Supported Runtimes: 3.6, 3.7, 3.8

How To Use

With Lambda Layers:

• Use the latest ARN version from these tables

With Serverless framework:

• Install the serverless-lumigo-plugin

Manually

• install with pip install lumigo_tracer
  
• import using from lumigo_tracer import lumigo_tracer
• wrap you lambda function using @lumigo_tracer or @lumigo_tracer(token='XXX'). As an example, your lambda should look like:

@lumigo_tracer(token='XXX')
def my_lambda(event, context):
    print('I can finally troubleshoot!')

Configuration

• You can turn on the debug logs by setting the environment variable LUMIGO_DEBUG=true
• You can prevent lumigo from sending keys that answer specific regexes by defining LUMIGO_SECRET_MASKING_REGEX=["regex1", "regex2"]. By default, we use the default regexes [".*pass.*", ".*key.*"]. All the regexes are case-insensitive.
• Similarly, you can prevent lumigo from sending the entire headers and body of specific domains using the environment variable LUMIGO_DOMAINS_SCRUBBER=[".*secret.*"] (give it a list which is a json parsable), or by specify the list of regexes with the key domains_scrubber in the tracer's decorator. By default, we will use ["secretsmanager\..*\.amazonaws\.com", "ssm\..*\.amazonaws\.com", "kms\..*\.amazonaws\.com"].
• In case of need, there is a kill switch, that stops all the interventions of lumigo immediately, without changing the code. Simply add an environment variable LUMIGO_SWITCH_OFF=true.

Logging Programmatic Errors

You can use report_error function to write logs which will be visible in the platform.
Add from lumigo_tracer import report_error.

Then use report_error("<msg>") from anywhere in your lambda code.

Adding Execution Tags

You can use add_execution_tag function to add an execution_tag with a dynamic value.
This value can be searched within the Lumigo platform.

Add from lumigo_tracer import add_execution_tag.
Then use add_execution_tag("<key>", "<value>") from anywhere in your lambda code.
Limitation:

• The maximum number of tags is 50.
• Key and value length should be between 1 and 50.

Step Functions

If this function is part of a step function, you can add the flag step_function=True or environment variable LUMIGO_STEP_FUNCTION=True, and we will track the states in the step function as a single transaction.

@lumigo_tracer(token='XXX', step_function=True)
def my_lambda(event, context):
    print('Step function visibility!')

Note: we will add the key "_lumigo" to the return value of the function.

If you override the "Parameters" configuration, simply add "_lumigo.$": "$._lumigo".
For example:

"States": {
    "state1": {
      "Type": "Task",
      "Resource": "arn:aws:lambda:us-west-2:ACCOUNT:function:FUNCTION_NAME",
      "Parameters": {
          "Changed": "parameters",
          "_lumigo.$": "$._lumigo"
        },
      "Next": "state2"
    },
    "state2": {
      "Type": "pass",
      "End": true
    }
}

Frameworks

Chalice

• In chalice, you should add the following lines to the your file:

from lumigo_tracer import LumigoChalice
...
app = Chalice(app_name='chalice')
app = LumigoChalice(app, token="XXX")

Sentry/Raven Lambda Integration

Add our decorator beneath the Raven decorator

from lumigo_tracer import lumigo_tracer
...
@RavenLambdaWrapper()
@lumigo_tracer(token='XXX')
def lambda_handler (event, context):  return  {
 'statusCode' :  200,
 'body' : json.dumps( 'Hi!' ) }

How To Contribute

Prepare your machine

• Create a virtualenv virtualenv venv -p python3
• Activate the virtualenv by running . venv/bin/activate
• Run pip install -r requirements.txt to install dependencies.
• cd src and python setup.py develop.
• If you use pycharm, make sure to change its virtualenv through the PyCharm -> Preferences -> Project -> Interpreter under the menu
• Run pre-commit install in your repository to install pre-commits hooks.

Test

• To run the unit tests, run py.test in the root folder.
• To deploy the services for the component tests, move to the root test directory and run sls deploy. This can be performed only once if the resources doesn't change.
• To run the component tests, run py.test --all.