Skip to content

Logger

Logger provides an opinionated logger with output structured as JSON.

Key features

  • Capture key fields from Lambda context, cold start and structures logging output as JSON
  • Log Lambda event when instructed (disabled by default)
  • Log sampling enables DEBUG log level for a percentage of requests (disabled by default)
  • Append additional keys to structured log at any point in time

Getting started

Tip

All examples shared in this documentation are available within the project repository.

Logger requires two settings:

Setting Description Environment variable Constructor parameter
Logging level Sets how verbose Logger should be (INFO, by default) LOG_LEVEL level
Service Sets service key that will be present across all log statements POWERTOOLS_SERVICE_NAME service
AWS Serverless Application Model (SAM) example
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
AWSTemplateFormatVersion: "2010-09-09"
Transform: AWS::Serverless-2016-10-31
Description: AWS Lambda Powertools Tracer doc examples

Globals:
  Function:
    Timeout: 5
    Runtime: python3.10
    Tracing: Active
    Environment:
      Variables:
        POWERTOOLS_SERVICE_NAME: payment
        LOG_LEVEL: INFO
    Layers:
      # Find the latest Layer version in the official documentation
      # https://awslabs.github.io/aws-lambda-powertools-python/latest/#lambda-layer
      - !Sub arn:aws:lambda:${AWS::Region}:017000801446:layer:AWSLambdaPowertoolsPythonV2:32

Resources:
  LoggerLambdaHandlerExample:
    Type: AWS::Serverless::Function
    Properties:
      CodeUri: ../src
      Handler: inject_lambda_context.handler

Standard structured keys

Your Logger will include the following keys to your structured logging:

Key Example Note
level: str INFO Logging level
location: str collect.handler:1 Source code location where statement was executed
message: Any Collecting payment Unserializable JSON values are casted as str
timestamp: str 2021-05-03 10:20:19,650+0200 Timestamp with milliseconds, by default uses local timezone
service: str payment Service name defined, by default service_undefined
xray_trace_id: str 1-5759e988-bd862e3fe1be46a994272793 When tracing is enabled, it shows X-Ray Trace ID
sampling_rate: float 0.1 When enabled, it shows sampling rate in percentage e.g. 10%
exception_name: str ValueError When logger.exception is used and there is an exception
exception: str Traceback (most recent call last).. When logger.exception is used and there is an exception

Capturing Lambda context info

You can enrich your structured logs with key Lambda context information via inject_lambda_context.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


@logger.inject_lambda_context
def lambda_handler(event: dict, context: LambdaContext) -> str:
    logger.info("Collecting payment")

    # You can log entire objects too
    logger.info({"operation": "collect_payment", "charge_id": event["charge_id"]})
    return "hello world"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
[
    {
        "level": "INFO",
        "location": "collect.handler:9",
        "message": "Collecting payment",
        "timestamp": "2021-05-03 11:47:12,494+0200",
        "service": "payment",
        "cold_start": true,
        "function_name": "test",
        "function_memory_size": 128,
        "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
        "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72"
    },
    {
        "level": "INFO",
        "location": "collect.handler:12",
        "message": {
            "operation": "collect_payment",
            "charge_id": "ch_AZFlk2345C0"
        },
        "timestamp": "2021-05-03 11:47:12,494+0200",
        "service": "payment",
        "cold_start": true,
        "function_name": "test",
        "function_memory_size": 128,
        "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
        "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72"
    }
]

When used, this will include the following keys:

Key Example
cold_start: bool false
function_name str example-powertools-HelloWorldFunction-1P1Z6B39FLU73
function_memory_size: int 128
function_arn: str arn:aws:lambda:eu-west-1:012345678910:function:example-powertools-HelloWorldFunction-1P1Z6B39FLU73
function_request_id: str 899856cb-83d1-40d7-8611-9e78f15f32f4

Logging incoming event

When debugging in non-production environments, you can instruct Logger to log the incoming event with log_event param or via POWERTOOLS_LOGGER_LOG_EVENT env var.

Warning

This is disabled by default to prevent sensitive info being logged

Logging incoming event
1
2
3
4
5
6
7
8
9
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


@logger.inject_lambda_context(log_event=True)
def lambda_handler(event: dict, context: LambdaContext) -> str:
    return "hello world"

Setting a Correlation ID

You can set a Correlation ID using correlation_id_path param by passing a JMESPath expression.

