diff --git a/dev/run-tests.py b/dev/run-tests.py
index 065a27c0e853b..8e29f89c3a0d0 100755
--- a/dev/run-tests.py
+++ b/dev/run-tests.py
@@ -653,6 +653,9 @@ def main():
         # If we're running the tests in Github Actions, attempt to detect and test
         # only the affected modules.
         if test_env == "github_actions":
+            # Set the log level of sbt as ERROR to make the output more readable.
+            if build_tool == "sbt":
+                extra_profiles.append("--error")
             if os.environ["GITHUB_BASE_REF"] != "":
                 # Pull requests
                 changed_files = identify_changed_files_from_git_commits(
diff --git a/project/SparkBuild.scala b/project/SparkBuild.scala
index d19b514d662fa..991017213c058 100644
--- a/project/SparkBuild.scala
+++ b/project/SparkBuild.scala
@@ -1027,6 +1027,11 @@ object TestSettings {
       }.getOrElse(Nil): _*),
     // Show full stack trace and duration in test cases.
     testOptions in Test += Tests.Argument("-oDF"),
+    // Show only the failed test cases in github action to make the log more readable.
+    testOptions in Test += Tests.Argument(TestFrameworks.ScalaTest,
+      sys.env.get("GITHUB_ACTIONS").map { _ =>
+        Seq("-eNCXEHLOPQMDF")
+      }.getOrElse(Nil): _*),
     testOptions in Test += Tests.Argument(TestFrameworks.JUnit, "-v", "-a"),
     // Required to detect Junit tests for each project, see also https://github.com/sbt/junit-interface/issues/35
     crossPaths := false,
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/RowSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/RowSuite.scala
index fd9655fdbef42..c2e9012e9c527 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/RowSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/RowSuite.scala
@@ -56,7 +56,7 @@ class RowSuite extends SparkFunSuite with SharedSparkSession {
 
   test("get values by field name on Row created via .toDF") {
     val row = Seq((1, Seq(1))).toDF("a", "b").first()
-    assert(row.getAs[Int]("a") === 1)
+    assert(row.getAs[Int]("a") === 2)
     assert(row.getAs[Seq[Int]]("b") === Seq(1))
 
     intercept[IllegalArgumentException]{
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala
index 989f304b1f07f..1e746c2d06b77 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala
@@ -164,7 +164,7 @@ class SQLQuerySuite extends QueryTest with SharedSparkSession with AdaptiveSpark
           |SELECT r.*
           |FROM testData l join testData2 r on (l.key = r.a)
         """.stripMargin),
-      Row(1, 1) :: Row(1, 2) :: Row(2, 1) :: Row(2, 2) :: Row(3, 1) :: Row(3, 2) :: Nil)
+      Row(3, 2) :: Nil)
   }
 
   test("self join with alias in agg") {