[fix]: Improve error logging for metric insertion failures in dogstatsd

crates/dogstatsd/src/aggregator_service.rs

@@ -106,8 +106,10 @@ impl AggregatorService {
                 AggregatorCommand::InsertBatch(metrics) => {
                     let mut insert_errors = 0;
                     for metric in metrics {
+                        let metric_name = metric.name;
                         // The only possible error here is an overflow
-                        if let Err(_e) = self.aggregator.insert(metric) {
+                        if let Err(e) = self.aggregator.insert(metric) {
+                            warn!("Failed to insert metric '{metric_name}': {e:?}");
                             insert_errors += 1;
                         }
                     }
@@ -255,4 +257,42 @@ mod tests {
         handle.shutdown().expect("Failed to shutdown");
         service_task.await.expect("Service task failed");
     }
+
+    #[tokio::test]
+    async fn test_aggregator_service_insert_overflow() {
+        // Create an aggregator with very small capacity to trigger overflow
+        let (service, handle) =
+            AggregatorService::new(EMPTY_TAGS, 2).expect("Failed to create aggregator service");
+
+        let service_task = tokio::spawn(service.run());
+
+        // Create metrics with different names and tags to exceed max_context
+        let metrics = vec![
+            parse("metric1:1|c|#tag:a").expect("metric parse failed"),
+            parse("metric2:2|c|#tag:b").expect("metric parse failed"),
+            parse("metric3:3|c|#tag:c").expect("metric parse failed"),
+            parse("metric4:4|c|#tag:d").expect("metric parse failed"),
+            parse("metric5:5|c|#tag:e").expect("metric parse failed"),
+        ];
+
+        // Insert batch - some should fail due to overflow
+        handle
+            .insert_batch(metrics)
+            .expect("Failed to send insert batch");
+
+        // Flush and verify that not all metrics were inserted
+        let response = handle.flush().await.expect("Failed to flush");
+
+        // With max_context=2, we should only successfully insert 2 unique metrics
+        // The rest should fail with overflow errors (which are logged with metric names)
+        assert!(
+            response.series[0].series.len() <= 2,
+            "Expected at most 2 metrics to be inserted, got {}",
+            response.series[0].series.len()
+        );
+
+        // Shutdown the service
+        handle.shutdown().expect("Failed to shutdown");
+        service_task.await.expect("Service task failed");
+    }
 }