Tip

You can retrieve correlation IDs via get_correlation_id method

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


@logger.inject_lambda_context(correlation_id_path="headers.my_request_id_header")
def lambda_handler(event: dict, context: LambdaContext) -> str:
    logger.debug(f"Correlation ID => {logger.get_correlation_id()}")
    logger.info("Collecting payment")

    return "hello world"

1
2
3
4
5
{
    "headers": {
        "my_request_id_header": "correlation_id_value"
    }
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
{
    "level": "INFO",
    "location": "collect.handler:10",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "cold_start": true,
    "function_name": "test",
    "function_memory_size": 128,
    "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
    "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72",
    "correlation_id": "correlation_id_value"
}

set_correlation_id method

You can also use set_correlation_id method to inject it anywhere else in your code. Example below uses Event Source Data Classes utility to easily access events properties.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.data_classes import APIGatewayProxyEvent
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


def lambda_handler(event: dict, context: LambdaContext) -> str:
    request = APIGatewayProxyEvent(event)

    logger.set_correlation_id(request.request_context.request_id)
    logger.info("Collecting payment")

    return "hello world"

1
2
3
4
5
{
    "requestContext": {
        "requestId": "correlation_id_value"
    }
}

1
2
3
4
5
6
7
8
{
    "level": "INFO",
    "location": "collect.handler:13",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "correlation_id": "correlation_id_value"
}

Known correlation IDs

To ease routine tasks like extracting correlation ID from popular event sources, we provide built-in JMESPath expressions.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
from aws_lambda_powertools import Logger
from aws_lambda_powertools.logging import correlation_paths
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


@logger.inject_lambda_context(correlation_id_path=correlation_paths.API_GATEWAY_REST)
def lambda_handler(event: dict, context: LambdaContext) -> str:
    logger.debug(f"Correlation ID => {logger.get_correlation_id()}")
    logger.info("Collecting payment")

    return "hello world"

1
2
3
4
5
{
    "requestContext": {
        "requestId": "correlation_id_value"
    }
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
{
    "level": "INFO",
    "location": "collect.handler:11",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "cold_start": true,
    "function_name": "test",
    "function_memory_size": 128,
    "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
    "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72",
    "correlation_id": "correlation_id_value"
}

Appending additional keys

Info: Custom keys are persisted across warm invocations

Always set additional keys as part of your handler to ensure they have the latest value, or explicitly clear them with clear_state=True.

You can append additional keys using either mechanism:

  • Persist new keys across all future log messages via append_keys method
  • Add additional keys on a per log message basis as a keyword=value, or via extra parameter

append_keys method

Warning

append_keys is not thread-safe, please see RFC.

You can append your own keys to your existing Logger via append_keys(**additional_key_values) method.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


def lambda_handler(event: dict, context: LambdaContext) -> str:
    order_id = event.get("order_id")

    # this will ensure order_id key always has the latest value before logging
    # alternative, you can use `clear_state=True` parameter in @inject_lambda_context
    logger.append_keys(order_id=order_id)
    logger.info("Collecting payment")

    return "hello world"

1
2
3
4
5
6
7
8
{
    "level": "INFO",
    "location": "collect.handler:11",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "order_id": "order_id_value"
}
Tip: Logger will automatically reject any key with a None value

If you conditionally add keys depending on the payload, you can follow the example above.

This example will add order_id if its value is not empty, and in subsequent invocations where order_id might not be present it'll remove it from the Logger.

ephemeral metadata

You can pass an arbitrary number of keyword arguments (kwargs) to all log level's methods, e.g. logger.info, logger.warning.

Two common use cases for this feature is to enrich log statements with additional metadata, or only add certain keys conditionally.

Any keyword argument added will not be persisted in subsequent messages.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


def lambda_handler(event: dict, context: LambdaContext) -> str:
    logger.info("Collecting payment", request_id="1123")

    return "hello world"

1
2
3
4
5
6
7
8
{
    "level": "INFO",
    "location": "collect.handler:8",
    "message": "Collecting payment",
    "timestamp": "2022-11-26 11:47:12,494+0200",
    "service": "payment",
    "request_id": "1123"
}

extra parameter

Extra parameter is available for all log levels' methods, as implemented in the standard logging library - e.g. logger.info, logger.warning.

It accepts any dictionary, and all keyword arguments will be added as part of the root structure of the logs for that log statement.

Any keyword argument added using extra will not be persisted in subsequent messages.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


def lambda_handler(event: dict, context: LambdaContext) -> str:
    fields = {"request_id": "1123"}
    logger.info("Collecting payment", extra=fields)

    return "hello world"

1
2
3
4
5
6
7
8
{
    "level": "INFO",
    "location": "collect.handler:9",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "request_id": "1123"
}

Removing additional keys

You can remove any additional key from Logger state using remove_keys.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


def lambda_handler(event: dict, context: LambdaContext) -> str:
    logger.append_keys(sample_key="value")
    logger.info("Collecting payment")

    logger.remove_keys(["sample_key"])
    logger.info("Collecting payment without sample key")

    return "hello world"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
[
    {
        "level": "INFO",
        "location": "collect.handler:9",
        "message": "Collecting payment",
        "timestamp": "2021-05-03 11:47:12,494+0200",
        "service": "payment",
        "sample_key": "value"
    },
    {
        "level": "INFO",
        "location": "collect.handler:12",
        "message": "Collecting payment without sample key",
        "timestamp": "2021-05-03 11:47:12,494+0200",
        "service": "payment"
    }
]

Clearing all state

Logger is commonly initialized in the global scope. Due to Lambda Execution Context reuse, this means that custom keys can be persisted across invocations. If you want all custom keys to be deleted, you can use clear_state=True param in inject_lambda_context decorator.

