Add timeout to endpointset metric collector

We have seen deadlocks with endpoint discovery caused by the metric
collector hanging and not releasing the store labels lock. This causes
the endpoint update to hang, which also makes all endpoint readers hang on
acquiring a read lock for the resolved endpoints slice.

This commit makes sure the Collect method on the metrics collector has
a built in timeout to guard against cases where an upstream call never
reads from the collection channel.

Signed-off-by: Filip Petkovski <filip.petkovsky@gmail.com>

pkg/query/endpointset.go

@@ -202,15 +202,17 @@ type endpointSetNodeCollector struct {
 	storeNodes      map[component.Component]map[string]int
 	storePerExtLset map[string]int
 
+	logger          log.Logger
 	connectionsDesc *prometheus.Desc
 	labels          []string
 }
 
-func newEndpointSetNodeCollector(labels ...string) *endpointSetNodeCollector {
+func newEndpointSetNodeCollector(logger log.Logger, labels ...string) *endpointSetNodeCollector {
 	if len(labels) == 0 {
 		labels = []string{string(ExternalLabels), string(StoreType)}
 	}
 	return &endpointSetNodeCollector{
+		logger:     logger,
 		storeNodes: map[component.Component]map[string]int{},
 		connectionsDesc: prometheus.NewDesc(
 			"thanos_store_nodes_grpc_connections",
@@ -277,7 +279,12 @@ func (c *endpointSetNodeCollector) Collect(ch chan<- prometheus.Metric) {
 					lbls = append(lbls, storeTypeStr)
 				}
 			}
-			ch <- prometheus.MustNewConstMetric(c.connectionsDesc, prometheus.GaugeValue, float64(occurrences), lbls...)
+			select {
+			case ch <- prometheus.MustNewConstMetric(c.connectionsDesc, prometheus.GaugeValue, float64(occurrences), lbls...):
+			case <-time.After(1 * time.Second):
+				level.Warn(c.logger).Log("msg", "failed to collect endpointset metrics", "timeout", 1*time.Second)
+				return
+			}
 		}
 	}
 }
@@ -319,7 +326,7 @@ func NewEndpointSet(
 	endpointInfoTimeout time.Duration,
 	endpointMetricLabels ...string,
 ) *EndpointSet {
-	endpointsMetric := newEndpointSetNodeCollector(endpointMetricLabels...)
+	endpointsMetric := newEndpointSetNodeCollector(logger, endpointMetricLabels...)
 	if reg != nil {
 		reg.MustRegister(endpointsMetric)
 	}