diff --git a/weed/admin/dash/admin_server.go b/weed/admin/dash/admin_server.go
index fd9abe2e0d..27958dec6d 100644
--- a/weed/admin/dash/admin_server.go
+++ b/weed/admin/dash/admin_server.go
@@ -2309,17 +2309,25 @@ func collectCollectionStats(topologyInfo *master_pb.TopologyInfo) map[string]col
 
 	// Fold EC per-volume counts into the collection totals. fileCount is
 	// deduped via max across every node reporting shards for the volume;
-	// deleteCount is summed across the same nodes.
+	// deleteCount is summed across the same nodes, and a tombstone recorded
+	// on more than one shard holder can push that sum past fileCount. This
+	// is an expected accounting artefact that recurs on every refresh, so
+	// it is summarized once per pass rather than logged per volume.
+	skipped := 0
 	for vid, agg := range ecVolumeAgg {
 		data := collectionMap[agg.collection]
 		if agg.fileCount >= agg.deleteCount {
 			data.FileCount += int64(agg.fileCount - agg.deleteCount)
 		} else {
-			glog.Warningf("ec volume %d in collection %q: summed delete_count=%d exceeds file_count=%d; skipping object count",
+			skipped++
+			glog.V(2).Infof("ec volume %d in collection %q: summed delete_count=%d exceeds file_count=%d; skipping object count",
 				vid, agg.collection, agg.deleteCount, agg.fileCount)
 		}
 		collectionMap[agg.collection] = data
 	}
+	if skipped > 0 {
+		glog.V(1).Infof("%d ec volumes have summed delete_count above file_count; skipping object count for them", skipped)
+	}
 
 	return collectionMap
 }
diff --git a/weed/admin/dash/collect_collection_stats_test.go b/weed/admin/dash/collect_collection_stats_test.go
index 5451ba6369..2d26610e3e 100644
--- a/weed/admin/dash/collect_collection_stats_test.go
+++ b/weed/admin/dash/collect_collection_stats_test.go
@@ -185,6 +185,47 @@ func TestCollectCollectionStatsECFileAndDeleteCountAggregation(t *testing.T) {
 	}
 }
 
+// TestCollectCollectionStatsECDeleteCountAboveFileCount verifies that an EC
+// volume whose summed delete_count exceeds its file_count adds nothing to the
+// collection's object count, while other volumes in the collection still do.
+func TestCollectCollectionStatsECDeleteCountAboveFileCount(t *testing.T) {
+	shard := func(vid uint32, bits uint32, fileCount, deleteCount uint64) *master_pb.VolumeEcShardInformationMessage {
+		return &master_pb.VolumeEcShardInformationMessage{
+			Id:          vid,
+			Collection:  "bucket-c",
+			EcIndexBits: bits,
+			ShardSizes:  []int64{1},
+			FileCount:   fileCount,
+			DeleteCount: deleteCount,
+		}
+	}
+	node := func(shards ...*master_pb.VolumeEcShardInformationMessage) *master_pb.DataNodeInfo {
+		return &master_pb.DataNodeInfo{
+			DiskInfos: map[string]*master_pb.DiskInfo{"disk1": {EcShardInfos: shards}},
+		}
+	}
+
+	topo := &master_pb.TopologyInfo{
+		DataCenterInfos: []*master_pb.DataCenterInfo{
+			{
+				RackInfos: []*master_pb.RackInfo{
+					{
+						DataNodeInfos: []*master_pb.DataNodeInfo{
+							node(shard(21, 1<<0, 100, 60), shard(22, 1<<0, 50, 10)),
+							node(shard(21, 1<<1, 100, 60)),
+						},
+					},
+				},
+			},
+		},
+	}
+
+	got := collectCollectionStats(topo)["bucket-c"]
+	if got.FileCount != 40 {
+		t.Errorf("FileCount: got %d, want 40 (volume 21 skipped, volume 22 has 50-10)", got.FileCount)
+	}
+}
+
 // TestCollectCollectionStatsECFileCountMaxDedupe verifies that EC file_count
 // is taken as the max across reporting nodes rather than the first-seen
 // value. A node that has not yet finished loading .ecx reports file_count=0,