testbed/aws-powertools__powertools-lambda-python/docs/core/tracer.md

---
title: Tracer
description: Core utility
---

Tracer is an opinionated thin wrapper for [AWS X-Ray Python SDK](https://github.com/aws/aws-xray-sdk-python/){target="_blank" rel="nofollow"}.

![Tracer showcase](../media/tracer_utility_showcase.png)

## Key features

* Auto capture cold start as annotation, and responses or full exceptions as metadata
* Auto-disable when not running in AWS Lambda environment
* Support tracing async methods, generators, and context managers
* Auto patch supported modules by AWS X-Ray

## Getting started

???+ tip
    All examples shared in this documentation are available within the [project repository](https://github.com/aws-powertools/powertools-lambda-python/tree/develop/examples){target="_blank"}.

!!! note "Tracer relies on AWS X-Ray SDK over [OpenTelememetry Distro (ADOT)](https://aws-otel.github.io/docs/getting-started/lambda){target="_blank" rel="nofollow"} for optimal cold start (lower latency)."

### Install

!!! info "This is not necessary if you're installing Powertools for AWS Lambda (Python) via [Lambda Layer/SAR](../index.md#lambda-layer){target="_blank"}"

Add `aws-lambda-powertools[tracer]` as a dependency in your preferred tool: _e.g._, _requirements.txt_, _pyproject.toml_. This will ensure you have the required dependencies before using Tracer.

### Permissions

Before your use this utility, your AWS Lambda function [must have permissions](https://docs.aws.amazon.com/lambda/latest/dg/services-xray.html#services-xray-permissions){target="_blank"} to send traces to AWS X-Ray.

```yaml hl_lines="9 12" title="AWS Serverless Application Model (SAM) example"
--8<-- "examples/tracer/sam/template.yaml"
```

### Lambda handler

You can quickly start by initializing `Tracer` and use `capture_lambda_handler` decorator for your Lambda handler.

```python hl_lines="1 4 12" title="Tracing Lambda handler with capture_lambda_handler"
--8<-- "examples/tracer/src/capture_lambda_handler.py"
```

`capture_lambda_handler` performs these additional tasks to ease operations:

* Creates a `ColdStart` annotation to easily filter traces that have had an initialization overhead
* Creates a `Service` annotation if `service` parameter or `POWERTOOLS_SERVICE_NAME` is set
* Captures any response, or full exceptions generated by the handler, and include as tracing metadata

### Annotations & Metadata

**Annotations** are key-values associated with traces and indexed by AWS X-Ray. You can use them to filter traces and to create [Trace Groups](https://aws.amazon.com/about-aws/whats-new/2018/11/aws-xray-adds-the-ability-to-group-traces/){target="_blank"} to slice and dice your transactions.

```python hl_lines="8" title="Adding annotations with put_annotation method"
--8<-- "examples/tracer/src/put_trace_annotations.py"
```

**Metadata** are key-values also associated with traces but not indexed by AWS X-Ray. You can use them to add additional context for an operation using any native object.

```python hl_lines="19" title="Adding arbitrary metadata with put_metadata method"
--8<-- "examples/tracer/src/put_trace_metadata.py"
```

### Synchronous functions

You can trace synchronous functions using the `capture_method` decorator.

```python hl_lines="7" title="Tracing an arbitrary function with capture_method"
--8<-- "examples/tracer/src/capture_method.py"
```

???+ note "Note: Function responses are auto-captured and stored as JSON, by default."

	Use [capture_response](#disabling-response-auto-capture) parameter to override this behaviour.

    The serialization is performed by aws-xray-sdk via `jsonpickle` module. This can cause
    side effects for file-like objects like boto S3 <a href="https://botocore.amazonaws.com/v1/documentation/api/latest/reference/response.html#botocore.response.StreamingBody">`StreamingBody`</a>, where its response will be read only once during serialization.

### Asynchronous and generator functions

???+ warning
	We do not support asynchronous Lambda handler

You can trace asynchronous functions and generator functions (including context managers) using `capture_method`.

=== "capture_method_async.py"

    ```python hl_lines="9"
    --8<-- "examples/tracer/src/capture_method_async.py"
    ```

=== "capture_method_context_manager.py"

    ```python hl_lines="12-13"
    --8<-- "examples/tracer/src/capture_method_context_manager.py"
    ```

=== "capture_method_generators.py"

    ```python hl_lines="9"
    --8<-- "examples/tracer/src/capture_method_generators.py"
    ```

### Environment variables

The following environment variables are available to configure Tracer at a global scope:

| Setting               | Description                                      | Environment variable                 | Default |
|-----------------------|--------------------------------------------------|--------------------------------------|---------|
| **Disable Tracing**   | Explicitly disables all tracing.                 | `POWERTOOLS_TRACE_DISABLED`          | `false` |
| **Response Capture**  | Captures Lambda or method return as metadata.    | `POWERTOOLS_TRACER_CAPTURE_RESPONSE` | `true`  |
| **Exception Capture** | Captures Lambda or method exception as metadata. | `POWERTOOLS_TRACER_CAPTURE_ERROR`    | `true`  |

Both [`POWERTOOLS_TRACER_CAPTURE_RESPONSE`](#disabling-response-auto-capture) and [`POWERTOOLS_TRACER_CAPTURE_ERROR`](#disabling-exception-auto-capture) can be set on a per-method basis, consequently overriding the environment variable value.