Tip: When is this useful?

It is useful when you add multiple custom keys conditionally, instead of setting a default None value if not present. Any key with None value is automatically removed by Logger.

Danger: This can have unintended side effects if you use Layers

Lambda Layers code is imported before the Lambda handler.

This means that clear_state=True will instruct Logger to remove any keys previously added before Lambda handler execution proceeds.

You can either avoid running any code as part of Lambda Layers global scope, or override keys with their latest value as part of handler's execution.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


@logger.inject_lambda_context(clear_state=True)
def lambda_handler(event: dict, context: LambdaContext) -> str:
    if event.get("special_key"):
        # Should only be available in the first request log
        # as the second request doesn't contain `special_key`
        logger.append_keys(debugging_key="value")

    logger.info("Collecting payment")

    return "hello world"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
{
    "level": "INFO",
    "location": "collect.handler:10",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "special_key": "debug_key",
    "cold_start": true,
    "function_name": "test",
    "function_memory_size": 128,
    "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
    "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72"
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
{
    "level": "INFO",
    "location": "collect.handler:10",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "cold_start": false,
    "function_name": "test",
    "function_memory_size": 128,
    "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
    "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72"
}

Logging exceptions

Use logger.exception method to log contextual information about exceptions. Logger will include exception_name and exception keys to aid troubleshooting and error enumeration.

Tip

You can use your preferred Log Analytics tool to enumerate and visualize exceptions across all your services using exception_name key.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
import requests

from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

ENDPOINT = "http://httpbin.org/status/500"
logger = Logger()


def lambda_handler(event: dict, context: LambdaContext) -> str:
    try:
        ret = requests.get(ENDPOINT)
        ret.raise_for_status()
    except requests.HTTPError as e:
        logger.exception("Received a HTTP 5xx error")
        raise RuntimeError("Unable to fullfil request") from e

    return "hello world"

1
2
3
4
5
6
7
8
9
{
    "level": "ERROR",
    "location": "collect.handler:15",
    "message": "Received a HTTP 5xx error",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "exception_name": "RuntimeError",
    "exception": "Traceback (most recent call last):\n  File \"<input>\", line 2, in <module> RuntimeError: Unable to fullfil request"
}

Uncaught exceptions

CAUTION: some users reported a problem that causes this functionality not to work in the Lambda runtime. We recommend that you don't use this feature for the time being.

Logger can optionally log uncaught exceptions by setting log_uncaught_exceptions=True at initialization.

Logger will replace any exception hook previously registered via sys.excepthook.

What are uncaught exceptions?

It's any raised exception that wasn't handled by the except statement, leading a Python program to a non-successful exit.

They are typically raised intentionally to signal a problem (raise ValueError), or a propagated exception from elsewhere in your code that you didn't handle it willingly or not (KeyError, jsonDecoderError, etc.).

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
import requests

from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

ENDPOINT = "http://httpbin.org/status/500"
logger = Logger(log_uncaught_exceptions=True)


def lambda_handler(event: dict, context: LambdaContext) -> str:
    ret = requests.get(ENDPOINT)
    # HTTP 4xx/5xx status will lead to requests.HTTPError
    # Logger will log this exception before this program exits non-successfully
    ret.raise_for_status()

    return "hello world"

1
2
3
4
5
6
7
8
9
{
    "level": "ERROR",
    "location": "log_uncaught_exception_hook:756",
    "message": "500 Server Error: INTERNAL SERVER ERROR for url: http://httpbin.org/status/500",
    "timestamp": "2022-11-16 13:51:29,198+0100",
    "service": "payment",
    "exception": "Traceback (most recent call last):\n  File \"<input>\", line 52, in <module>\n    handler({}, {})\n  File \"<input>\", line 17, in handler\n    ret.raise_for_status()\n  File \"<input>/lib/python3.9/site-packages/requests/models.py\", line 1021, in raise_for_status\n    raise HTTPError(http_error_msg, response=self)\nrequests.exceptions.HTTPError: 500 Server Error: INTERNAL SERVER ERROR for url: http://httpbin.org/status/500",
    "exception_name": "HTTPError"
}

Date formatting

Logger uses Python's standard logging date format with the addition of timezone: 2021-05-03 11:47:12,494+0200.

You can easily change the date format using one of the following parameters:

  • datefmt. You can pass any strftime format codes. Use %F if you need milliseconds.
  • use_rfc3339. This flag will use a format compliant with both RFC3339 and ISO8601: 2022-10-27T16:27:43.738+02:00
Prefer using datetime string formats?

