AWS Python SDK Guide

With boto3, your application continues to use the normal AWS service model and request dictionaries. Configure the endpoint for the process or for one client.

Start a local adapter

This guide’s bounded example uses S3 backed by Google Cloud Storage:

Install or update the Tensor9 CLI

Install the customer CLI with Homebrew, then confirm the binary is available:

brew tap tensor9ine/tensor9
brew trust tensor9ine/tensor9
brew install tensor9
tensor9 --version

If tensor9 is already installed, run brew upgrade tensor9 instead of the install command.

Authenticate with Tensor9

A Tensor9 API key is required. Use the key issued during onboarding, or email [email protected] to request one for your organization. Do not share your account or key with another customer or organization.If your secret manager has already supplied T9_API_KEY, skip the prompt. Otherwise, enter your key at a hidden prompt in the terminal where you will run tensor9:

Bash
Zsh

set +x
IFS= read -r -s -p 'Tensor9 API key: ' T9_API_KEY
printf '\n'
export T9_API_KEY

set +x
IFS= read -r -s 'T9_API_KEY?Tensor9 API key: '
printf '\n'
export T9_API_KEY

Check that the variable is set without printing it, then verify your Tensor9 identity:

: "${T9_API_KEY:?Set T9_API_KEY before continuing}"
tensor9 whoami

Confirm that the reported identity belongs to the intended customer account. Do not continue if authentication fails or the account is wrong. Keep shell tracing disabled while using the key. Do not paste the key into commands, shell profiles, source files, screenshots or support messages.See Install and Authenticate for the planned license lifecycle, entitlements and credential boundaries.

Inspect the S3 to Google Cloud Storage mapping

Ask Tensor9 to describe the exact directed service mapping used by this guide:

tensor9 explain \
  -origin aws::1.0.0::s3 \
  -target google::1.0.0::gcs \
  -fmt Human

Read the supported S3 operations and Google Cloud Storage limitations before starting the process. The mapping is directional: this command describes S3 requests adapted to Google Cloud Storage, not the reverse direction.

Prepare Google Cloud access

Authenticate with Google Cloud using its standard Application Default Credentials chain. The adapter uses that identity when it calls Google Cloud Storage.Use a dedicated test project and narrowly scoped credentials. Do not put credentials in the command line, source code, or screenshots.

Start the S3 service adapter

Run the S3 to Google Cloud Storage adapter as a foreground process:

tensor9 adapt svc run \
  --origin aws::1.0.0::s3 \
  --backend google::1.0.0::gcs

Leave this terminal open. This is the current foreground launcher, not the pending Docker launcher for local development mode. It calls your actual target-cloud services; it is separate from the in-memory emulator. Acquisition and license renewal require Tensor9 authentication. Keep the endpoint local to your test.

Copy the listening endpoint

When the command reports its listening endpoint, copy that exact value into a second terminal:

export T9_CLOUD_ADAPTER_ENDPOINT="<endpoint printed by tensor9>"

Keep the endpoint local to the test unless you have deliberately configured network access and authentication for another machine.

Run one bounded S3 request

Continue with this page’s client or tool instructions. Point only S3 at T9_CLOUD_ADAPTER_ENDPOINT, then perform one operation whose Google Cloud Storage effect you can verify independently.Check both sides: read the S3-shaped response and inspect the native object or bucket in Google Cloud Storage. A successful client response alone does not prove that the target state is correct.

Stop the local adapter

Return to the first terminal and press Control-C. Confirm the process exits before deleting the test object or bucket with Google Cloud’s native client.If the request failed, preserve the request ID, adapter endpoint, S3 operation name, Google Cloud request ID, and any x-t9-explain-* receipt headers before stopping the process.

Set it for the process

export AWS_ENDPOINT_URL_S3="$T9_CLOUD_ADAPTER_ENDPOINT"

Then create the client without endpoint_url:

import os
import boto3
from botocore.config import Config

session = boto3.Session(profile_name="adapter-test")
s3 = session.client(
    "s3",
    region_name=os.environ.get("AWS_REGION", "us-east-1"),
    config=Config(s3={"addressing_style": "path"}),
)

Set it on one client

: "${T9_CLOUD_ADAPTER_ENDPOINT:?start the local adapter and copy its endpoint first}"

import os
import boto3
from botocore.config import Config

session = boto3.Session(profile_name="adapter-test")
s3 = session.client(
    "s3",
    endpoint_url=os.environ["T9_CLOUD_ADAPTER_ENDPOINT"],
    region_name=os.environ.get("AWS_REGION", "us-east-1"),
    config=Config(
        s3={"addressing_style": "path"},
        retries={"mode": "standard", "max_attempts": 3},
        connect_timeout=5,
        read_timeout=30,
    ),
)

Use a separate Session or dependency-injected client for Cloud Adapter. Avoid changing a process-wide endpoint behind unrelated AWS clients. An explicit endpoint_url takes precedence over AWS_ENDPOINT_URL_S3.

Verify object behavior

import time

bucket = os.environ["TEST_BUCKET"]
key = f"python/{int(time.time())}-hello.txt"
expected = b"hello through Cloud Adapter\n"

s3.put_object(Bucket=bucket, Key=key, Body=expected, ContentType="text/plain")
head = s3.head_object(Bucket=bucket, Key=key)
downloaded = s3.get_object(Bucket=bucket, Key=key)["Body"].read()

assert downloaded == expected
assert head["ContentLength"] == len(expected)

s3.delete_object(Bucket=bucket, Key=key)

Inspect the corresponding native target object after put_object. If the application depends on metadata, ETags, version IDs, ranges, conditionals, or multipart uploads, assert those behaviors separately against the S3 target profile.

Distinguish service and transport failures

from botocore.exceptions import ClientError, EndpointConnectionError, ReadTimeoutError

try:
    s3.head_object(Bucket=bucket, Key="missing-key")
except ClientError as error:
    response = error.response
    print({
        "code": response.get("Error", {}).get("Code"),
        "status": response.get("ResponseMetadata", {}).get("HTTPStatusCode"),
        "request_id": response.get("ResponseMetadata", {}).get("RequestId"),
    })
except (EndpointConnectionError, ReadTimeoutError) as error:
    print({"transport_error": type(error).__name__, "message": str(error)})

A ClientError means an HTTP service response reached Botocore. A connection or read timeout may have no response metadata at all. After a timed-out write, check target state before deciding whether to retry.

Production checklist

Pin boto3 and Botocore versions together.
Keep the endpoint, region, retry policy, and credentials explicit.
Reuse clients across requests.
Do not log Authorization, signed URLs, secret values, or object bodies.
Capture ResponseMetadata and Cloud Adapter diagnostic headers.
Add integration tests for paginators, waiters, retries, and error codes used by the application.

Backend credentials are configured on Cloud Adapter. They do not belong in the Python application’s AWS session.

Tensor9 Cloud Adapter

How It Works

Getting Started

Configuration

Deployment

Support

Amazon Web Services

Google Cloud

Microsoft Azure

Making Trade-offs

Tuning

Debugging

Testing

Guides

Reference

AWS Python SDK Guide

Start a local adapter

Set it for the process

Set it on one client

Verify object behavior

Distinguish service and transport failures

Production checklist

​Start a local adapter

​Set it for the process

​Set it on one client

​Verify object behavior

​Distinguish service and transport failures

​Production checklist

Start a local adapter

Set it for the process

Set it on one client

Verify object behavior

Distinguish service and transport failures

Production checklist