## Advanced

### Patching modules

Tracer automatically patches all [supported libraries by X-Ray](https://docs.aws.amazon.com/xray/latest/devguide/xray-sdk-python-patching.html){target="_blank"} during initialization, by default. Underneath, AWS X-Ray SDK checks whether a supported library has been imported before patching.

If you're looking to shave a few microseconds, or milliseconds depending on your function memory configuration, you can patch specific modules using `patch_modules` param:

```python hl_lines="8" title="Example of explicitly patching requests only"
--8<-- "examples/tracer/src/patch_modules.py"
```

### Disabling response auto-capture

Use **`capture_response=False`** parameter in both `capture_lambda_handler` and `capture_method` decorators to instruct Tracer **not** to serialize function responses as metadata.

???+ info "Info: This is useful in three common scenarios"
    1. You might **return sensitive** information you don't want it to be added to your traces
    2. You might manipulate **streaming objects that can be read only once**; this prevents subsequent calls from being empty
    3. You might return **more than 64K** of data _e.g., `message too long` error_

=== "disable_capture_response.py"

    ```python hl_lines="8 15"
    --8<-- "examples/tracer/src/disable_capture_response.py"
    ```

=== "disable_capture_response_streaming_body.py"

    ```python hl_lines="19"
    --8<-- "examples/tracer/src/disable_capture_response_streaming_body.py"
    ```

### Disabling exception auto-capture

Use **`capture_error=False`** parameter in both `capture_lambda_handler` and `capture_method` decorators to instruct Tracer **not** to serialize exceptions as metadata.

???+ info
	Useful when returning sensitive information in exceptions/stack traces you don't control

```python hl_lines="16 26" title="Disabling exception auto-capture for tracing metadata"
--8<-- "examples/tracer/src/disable_capture_error.py"
```

### Ignoring certain HTTP endpoints

You might have endpoints you don't want requests to be traced, perhaps due to the volume of calls or sensitive URLs.

You can use `ignore_endpoint` method with the hostname and/or URLs you'd like it to be ignored - globs (`*`) are allowed.

```python hl_lines="12-13" title="Ignoring certain HTTP endpoints from being traced"
--8<-- "examples/tracer/src/ignore_endpoints.py"
```

### Tracing aiohttp requests

???+ info
	This snippet assumes you have aiohttp as a dependency

You can use `aiohttp_trace_config` function to create a valid [aiohttp trace_config object](https://docs.aiohttp.org/en/stable/tracing_reference.html){target="_blank" rel="nofollow"}. This is necessary since X-Ray utilizes [aiohttp](https://docs.aiohttp.org/en/stable/){target="_blank" rel="nofollow"} trace hooks to capture requests end-to-end.

```python hl_lines="7 17" title="Tracing aiohttp requests"
--8<-- "examples/tracer/src/tracing_aiohttp.py"
```

### Escape hatch mechanism

You can use `tracer.provider` attribute to access all methods provided by AWS X-Ray `xray_recorder` object.

This is useful when you need a feature available in X-Ray that is not available in the Tracer utility, for example [thread-safe](https://github.com/aws/aws-xray-sdk-python/#user-content-trace-threadpoolexecutor){target="_blank"}, or [context managers](https://github.com/aws/aws-xray-sdk-python/#user-content-start-a-custom-segmentsubsegment){target="_blank"}.

```python hl_lines="14" title="Tracing a code block with in_subsegment escape hatch"
--8<-- "examples/tracer/src/sdk_escape_hatch.py"
```

### Concurrent asynchronous functions

???+ warning
	[X-Ray SDK will raise an exception](https://github.com/aws/aws-xray-sdk-python/issues/164){target="_blank"} when async functions are run and traced concurrently

A safe workaround mechanism is to use `in_subsegment_async` available via Tracer escape hatch (`tracer.provider`).

```python hl_lines="10 17 24" title="Workaround to safely trace async concurrent functions"
--8<-- "examples/tracer/src/capture_method_async_concurrency.py"
```

### Reusing Tracer across your code

Tracer keeps a copy of its configuration after the first initialization. This is useful for scenarios where you want to use Tracer in more than one location across your code base.

???+ warning "Warning: Import order matters when using Lambda Layers or multiple modules"
    **Do not set `auto_patch=False`** when reusing Tracer in Lambda Layers, or in multiple modules.

    This can result in the first Tracer config being inherited by new instances, and their modules not being patched.

	Tracer will automatically ignore imported modules that have been patched.

=== "tracer_reuse.py"

    ```python hl_lines="1 6"
    --8<-- "examples/tracer/src/tracer_reuse.py"
    ```

=== "tracer_reuse_module.py"
    A new instance of Tracer will be created but will reuse the previous Tracer instance configuration, similar to a Singleton.

    ```python hl_lines="3"
    --8<-- "examples/tracer/src/tracer_reuse_module.py"
    ```

## Testing your code

Tracer is disabled by default when not running in the AWS Lambda environment, including AWS SAM CLI and Chalice environments. This means no code changes or environment variables to be set.

## Tips

* Use annotations on key operations to slice and dice traces, create unique views, and create metrics from it via Trace Groups
* Use a namespace when adding metadata to group data more easily
* Annotations and metadata are added to the current subsegment opened. If you want them in a specific subsegment, use a [context manager](https://github.com/aws/aws-xray-sdk-python/#start-a-custom-segmentsubsegment){target="_blank"} via the escape hatch mechanism