Use use_datetime_directive flag along with datefmt to instruct Logger to use datetime instead of time.strftime.

1
2
3
4
5
6
7
8
9
from aws_lambda_powertools import Logger

date_format = "%m/%d/%Y %I:%M:%S %p"

logger = Logger(service="payment", use_rfc3339=True)
logger.info("Collecting payment")

logger_custom_format = Logger(service="loyalty", datefmt=date_format)
logger_custom_format.info("Calculating points")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
[
    {
        "level": "INFO",
        "location": "<module>:6",
        "message": "Collecting payment",
        "timestamp": "2022-10-28T14:35:03.210+02:00",
        "service": "payment"
    },
    {
        "level": "INFO",
        "location": "<module>:9",
        "message": "Calculating points",
        "timestamp": "10/28/2022 02:35:03 PM",
        "service": "loyalty"
    }
]

Advanced

Built-in Correlation ID expressions

You can use any of the following built-in JMESPath expressions as part of inject_lambda_context decorator.

Note: Any object key named with - must be escaped

For example, request.headers."x-amzn-trace-id".

Name Expression Description
API_GATEWAY_REST "requestContext.requestId" API Gateway REST API request ID
API_GATEWAY_HTTP "requestContext.requestId" API Gateway HTTP API request ID
APPSYNC_RESOLVER 'request.headers."x-amzn-trace-id"' AppSync X-Ray Trace ID
APPLICATION_LOAD_BALANCER 'headers."x-amzn-trace-id"' ALB X-Ray Trace ID
EVENT_BRIDGE "id" EventBridge Event ID

Reusing Logger across your code

Similar to Tracer, a new instance that uses the same service name - env var or explicit parameter - will reuse a previous Logger instance. Just like logging.getLogger("logger_name") would in the standard library if called with the same logger name.

Notice in the CloudWatch Logs output how payment_id appeared as expected when logging in collect.py.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
from logger_reuse_payment import inject_payment_id

from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


@logger.inject_lambda_context
def lambda_handler(event: dict, context: LambdaContext) -> str:
    inject_payment_id(context=event)
    logger.info("Collecting payment")
    return "hello world"

1
2
3
4
5
6
7
from aws_lambda_powertools import Logger

logger = Logger()


def inject_payment_id(context):
    logger.append_keys(payment_id=context.get("payment_id"))

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
{
    "level": "INFO",
    "location": "collect.handler:12",
    "message": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494+0200",
    "service": "payment",
    "cold_start": true,
    "function_name": "test",
    "function_memory_size": 128,
    "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
    "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72",
    "payment_id": "968adaae-a211-47af-bda3-eed3ca2c0ed0"
}
Note: About Child Loggers

Coming from standard library, you might be used to use logging.getLogger(__name__). This will create a new instance of a Logger with a different name.

In Powertools, you can have the same effect by using child=True parameter: Logger(child=True). This creates a new Logger instance named after service.<module>. All state changes will be propagated bi-directionally between Child and Parent.

For that reason, there could be side effects depending on the order the Child Logger is instantiated, because Child Loggers don't have a handler.

For example, if you instantiated a Child Logger and immediately used logger.append_keys/remove_keys/set_correlation_id to update logging state, this might fail if the Parent Logger wasn't instantiated.

In this scenario, you can either ensure any calls manipulating state are only called when a Parent Logger is instantiated (example above), or refrain from using child=True parameter altogether.

Sampling debug logs

Use sampling when you want to dynamically change your log level to DEBUG based on a percentage of your concurrent/cold start invocations.

You can use values ranging from 0.0 to 1 (100%) when setting POWERTOOLS_LOGGER_SAMPLE_RATE env var, or sample_rate parameter in Logger.

Tip: When is this useful?

Let's imagine a sudden spike increase in concurrency triggered a transient issue downstream. When looking into the logs you might not have enough information, and while you can adjust log levels it might not happen again.

This feature takes into account transient issues where additional debugging information can be useful.

Sampling decision happens at the Logger initialization. This means sampling may happen significantly more or less than depending on your traffic patterns, for example a steady low number of invocations and thus few cold starts.

Note

Open a feature request if you want Logger to calculate sampling for every invocation

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

# Sample 10% of debug logs e.g. 0.1
# NOTE: this evaluation will only occur at cold start
logger = Logger(service="payment", sample_rate=0.1)


def lambda_handler(event: dict, context: LambdaContext):
    logger.debug("Verifying whether order_id is present")
    logger.info("Collecting payment")

    return "hello world"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
[
    {
        "level": "DEBUG",
        "location": "collect.handler:7",
        "message": "Verifying whether order_id is present",
        "timestamp": "2021-05-03 11:47:12,494+0200",
        "service": "payment",
        "cold_start": true,
        "function_name": "test",
        "function_memory_size": 128,
        "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
        "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72",
        "sampling_rate": 0.1
    },
    {
        "level": "INFO",
        "location": "collect.handler:7",
        "message": "Collecting payment",
        "timestamp": "2021-05-03 11:47:12,494+0200",
        "service": "payment",
        "cold_start": true,
        "function_name": "test",
        "function_memory_size": 128,
        "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
        "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72",
        "sampling_rate": 0.1
    }
]

LambdaPowertoolsFormatter

Logger propagates a few formatting configurations to the built-in LambdaPowertoolsFormatter logging formatter.

