[eBook Presented by AWS & Thundra] Mastering Observability on the Cloud đź“– Download:

Introducing: Thundra Integration for Honeycomb

Nov 6, 2018


We believe that your data should not exist in a silo and you should be able to view, manipulate, and analyze it in the way best suited for your business, using your favorite platforms. With this goal in mind, we feel integrations with popular data platforms are an important part of Thundra’s offering to the serverless monitoring and observability space.

When we released our Splunk integration, we knew it would be a powerful way to view and analyze Thundra monitoring data. Many enterprise organizations use Splunk and we wanted to provide a way for users to monitor and observe their serverless environments within the platforms they are most comfortable using. That’s why we are proud to introduce our new integration with Honeycomb today.

At the beginning of June, Honeycomb announced the general availability of tracing in Honeycomb. We share a common understanding of observability with Honeycomb team and felt that anyone interested in debugging their serverless apps with Honeycomb would find a Thundra integration valuable. Using Thundra’s Honeycomb integration, serverless communities will be able to visualize data generated by Thundra in their Honeycomb deployment.

You can set up the Thundra-Honeycomb integration very easily - in minutes. First, you need to obtain a license key, which gives you a month long free trial of our integration. Request your license key, here, and the Thundra team will send you a license key as soon as they receive the request. Then, you will install Thundra-Honeycomb integrator to your AWS account and point that account to your Honeycomb deployment. After you complete the setup, Thundra Integrator for Honeycomb will collect, transform, and forward AWS Lambda serverless monitoring data to Honeycomb.

After set up, your monitoring data will flow directly from your AWS account into your Honeycomb deployment. You will see the Thundra datasets in your Honeycomb instance within minutes.

You can find detailed information about the steps below. 

Set up the Honeycomb Integrator using a CloudFormation

We provide a CloudFormation template to install the Thundra Integrator for Honeycomb. It helps you create and configure the Integrator EC2 instance. When installing, follow the instructions in our documentation. Make sure your Integrator instance is publicly available since Thundra agents will send the collected data to this instance.

We also made an Amazon Linux-based AMI for our Honeycomb Integrator that will soon be available on the AWS marketplace for easy setup and config.

There are two packages in the instance:

  • Our Honeycomb Integrator, which is a simple Java-based web app that uses a RESTful API to accept incoming data from Thundra Lambda agents. It modifies the Thundra data to the Honeycomb trace format, adding fields such as id, parentId, and serviceName to the collected trace data.
  • Honeycomb’s Honeytail agent which forwards the data to Honeycomb.

Update your Lambda Functions’ Environment Variables

After setting up the Integrator, the last step is to set your environment variables. Set the `thundra_agent_lambda_report_rest_baseUrl` environment variable of your AWS Lambda function(s) to the URL of the Integrator EC2 instance. You can find the URL of the instance in the Output tab of the CloudFormation stack. Get detailed instructions here. Now, you’ve completed the set up of your Thundra and Honeycomb integration! Congratulations!




Analyze Serverless Data in Honeycomb

Now, let’s dive into serverless monitoring data in Honeycomb! The Thundra integration provides five datasets:

  • metrics, logs, and traces (which forms the three pillars of observability)
  • Spans
  • An invocation dataset that contains detailed information such as function duration, application runtime environment, and more.

You can perform various actions on all these datasets such as breaking down the data, filtering it, and aggregating it as you require.

Let’s walk through one example. You may find the trace dataset particularly interesting because the trace charts allow you to easily follow the path of your invocation. There are two ways to visualize trace charts in Honeycomb.

  • Method 1: Hover your mouse over the graph and a small circle will appear with trace metadata. Click on that circle to view trace chart.
  • Method 2: Add a break down by traceId field to your query to list all available traces. Then, click on the traceId field in the results table to view trace chart.



After you navigate to the trace charts, you can now see the details of your AWS Lambda function (see the chart below). You can see spans and their respective durations on this chart and each is clickable to explore additional information. When you click on span information, details about that span comes up on the right of the screen, including id, timestamp, duration, parentId etc.




There you have it! Hopefully, this introduction has been helpful in showing you how you can use Honeycomb to dive into our AWS Lambda monitoring data with the Thundra Integrator for Honeycomb. We’d love to hear your thoughts on this integration as well as what other integrations would be useful for you. Visit our website and follow us on Twitter and LinkedIn to keep up with the latest news and features.