AWS Lambda is a cloud service offered by Amazon Web Services (AWS) that lets developers run their code in response to a trigger, such as an HTTP request or a message in a message queue. It is the primary component of the AWS serverless computing platform. The code is executed in a serverless environment, which means that the developer does not have to worry about the underlying infrastructure, such as servers or virtual machines.
AWS Lambda automatically logs the events and the results of function execution, including any errors or exceptions that may have occurred. These logs can be accessed through the AWS Management Console, the AWS CLI, or the AWS CloudWatch Logs API.
By default, AWS Lambda logs all function invocations and the results to Amazon CloudWatch Logs. CloudWatch Logs is a cloud service that collects, monitors, and stores logs from various sources, including AWS Lambda. It provides a way to view, search, and set alarms for the logs, as well as to export the logs to other destinations.
In addition to the logs generated by AWS Lambda, developers can also add their own logs to the function execution by using the logging libraries or functions provided by their programming language of choice. These custom log messages can also be accessed through CloudWatch Logs and can be used to debug and troubleshoot issues with the function.
This is part of a series of articles about serverless debugging.
In this article
AWS Lambda automatically reports metrics for your Lambda functions, logging all requests handled by the function and storing the logs in Amazon CloudWatch. You can insert logging statements into function code, and these custom logs will also be stored in CloudWatch. All the logs are stored to a CloudWatch Logs group associated with the Lambda function, named /aws/lambda/<function name>.
You can access the logs for a Lambda function in the following ways:
To access Lambda logs using CloudWatch, follow these steps:
The logs for the function execution will be displayed in the log stream. You can use the search and filter options to find specific log messages.
In order to use the AWS CLI, you will need to create an AWS Access Key and Key Secret in Amazon IAM User Management. If you do not properly configure access, you will not be able to run the code below.
To access Lambda logs using the AWS CLI, follow these steps:
aws logs filter-log-events --log-group-name "/aws/lambda/<function-name>" --start-time <start-time> --end-time <end-time>
<function-name> with the name of your function, and <start-time> and <end-time> with the time range for the logs you want to retrieve, in Unix timestamp format.To access Lambda logs using the AWS Lambda console, follow these steps:
Related content: Read our guide to Lambda tracing
When you invoke a Lambda function for the first time, a log group will automatically be created. There can be multiple versions of a Lambda function, allowing you to choose the version you want to run. A single log group stores all the logs for the invocations of a Lambda function.
A logging library is useful for formatting and categorizing log messages. For instance, you might use the Python logging module for Lambda functions written in Python—this will make it easier to control your output format and log messages. Another tip is to log messages in the JSON format to make querying with CloudWatch Logs Insights easier. This also makes filtering and exporting easier.
Another good practice is using a variable to set and adjust the logging level and based on your requirements and environment. The Lambda function’s code can generate smaller or larger volumes of log data at different logging levels, impacting logging performance and costs.
With Lambda, you can specify environment variables for the function’s runtime environment without updating the code. For instance, you can start by creating a LAMBDA_LOG_LEVEL environment variable that defines the logging level for your code. Here is an example of how to import and use the LAMBDA_LOG_LEVEL variable and to define a “logger”:
import os
import logging
# Get the value of the LAMBDA_LOG_LEVEL environment variable
log_level = os.environ.get('LAMBDA_LOG_LEVEL', 'INFO')
# Configure the logger
logger = logging.getLogger()
logger.setLevel(log_level)
# Your code goes here
def my_function(event, context):
logger.debug('This is a debug message')
logger.info('This is an info message')
logger.warning('This is a warning message')
logger.error('This is an error message')
In this example, the value of the LAMBDA_LOG_LEVEL environment variable is used to set the log level for the logger. If the environment variable is not set, the default log level of INFO is used. The log level determines the type of messages that will be logged.
To send Amazon Lambda logs to other destinations from CloudWatch, you can use the CloudWatch Logs Subscription Filter feature. This allows you to set up a real-time feed of your logs to other services, such as Amazon Elasticsearch Service, Amazon Kinesis Data Streams, or a third-party logging service, using a custom AWS Lambda function as the delivery stream.
To set up a CloudWatch Logs Subscription Filter:
Once the subscription filter is set up, CloudWatch will automatically send new log events to the delivery stream in real-time, where they can be processed and forwarded to the destination service.
You can also use the CloudWatch Logs API or the CloudWatch Logs Agent to set up a subscription filter programmatically.
There are several best practices for logging in AWS Lambda functions:
By following these best practices, you can improve the quality and usefulness of your AWS Lambda logs and better understand the behavior and performance of your serverless applications.
Lumigo is a distributed tracing platform purpose-built for troubleshooting microservices in production. Developers building serverless apps with AWS Lambda and other serverless services use Lumigo to monitor, trace and troubleshoot their serverless applications. Deployed with no changes and automated in one-click, Lumigo stitches together every interaction between micro and managed services into end-to-end stack traces, giving complete visibility into serverless environments. Using Lumigo to monitor and troubleshoot their applications, developers get: