Our customers love Thundra’s trace charts because the charts help identify which parts are problematic in an AWS Lambda invocation. In one of our previous blog posts, we walked through how to detect a real life invocation problem with the Thundra trace charts.
As of today, we are proud to announce that we’ve enriched our trace charts with new metric information aimed at providing even more value and support for AWS Lambda developers. Previously, we only showed trace chart for an invocation. Now, with this improvement, you can gather more information about any single invocation in what we hope is a more clear and concise way.
From now on, we will call this page as “Invocation Detail”. In order to open it on the Thundra Web Console, click on the name of a function to navigate to details of that function.
Then, navigate to the “Invocations” tab in order to see the full list of invocations of that selected function.
Note, this list is quickly and easily filterable according to date, region, and profile. It is also possible to filter cold started, and/or erroneous invocations, or invocations that took more than a threshold time to execute.
Now, you are one click away from seeing the detail of a single invocation. Just click on an invocation in the list to dive into the details.
You have arrived! The “Invocation Details” page is designed to help you understand what’s happening with your AWS Lambda invocation. Let’s walk through this updated page to see what you can learn about your invocation.
The “Cards” section on top lets you to see the summary view of your invocation. This includes duration of an invocation, memory and CPU usages, and errors. In a healthy serverless system, you will want to see all these cards as green.
The “Basic Information” section includes some detailed, yet helpful, information. If you expand this section, it will show the link to your invocation’s CloudWatch(CW) logs, your function’s Amazon Resource Name (ARN), your invocation’s request id, and more. Using this information, you can check the invocation against CW logs.
We’ve updated our original “Trace chart” section with a refreshed user interface. In our original implementation, we only provided the raw trace data. However, if your raw data included tons and tons of lines, it was really hard to track and understand what was really going on with your invocation. To remedy this, we use the “Arguments & Return” tab to provide critical information as smart cards. In the “Trace Log” tab, you can see the span’s log. And, if you want to still see the raw JSON you can reach it under “Raw Data” tab.
We are all familiar with CW logs, right? For this reason, we felt it important that you can see the logs of the invocation in a single pane. We provide these at the bottom of the page in the “Invocation Log” section. In this section, you can see the logs colored with respect to log levels.
At Thundra, we always try to put ourselves into the shoes of the developer and make every release and every update with this mindset as our top priority. We believe this new invocation detail page will help you detect, debug, and solve the problems more effectively as well as learn almost anything you need to about the invocation you are analyzing. We want to get your feedback! Do you think this new page is an improvement and will help you dive deeper into your invocations?
Try out our free beta and let us know what you think!
And, connect with us in our Slack channel and give us your feedback, ideas, requests, and tell us about your serverless observability use cases and needs. We are in beta mode and actively developing Thundra based upon user feedback, which means this is your opportunity to help us create a platform which satisfies your needs. Want to learn more about Thundra’s other features? Visit our website or request a demo today.