpts{
+			Name: "stolon_proxy_update_proxy_info_errors",
+			Help: "Count of update proxyInfo failures",
+		},
+	)
+)
+
+func init() {
+	prometheus.MustRegister(proxyHealthGauge)
+	prometheus.MustRegister(clusterdataLastValidUpdateSeconds)
+	prometheus.MustRegister(proxyListenerStartedSeconds)
+	prometheus.MustRegister(getClusterInfoErrors)
+	prometheus.MustRegister(updateProxyInfoErrors)
+}
diff --git a/cmd/proxy/cmd/proxy.go b/cmd/proxy/cmd/proxy.go
index 39f1f3b99..29711dd65 100644
--- a/cmd/proxy/cmd/proxy.go
+++ b/cmd/proxy/cmd/proxy.go
@@ -148,6 +148,8 @@ func (c *ClusterChecker) startPollonProxy() error {
 		c.endPollonProxyCh <- c.pp.Start()
 	}()
 
+	proxyHealthGauge.Set(1)
+	proxyListenerStartedSeconds.SetToCurrentTime()
 	return nil
 }
 
@@ -160,6 +162,7 @@ func (c *ClusterChecker) stopPollonProxy() {
 		c.pp = nil
 		c.listener.Close()
 		c.listener = nil
+		proxyHealthGauge.Set(0)
 	}
 }
 
@@ -190,6 +193,7 @@ func (c *ClusterChecker) SetProxyInfo(e store.Store, generation int64, proxyTime
 func (c *ClusterChecker) Check() error {
 	cd, _, err := c.e.GetClusterData(context.TODO())
 	if err != nil {
+		getClusterInfoErrors.Inc()
 		return fmt.Errorf("cannot get cluster data: %v", err)
 	}
 
@@ -206,13 +210,20 @@ func (c *ClusterChecker) Check() error {
 	}
 	if cd.FormatVersion != cluster.CurrentCDFormatVersion {
 		c.sendPollonConfData(pollon.ConfData{DestAddr: nil})
+		getClusterInfoErrors.Inc()
 		return fmt.Errorf("unsupported clusterdata format version: %d", cd.FormatVersion)
 	}
 	if err = cd.Cluster.Spec.Validate(); err != nil {
 		c.sendPollonConfData(pollon.ConfData{DestAddr: nil})
+		getClusterInfoErrors.Inc()
 		return fmt.Errorf("clusterdata validation failed: %v", err)
 	}
 
+	// Mark that the clusterdata we've received is valid. We'll use this metric to detect
+	// when our store is failing to serve a valid clusterdata, so it's important we only
+	// update the metric here.
+	clusterdataLastValidUpdateSeconds.SetToCurrentTime()
+
 	cdProxyCheckInterval := cd.Cluster.DefSpec().ProxyCheckInterval.Duration
 	cdProxyTimeout := cd.Cluster.DefSpec().ProxyTimeout.Duration
 
@@ -231,6 +242,7 @@ func (c *ClusterChecker) Check() error {
 		c.sendPollonConfData(pollon.ConfData{DestAddr: nil})
 		// ignore errors on setting proxy info
 		if err = c.SetProxyInfo(c.e, cluster.NoGeneration, proxyTimeout); err != nil {
+			updateProxyInfoErrors.Inc()
 			log.Errorw("failed to update proxyInfo", zap.Error(err))
 		} else {
 			// update proxyCheckinterval and proxyTimeout only if we successfully updated our proxy info
@@ -248,6 +260,7 @@ func (c *ClusterChecker) Check() error {
 		c.sendPollonConfData(pollon.ConfData{DestAddr: nil})
 		// ignore errors on setting proxy info
 		if err = c.SetProxyInfo(c.e, proxy.Generation, proxyTimeout); err != nil {
+			updateProxyInfoErrors.Inc()
 			log.Errorw("failed to update proxyInfo", zap.Error(err))
 		} else {
 			// update proxyCheckinterval and proxyTimeout only if we successfully updated our proxy info
@@ -271,6 +284,7 @@ func (c *ClusterChecker) Check() error {
 		// cannot ignore this error since the sentinel won't know that we exist
 		// and are sending connections to a master so, when electing a new
 		// master, it'll not wait for us to close connections to the old one.
+		updateProxyInfoErrors.Inc()
 		return fmt.Errorf("failed to update proxyInfo: %v", err)
 	} else {
 		// update proxyCheckinterval and proxyTimeout only if we successfully updated our proxy info