If you prefer configuring it separately, or you'd want to bring this JSON Formatter to another application, these are the supported settings:

Parameter Description Default
json_serializer function to serialize obj to a JSON formatted str json.dumps
json_deserializer function to deserialize str, bytes, bytearray containing a JSON document to a Python obj json.loads
json_default function to coerce unserializable values, when no custom serializer/deserializer is set str
datefmt string directives (strftime) to format log timestamp %Y-%m-%d %H:%M:%S,%F%z, where %F is a custom ms directive
use_datetime_directive format the datefmt timestamps using datetime, not time (also supports the custom %F directive for milliseconds) False
utc set logging timestamp to UTC False
log_record_order set order of log keys when logging ["level", "location", "message", "timestamp"]
kwargs key-value to be included in log messages None
Info

When POWERTOOLS_DEV env var is present and set to "true", Logger's default serializer (json.dumps) will pretty-print log messages for easier readability.

Pre-configuring Lambda Powertools Formatter
1
2
3
4
5
6
7
8
from aws_lambda_powertools import Logger
from aws_lambda_powertools.logging.formatter import LambdaPowertoolsFormatter

# NOTE: Check docs for all available options
# https://awslabs.github.io/aws-lambda-powertools-python/latest/core/logger/#lambdapowertoolsformatter

formatter = LambdaPowertoolsFormatter(utc=True, log_record_order=["message"])
logger = Logger(service="example", logger_formatter=formatter)

Observability providers

In this context, an observability provider is an AWS Lambda Partner offering a platform for logging, metrics, traces, etc.

You can send logs to the observability provider of your choice via Lambda Extensions. In most cases, you shouldn't need any custom Logger configuration, and logs will be shipped async without any performance impact.

Built-in formatters

In rare circumstances where JSON logs are not parsed correctly by your provider, we offer built-in formatters to make this transition easier.

Provider Formatter Notes
Datadog DatadogLogFormatter Modifies default timestamp to use RFC3339 by default

You can use import and use them as any other Logger formatter via logger_formatter parameter:

Using built-in Logger Formatters
1
2
3
4
5
from aws_lambda_powertools import Logger
from aws_lambda_powertools.logging.formatters.datadog import DatadogLogFormatter

logger = Logger(service="payment", logger_formatter=DatadogLogFormatter())
logger.info("hello")

Migrating from other Loggers

If you're migrating from other Loggers, there are few key points to be aware of: Service parameter, Inheriting Loggers, Overriding Log records, and Logging exceptions.

The service parameter

Service is what defines the Logger name, including what the Lambda function is responsible for, or part of (e.g payment service).

For Logger, the service is the logging key customers can use to search log operations for one or more functions - For example, search for all errors, or messages like X, where service is payment.

Inheriting Loggers

Tip: Prefer Logger Reuse feature over inheritance unless strictly necessary, see caveats.

Python Logging hierarchy happens via the dot notation: service, service.child, service.child_2

For inheritance, Logger uses a child=True parameter along with service being the same value across Loggers.

For child Loggers, we introspect the name of your module where Logger(child=True, service="name") is called, and we name your Logger as {service}.{filename}.

Danger

A common issue when migrating from other Loggers is that service might be defined in the parent Logger (no child param), and not defined in the child Logger:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
from logging_inheritance_module import inject_payment_id

from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

# NOTE: explicit service name differs from Child
# meaning we will have two Logger instances with different state
# and an orphan child logger who won't be able to manipulate state
logger = Logger(service="payment")


@logger.inject_lambda_context
def lambda_handler(event: dict, context: LambdaContext) -> str:
    inject_payment_id(context=event)

    return "hello world"

1
2
3
4
5
6
7
from aws_lambda_powertools import Logger

logger = Logger(child=True)


def inject_payment_id(context):
    logger.append_keys(payment_id=context.get("payment_id"))

In this case, Logger will register a Logger named payment, and a Logger named service_undefined. The latter isn't inheriting from the parent, and will have no handler, resulting in no message being logged to standard output.

Tip

This can be fixed by either ensuring both has the service value as payment, or simply use the environment variable POWERTOOLS_SERVICE_NAME to ensure service value will be the same across all Loggers when not explicitly set.

