Merge pull request #65 from DataDog/chris.agocs/docs_for_adding_request_id_to_logs

Add documentation and example for adding request ID to logs.

Author: agocs

README.md

@@ -87,11 +87,8 @@ If you set the value of this variable to "true" then the Lambda layer will incre
 
 ### DD_LOGS_INJECTION
 
-By default, the Datadog trace id gets automatically injected into the logs for correlation, if using `console` or a logging library supported for [automatic](https://docs.datadoghq.com/tracing/connect_logs_and_traces/?tab=nodejs#automatic-trace-id-injection)  trace id injection.
-
-See instructions for [manual](https://docs.datadoghq.com/tracing/connect_logs_and_traces/?tab=nodejs#manual-trace-id-injection) trace id injection, if using other logging libraries.
-
-Set the environment variable `DD_LOGS_INJECTION` to `false` to disable this feature.
+Controlls whether or not the Datadog Trace ID is injected into log lines. See [DD_LOGS_INJECTION](#DD_LOGS_INJECTION-environment-variable) under
+the Trace & Log Correlation section below.
 
 ## Usage
 
@@ -136,6 +133,52 @@ sendDistributionMetric(
 
 If your Lambda function is associated with a VPC, you need to ensure it has access to the [public internet](https://aws.amazon.com/premiumsupport/knowledge-center/internet-access-lambda-function/).
 
+## Trace & Log Correlation
+
+### Using the Datadog Tracer
+
+If you are using the [Datadog Tracer](#datadog-tracer-experimental), follow [these instructions](https://docs.datadoghq.com/tracing/connect_logs_and_traces/nodejs/#automatic-trace-id-injection)
+to set up automatic trace-log correlation.
+
+### Without using the Datadog Tracer
+
+In order to correlate logs emitted by your Lambda with specific invocations, it
+is necessary to add the AWS Request ID to your logs. This is done automatically
+for `console.log()`, but you will have to implement this for other logging libraries.
+
+The AWS Request ID is available in the context that is passed to your lambda handler,
+as `context.awsRequestId`. It should be included in your log line as `lambda.request_id`.
+
+For example, using the [Pino](https://getpino.io/) logger:
+
+```typescript
+const logger = require('pino')();
+
+exports.handler = async function(event, context) {
+ 
+  //This sets up your request-specific logger to emit logs with the Request ID property.
+  const req_logger = logger.child({ 'lambda.request_id': context.awsRequestId });
+
+  //Carry on with whatever the lambda needs to do
+  const work = do.Work();
+
+  //Write a log message
+  req_logger.info("Work complete");
+
+  return work;
+}
+
+```
+
+### DD_LOGS_INJECTION environment variable
+
+By default, the Datadog trace id gets automatically injected into the logs for correlation, if using `console` or a logging library supported for [automatic](https://docs.datadoghq.com/tracing/connect_logs_and_traces/?tab=nodejs#automatic-trace-id-injection)  trace id injection.
+
+See instructions for [manual](https://docs.datadoghq.com/tracing/connect_logs_and_traces/?tab=nodejs#manual-trace-id-injection) trace id injection, if using other logging libraries.
+
+Set the environment variable `DD_LOGS_INJECTION` to `false` to disable this feature.
+
+
 ## Distributed Tracing
 
 [Distributed tracing](https://docs.datadoghq.com/tracing/guide/distributed_tracing/?tab=nodejs) allows you to propagate a trace context from a service running on a host to a service running on AWS Lambda, and vice versa, so you can see performance end-to-end. Linking is implemented by injecting Datadog trace context into the HTTP request headers.