diff --git a/tests/integration/instant_benchmark.py b/tests/integration/instant_benchmark.py
index 3d0f3ed0dc..365c421894 100644
--- a/tests/integration/instant_benchmark.py
+++ b/tests/integration/instant_benchmark.py
@@ -247,14 +247,16 @@ def build_running_script(template, job, instance, record):
                           job_template['requirements'], job_template['env'])
     container = job_template['container']
 
-    benchmark_command = []
+    benchmark_command = ['set -x']
     record_benchmark = ('python3 record_benchmark.py --template template.json '
                         f'--job {job} --instance {instance} '
                         f'--model models/test --record {record}')
 
     for key, value in job_template['awscurl'].items():
+        benchmark_command.append("rm -rf benchmark_result.json benchmark.log")
         benchmark_command.append(value)
-        benchmark_command.append(record_benchmark + f' --benchmark-vars "{key}"')
+        benchmark_command.append(record_benchmark +
+                                 f' --benchmark-vars "{key}"')
 
     bash_command = [
         'set -euo pipefail',
diff --git a/tests/integration/record_benchmark.py b/tests/integration/record_benchmark.py
index df6cdcaeba..a7a8697f19 100755
--- a/tests/integration/record_benchmark.py
+++ b/tests/integration/record_benchmark.py
@@ -124,7 +124,9 @@ def data_basic(data: dict):
 def data_from_client(data: dict):
     if os.path.exists("benchmark_result.json"):
         with open("benchmark_result.json", "r") as f:
-            data.update(json.load(f))
+            benchmark_metrics = json.load(f)
+            data.update(benchmark_metrics)
+            print(f"found awscurl metrics json: {benchmark_metrics}")
     elif os.path.exists("benchmark.log"):
         with open("benchmark.log", "r") as f:
             for line in f.readlines():