Do this instead:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
from logging_inheritance_module import inject_payment_id

from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

# NOTE: explicit service name matches any new Logger
# because we're using POWERTOOLS_SERVICE_NAME env var
# but we could equally use the same string as service value, e.g. "payment"
logger = Logger()


@logger.inject_lambda_context
def lambda_handler(event: dict, context: LambdaContext) -> str:
    inject_payment_id(context=event)

    return "hello world"

1
2
3
4
5
6
7
from aws_lambda_powertools import Logger

logger = Logger(child=True)


def inject_payment_id(context):
    logger.append_keys(payment_id=context.get("payment_id"))

Overriding Log records

You might want to continue to use the same date formatting style, or override location to display the package.function_name:line_number as you previously had.

Logger allows you to either change the format or suppress the following keys at initialization: location, timestamp, xray_trace_id.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
from aws_lambda_powertools import Logger

location_format = "[%(funcName)s] %(module)s"

# override location and timestamp format
logger = Logger(service="payment", location=location_format)
logger.info("Collecting payment")

# suppress keys with a None value
logger_two = Logger(service="loyalty", location=None)
logger_two.info("Calculating points")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
[
    {
        "level": "INFO",
        "location": "[<module>] overriding_log_records",
        "message": "Collecting payment",
        "timestamp": "2022-10-28 14:40:43,801+0200",
        "service": "payment"
    },
    {
        "level": "INFO",
        "message": "Calculating points",
        "timestamp": "2022-10-28 14:40:43,801+0200",
        "service": "loyalty"
    }
]

Reordering log keys position

You can change the order of standard Logger keys or any keys that will be appended later at runtime via the log_record_order parameter.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
from aws_lambda_powertools import Logger

# make message as the first key
logger = Logger(service="payment", log_record_order=["message"])

# make request_id that will be added later as the first key
logger_two = Logger(service="order", log_record_order=["request_id"])
logger_two.append_keys(request_id="123")

logger.info("hello world")
logger_two.info("hello world")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
[
    {
        "message": "hello world",
        "level": "INFO",
        "location": "<module>:11",
        "timestamp": "2022-06-24 11:25:40,143+0200",
        "service": "payment"
    },
    {
        "request_id": "123",
        "level": "INFO",
        "location": "<module>:12",
        "timestamp": "2022-06-24 11:25:40,144+0200",
        "service": "order",
        "message": "hello universe"
    }
]

Setting timestamp to UTC

By default, this Logger and standard logging library emits records using local time timestamp. You can override this behavior via utc parameter:

1
2
3
4
5
6
7
from aws_lambda_powertools import Logger

logger = Logger(service="payment")
logger.info("Local time")

logger_in_utc = Logger(service="order", utc=True)
logger_in_utc.info("GMT time zone")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
[
    {
        "level": "INFO",
        "location": "<module>:4",
        "message": "Local time",
        "timestamp": "2022-06-24 11:39:49,421+0200",
        "service": "payment"
    },
    {
        "level": "INFO",
        "location": "<module>:7",
        "message": "GMT time zone",
        "timestamp": "2022-06-24 09:39:49,421+0100",
        "service": "order"
    }
]

Custom function for unserializable values

By default, Logger uses str to handle values non-serializable by JSON. You can override this behavior via json_default parameter by passing a Callable:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
from datetime import date, datetime

from aws_lambda_powertools import Logger


def custom_json_default(value: object) -> str:
    if isinstance(value, (datetime, date)):
        return value.isoformat()

    return f"<non-serializable: {type(value).__name__}>"


class Unserializable:
    pass


logger = Logger(service="payment", json_default=custom_json_default)

logger.info({"ingestion_time": datetime.utcnow(), "serialize_me": Unserializable()})

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
{
    "level": "INFO",
    "location": "<module>:19",
    "message": {
        "ingestion_time": "2022-06-24T10:12:09.526365",
        "serialize_me": "<non-serializable: Unserializable>"
    },
    "timestamp": "2022-06-24 12:12:09,526+0200",
    "service": "payment"
}

Bring your own handler

By default, Logger uses StreamHandler and logs to standard output. You can override this behavior via logger_handler parameter:

Configure Logger to output to a file
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
import logging
from pathlib import Path

from aws_lambda_powertools import Logger

log_file = Path("/tmp/log.json")
log_file_handler = logging.FileHandler(filename=log_file)

logger = Logger(service="payment", logger_handler=log_file_handler)

logger.info("hello world")

Bring your own formatter

By default, Logger uses LambdaPowertoolsFormatter that persists its custom structure between non-cold start invocations. There could be scenarios where the existing feature set isn't sufficient to your formatting needs.

Info

The most common use cases are remapping keys by bringing your existing schema, and redacting sensitive information you know upfront.

