diff --git a/airbyte-config/init/src/main/resources/seed/source_definitions.yaml b/airbyte-config/init/src/main/resources/seed/source_definitions.yaml
index 785db71a4241..ee2395a9c285 100644
--- a/airbyte-config/init/src/main/resources/seed/source_definitions.yaml
+++ b/airbyte-config/init/src/main/resources/seed/source_definitions.yaml
@@ -271,7 +271,7 @@
 - name: File
   sourceDefinitionId: 778daa7c-feaf-4db6-96f3-70fd645acc77
   dockerRepository: airbyte/source-file
-  dockerImageTag: 0.2.16
+  dockerImageTag: 0.2.17
   documentationUrl: https://docs.airbyte.io/integrations/sources/file
   icon: file.svg
   sourceType: file
diff --git a/airbyte-config/init/src/main/resources/seed/source_specs.yaml b/airbyte-config/init/src/main/resources/seed/source_specs.yaml
index 079dad4ead84..4c45cc95ad2c 100644
--- a/airbyte-config/init/src/main/resources/seed/source_specs.yaml
+++ b/airbyte-config/init/src/main/resources/seed/source_specs.yaml
@@ -2255,7 +2255,7 @@
     supportsNormalization: false
     supportsDBT: false
     supported_destination_sync_modes: []
-- dockerImage: "airbyte/source-file:0.2.16"
+- dockerImage: "airbyte/source-file:0.2.17"
   spec:
     documentationUrl: "https://docs.airbyte.io/integrations/sources/file"
     connectionSpecification:
diff --git a/airbyte-integrations/connectors/source-file-secure/Dockerfile b/airbyte-integrations/connectors/source-file-secure/Dockerfile
index dcb20379a8b8..17f2d54f0d66 100644
--- a/airbyte-integrations/connectors/source-file-secure/Dockerfile
+++ b/airbyte-integrations/connectors/source-file-secure/Dockerfile
@@ -1,4 +1,4 @@
-FROM airbyte/source-file:0.2.16
+FROM airbyte/source-file:0.2.17
 
 WORKDIR /airbyte/integration_code
 COPY source_file_secure ./source_file_secure
@@ -9,5 +9,5 @@ RUN pip install .
 ENV AIRBYTE_ENTRYPOINT "python /airbyte/integration_code/main.py"
 ENTRYPOINT ["python", "/airbyte/integration_code/main.py"]
 
-LABEL io.airbyte.version=0.2.16
+LABEL io.airbyte.version=0.2.17
 LABEL io.airbyte.name=airbyte/source-file-secure
diff --git a/airbyte-integrations/connectors/source-file/Dockerfile b/airbyte-integrations/connectors/source-file/Dockerfile
index 9cbf49648e62..f91a6e7171b5 100644
--- a/airbyte-integrations/connectors/source-file/Dockerfile
+++ b/airbyte-integrations/connectors/source-file/Dockerfile
@@ -17,5 +17,5 @@ COPY source_file ./source_file
 ENV AIRBYTE_ENTRYPOINT "python /airbyte/integration_code/main.py"
 ENTRYPOINT ["python", "/airbyte/integration_code/main.py"]
 
-LABEL io.airbyte.version=0.2.16
+LABEL io.airbyte.version=0.2.17
 LABEL io.airbyte.name=airbyte/source-file
diff --git a/airbyte-integrations/connectors/source-file/setup.py b/airbyte-integrations/connectors/source-file/setup.py
index 352340b965f0..8ec1aba3f4b7 100644
--- a/airbyte-integrations/connectors/source-file/setup.py
+++ b/airbyte-integrations/connectors/source-file/setup.py
@@ -13,6 +13,7 @@
     "pandas==1.4.3",
     "paramiko==2.11.0",
     "s3fs==2022.7.1",
+    "boto3==1.21.21",
     "smart-open[all]==6.0.0",
     "lxml==4.9.1",
     "html5lib==1.1",
@@ -23,7 +24,7 @@
     "pyxlsb==1.0.9",
 ]
 
-TEST_REQUIREMENTS = ["boto3==1.21.21", "pytest==7.1.2", "pytest-docker==1.0.0", "pytest-mock~=3.8.2"]
+TEST_REQUIREMENTS = ["pytest~=6.2", "pytest-docker==1.0.0", "pytest-mock~=3.6.1"]
 
 setup(
     name="source_file",
diff --git a/airbyte-integrations/connectors/source-file/source_file/client.py b/airbyte-integrations/connectors/source-file/source_file/client.py
index 02e256cb5b35..5e0c68ed3139 100644
--- a/airbyte-integrations/connectors/source-file/source_file/client.py
+++ b/airbyte-integrations/connectors/source-file/source_file/client.py
@@ -4,6 +4,7 @@
 
 
 import json
+import tempfile
 import traceback
 from os import environ
 from typing import Iterable
@@ -352,10 +353,20 @@ def read(self, fields: Iterable = None) -> Iterable[dict]:
                 yield from df[columns].to_dict(orient="records")
             else:
                 fields = set(fields) if fields else None
+                if self.binary_source:
+                    fp = self._cache_stream(fp)
                 for df in self.load_dataframes(fp):
                     columns = fields.intersection(set(df.columns)) if fields else df.columns
                     df = df.where(pd.notnull(df), None)
-                    yield from df[columns].to_dict(orient="records")
+                    yield from df[list(columns)].to_dict(orient="records")
+
+    def _cache_stream(self, fp):
+        """cache stream to file"""
+        fp_tmp = tempfile.TemporaryFile(mode="w+b")
+        fp_tmp.write(fp.read())
+        fp_tmp.seek(0)
+        fp.close()
+        return fp_tmp
 
     def _stream_properties(self, fp):
         if self._reader_format == "yaml":
diff --git a/docs/integrations/sources/file.md b/docs/integrations/sources/file.md
index 77f7ef18ea9e..2578eacc5bd0 100644
--- a/docs/integrations/sources/file.md
+++ b/docs/integrations/sources/file.md
@@ -127,6 +127,7 @@ In order to read large files from a remote location, this connector uses the [sm
 
 | Version | Date       | Pull Request                                             | Subject                                           |
 |---------|------------|----------------------------------------------------------|---------------------------------------------------|
+| 0.2.17  | 2022-08-11 | [15501](https://github.com/airbytehq/airbyte/pull/15501) | Cache binary stream to file                       |
 | 0.2.16  | 2022-08-10 | [15293](https://github.com/airbytehq/airbyte/pull/15293) | added support for encoding reader option          |
 | 0.2.15  | 2022-08-05 | [15269](https://github.com/airbytehq/airbyte/pull/15269) | Bump `smart-open` version to 6.0.0                |
 | 0.2.12  | 2022-07-12 | [14535](https://github.com/airbytehq/airbyte/pull/14535) | Fix invalid schema generation for JSON files      |