diff --git a/dbms/src/Interpreters/InterpreterSelectQuery.cpp b/dbms/src/Interpreters/InterpreterSelectQuery.cpp
index 7c1f76f99e9..5a0230e6798 100644
--- a/dbms/src/Interpreters/InterpreterSelectQuery.cpp
+++ b/dbms/src/Interpreters/InterpreterSelectQuery.cpp
@@ -380,7 +380,7 @@ BlockIO InterpreterSelectQuery::execute()
 {
     Pipeline pipeline;
     executeImpl(pipeline, input);
-    executeUnion(pipeline);
+    executeUnion(pipeline, getSampleBlock());
 
     BlockIO res;
     res.in = pipeline.firstStream();
@@ -391,6 +391,7 @@ BlockInputStreams InterpreterSelectQuery::executeWithMultipleStreams()
 {
     Pipeline pipeline;
     executeImpl(pipeline, input);
+    unifyStreams(pipeline, getSampleBlock());
     return pipeline.streams;
 }
 
@@ -2221,12 +2222,12 @@ void InterpreterSelectQuery::executeDistinct(QueryPipeline & pipeline, bool befo
 }
 
 
-void InterpreterSelectQuery::executeUnion(Pipeline & pipeline)
+void InterpreterSelectQuery::executeUnion(Pipeline & pipeline, Block header)
 {
     /// If there are still several streams, then we combine them into one
     if (pipeline.hasMoreThanOneStream())
     {
-        unifyStreams(pipeline);
+        unifyStreams(pipeline, header);
 
         pipeline.firstStream() = std::make_shared<UnionBlockInputStream>(pipeline.streams, pipeline.stream_with_non_joined_data, max_streams);
         pipeline.stream_with_non_joined_data = nullptr;
@@ -2454,18 +2455,27 @@ void InterpreterSelectQuery::executeSubqueriesInSetsAndJoins(QueryPipeline & pip
 }
 
 
-void InterpreterSelectQuery::unifyStreams(Pipeline & pipeline)
+void InterpreterSelectQuery::unifyStreams(Pipeline & pipeline, Block header)
 {
-    if (pipeline.hasMoreThanOneStream())
+    if (pipeline.hasMoreThanOneStream() || header)
     {
         /// Unify streams in case they have different headers.
-        auto first_header = pipeline.streams.at(0)->getHeader();
-        for (size_t i = 1; i < pipeline.streams.size(); ++i)
+
+        size_t start = 0;
+
+        if (!header)
+        {
+            header = pipeline.streams.at(0)->getHeader();
+            start = 1;
+        }
+
+        for (size_t i = start; i < pipeline.streams.size(); ++i)
         {
             auto & stream = pipeline.streams[i];
-            auto header = stream->getHeader();
+            auto stream_header = stream->getHeader();
             auto mode = ConvertingBlockInputStream::MatchColumnsMode::Name;
-            if (!blocksHaveEqualStructure(first_header, header))
+
+            if (!blocksHaveEqualStructure(header, stream_header))
                 stream = std::make_shared<ConvertingBlockInputStream>(context, stream, first_header, mode);
         }
     }
diff --git a/dbms/src/Interpreters/InterpreterSelectQuery.h b/dbms/src/Interpreters/InterpreterSelectQuery.h
index d71bd310ee6..7c7b657f5ce 100644
--- a/dbms/src/Interpreters/InterpreterSelectQuery.h
+++ b/dbms/src/Interpreters/InterpreterSelectQuery.h
@@ -206,7 +206,7 @@ private:
     void executeOrder(Pipeline & pipeline, SortingInfoPtr sorting_info);
     void executeMergeSorted(Pipeline & pipeline);
     void executePreLimit(Pipeline & pipeline);
-    void executeUnion(Pipeline & pipeline);
+    void executeUnion(Pipeline & pipeline, Block header = {});
     void executeLimitBy(Pipeline & pipeline);
     void executeLimit(Pipeline & pipeline);
     void executeProjection(Pipeline & pipeline, const ExpressionActionsPtr & expression);
@@ -231,7 +231,7 @@ private:
     void executeSubqueriesInSetsAndJoins(QueryPipeline & pipeline, std::unordered_map<String, SubqueryForSet> & subqueries_for_sets);
 
     /// If pipeline has several streams with different headers, add ConvertingBlockInputStream to first header.
-    void unifyStreams(Pipeline & pipeline);
+    void unifyStreams(Pipeline & pipeline, Block header);
 
     enum class Modificator
     {