For these, you can override the serialize method from LambdaPowertoolsFormatter.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
from aws_lambda_powertools import Logger
from aws_lambda_powertools.logging.formatter import LambdaPowertoolsFormatter


class CustomFormatter(LambdaPowertoolsFormatter):
    def serialize(self, log: dict) -> str:
        """Serialize final structured log dict to JSON str"""
        log["event"] = log.pop("message")  # rename message key to event
        return self.json_serializer(log)  # use configured json serializer


logger = Logger(service="payment", logger_formatter=CustomFormatter())
logger.info("hello")

1
2
3
4
5
6
7
{
    "level": "INFO",
    "location": "<module>:16",
    "timestamp": "2021-12-30 13:41:53,413+0100",
    "service": "payment",
    "event": "hello"
}

The log argument is the final log record containing our standard keys, optionally Lambda context keys, and any custom key you might have added via append_keys or the extra parameter.

For exceptional cases where you want to completely replace our formatter logic, you can subclass BasePowertoolsFormatter.

Warning

You will need to implement append_keys, clear_state, override format, and optionally remove_keys to keep the same feature set Powertools Logger provides. This also means keeping state of logging keys added.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
import json
import logging
from typing import Iterable, List, Optional

from aws_lambda_powertools import Logger
from aws_lambda_powertools.logging.formatter import BasePowertoolsFormatter


class CustomFormatter(BasePowertoolsFormatter):
    def __init__(self, log_record_order: Optional[List[str]] = None, *args, **kwargs):
        self.log_record_order = log_record_order or ["level", "location", "message", "timestamp"]
        self.log_format = dict.fromkeys(self.log_record_order)
        super().__init__(*args, **kwargs)

    def append_keys(self, **additional_keys):
        # also used by `inject_lambda_context` decorator
        self.log_format.update(additional_keys)

    def remove_keys(self, keys: Iterable[str]):
        for key in keys:
            self.log_format.pop(key, None)

    def clear_state(self):
        self.log_format = dict.fromkeys(self.log_record_order)

    def format(self, record: logging.LogRecord) -> str:  # noqa: A003
        """Format logging record as structured JSON str"""
        return json.dumps(
            {
                "event": super().format(record),
                "timestamp": self.formatTime(record),
                "my_default_key": "test",
                **self.log_format,
            }
        )


logger = Logger(service="payment", logger_formatter=CustomFormatter())


@logger.inject_lambda_context
def lambda_handler(event, context):
    logger.info("Collecting payment")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
{
    "event": "Collecting payment",
    "timestamp": "2021-05-03 11:47:12,494",
    "my_default_key": "test",
    "cold_start": true,
    "function_name": "test",
    "function_memory_size": 128,
    "function_arn": "arn:aws:lambda:eu-west-1:12345678910:function:test",
    "function_request_id": "52fdfc07-2182-154f-163f-5f0f9a621d72"
}

Bring your own JSON serializer

By default, Logger uses json.dumps and json.loads as serializer and deserializer respectively. There could be scenarios where you are making use of alternative JSON libraries like orjson.

As parameters don't always translate well between them, you can pass any callable that receives a dict and return a str:

Using Rust orjson library as serializer
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
import functools

import orjson

from aws_lambda_powertools import Logger

custom_serializer = orjson.dumps
custom_deserializer = orjson.loads

logger = Logger(service="payment", json_serializer=custom_serializer, json_deserializer=custom_deserializer)

# NOTE: when using parameters, you can pass a partial
custom_serializer_with_parameters = functools.partial(orjson.dumps, option=orjson.OPT_SERIALIZE_NUMPY)

logger_two = Logger(
    service="payment", json_serializer=custom_serializer_with_parameters, json_deserializer=custom_deserializer
)

Testing your code

Inject Lambda Context

When unit testing your code that makes use of inject_lambda_context decorator, you need to pass a dummy Lambda Context, or else Logger will fail.

This is a Pytest sample that provides the minimum information necessary for Logger to succeed:

Note that dataclasses are available in Python 3.7+ only.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
from dataclasses import dataclass

import fake_lambda_context_for_logger_module  # sample module for completeness
import pytest


@pytest.fixture
def lambda_context():
    @dataclass
    class LambdaContext:
        function_name: str = "test"
        memory_limit_in_mb: int = 128
        invoked_function_arn: str = "arn:aws:lambda:eu-west-1:809313241:function:test"
        aws_request_id: str = "52fdfc07-2182-154f-163f-5f0f9a621d72"

    return LambdaContext()


def test_lambda_handler(lambda_context):
    test_event = {"test": "event"}
    fake_lambda_context_for_logger_module.handler(test_event, lambda_context)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()


@logger.inject_lambda_context
def lambda_handler(event: dict, context: LambdaContext) -> str:
    logger.info("Collecting payment")

    return "hello world"
Tip

Check out the built-in Pytest caplog fixture to assert plain log messages

Pytest live log feature

Pytest Live Log feature duplicates emitted log messages in order to style log statements according to their levels, for this to work use POWERTOOLS_LOG_DEDUPLICATION_DISABLED env var.

Disabling log deduplication to use Pytest live log
1
POWERTOOLS_LOG_DEDUPLICATION_DISABLED="1" pytest -o log_cli=1
Warning

This feature should be used with care, as it explicitly disables our ability to filter propagated messages to the root logger (if configured).

FAQ

How can I enable boto3 and botocore library logging?

You can enable the botocore and boto3 logs by using the set_stream_logger method, this method will add a stream handler for the given name and level to the logging module. By default, this logs all boto3 messages to stdout.

Enabling AWS SDK logging
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
from typing import Dict, List

import boto3

from aws_lambda_powertools import Logger
from aws_lambda_powertools.utilities.typing import LambdaContext

boto3.set_stream_logger()
boto3.set_stream_logger("botocore")

logger = Logger()
client = boto3.client("s3")


def lambda_handler(event: Dict, context: LambdaContext) -> List:
    response = client.list_buckets()

    return response.get("Buckets", [])

How can I enable Powertools logging for imported libraries?

You can copy the Logger setup to all or sub-sets of registered external loggers. Use the copy_config_to_registered_logger method to do this.

Tip

To help differentiate between loggers, we include the standard logger name attribute for all loggers we copied configuration to.

By default all registered loggers will be modified. You can change this behavior by providing include and exclude attributes. You can also provide optional log_level attribute external loggers will be configured with.

Cloning Logger config to all other registered standard loggers
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
import logging

from aws_lambda_powertools import Logger
from aws_lambda_powertools.logging import utils

logger = Logger()

external_logger = logging.getLogger()

utils.copy_config_to_registered_loggers(source_logger=logger)
external_logger.info("test message")

How can I add standard library logging attributes to a log record?

The Python standard library log records contains a large set of attributes, however only a few are included in Powertools Logger log record by default.

You can include any of these logging attributes as key value arguments (kwargs) when instantiating Logger or LambdaPowertoolsFormatter.

You can also add them later anywhere in your code with append_keys, or remove them with remove_keys methods.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
from aws_lambda_powertools import Logger

logger = Logger(service="payment", name="%(name)s")

logger.info("Name should be equal service value")

additional_log_attributes = {"process": "%(process)d", "processName": "%(processName)s"}
logger.append_keys(**additional_log_attributes)
logger.info("This will include process ID and name")
logger.remove_keys(["processName"])

# further messages will not include processName

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
[
    {
        "level": "INFO",
        "location": "<module>:16",
        "message": "Name should be equal service value",
        "name": "payment",
        "service": "payment",
        "timestamp": "2022-07-01 07:09:46,330+0000"
    },
    {
        "level": "INFO",
        "location": "<module>:23",
        "message": "This will include process ID and name",
        "name": "payment",
        "process": "9",
        "processName": "MainProcess",
        "service": "payment",
        "timestamp": "2022-07-01 07:09:46,330+0000"
    }
]

For log records originating from Powertools Logger, the name attribute will be the same as service, for log records coming from standard library logger, it will be the name of the logger (i.e. what was used as name argument to logging.getLogger).

What's the difference between append_keys and extra?

Keys added with append_keys will persist across multiple log messages while keys added via extra will only be available in a given log message operation.

Here's an example where we persist payment_id not request_id. Note that payment_id remains in both log messages while booking_id is only available in the first message.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
import os

import requests

from aws_lambda_powertools import Logger

ENDPOINT = os.getenv("PAYMENT_API", "")
logger = Logger(service="payment")


class PaymentError(Exception):
    ...


def lambda_handler(event, context):
    logger.append_keys(payment_id="123456789")
    charge_id = event.get("charge_id", "")

    try:
        ret = requests.post(url=f"{ENDPOINT}/collect", data={"charge_id": charge_id})
        ret.raise_for_status()

        logger.info("Charge collected successfully", extra={"charge_id": charge_id})
        return ret.json()
    except requests.HTTPError as e:
        raise PaymentError(f"Unable to collect payment for charge {charge_id}") from e

    logger.info("goodbye")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
[
    {
        "level": "INFO",
        "location": "<module>:22",
        "message": "Charge collected successfully",
        "timestamp": "2021-01-12 14:09:10,859",
        "service": "payment",
        "sampling_rate": 0.0,
        "payment_id": "123456789",
        "charge_id": "75edbad0-0857-4fc9-b547-6180e2f7959b"
    },
    {
        "level": "INFO",
        "location": "<module>:27",
        "message": "goodbye",
        "timestamp": "2021-01-12 14:09:10,860",
        "service": "payment",
        "sampling_rate": 0.0,
        "payment_id": "123456789"
    }
]

How do I aggregate and search Powertools logs across accounts?

As of now, ElasticSearch (ELK) or 3rd party solutions are best suited to this task. Please refer to this discussion for more details

Last update: 2023-05-01