Thundra APM
Search…
Thundra APM
Welcome to Thundra APM
Switch to Sidekick Docs
Switch to Foresight Docs
Getting Started
Getting Started with Thundra APM
Quick Start Guide
FAQ
Performance Benchmarks
Thundra Web Console
Introduction
Functions List Page
Dashboard Page
Function Details Page
Invocation Detail Page
Applications Page
Applications Detail Page
Query Helper
Architecture Page
Unique Traces Page
Unique Trace Details Page
Operations Page
Logs page
Alerts Page
APIs Page
API Details Page
Account & Settings
Error Inspector Page
Node.js
Integration Options for Node.js SDK
Integration Options for Node.js SDK in Containers and VMs
Configuration Options for Node.js
Enrich Tracing in Node.js
AWS Lambda Debugging for Serverless Node.js
Node.js Example Projects
FAQ for Node.js SDK
Python
Integration Options for Python SDK
Configuration Options for Python
Enrich Tracing in Python
AWS Lambda Debugging for Serverless Python
Python Example Projects
FAQ for Python SDK
Java
Integration Options for Java SDK
Integration Options for Java SDK in Containers and VMs
Configuration Options for Java
Enrich Tracing in Java
AWS Lambda Debugging for Serverless Java
Java Example Projects
FAQ for Java SDK
Go
Integration Options for Go SDK
Configuration Options for Go
Enrich Tracing in Go
Golang Example Projects
FAQ for Go SDK
.Net
Integration Options for .NET SDK
Integration Options for .NET SDK in Containers and VMs
Upgrading from v1.x to v2.x
Configuration Options for .NET SDK
Enrich Tracing in .NET
.NET Example Projects
FAQ for .NET SDK
Web Frameworks
Spring
Express.js
Hapi
Koa.js
Django
Flask
.NET Core 2.1
.NET Core 3.1
Fastapi
Tornado
Debugging
Time-Travel Debugging
AWS Lambda Debugger
IDE Integrations
VSCode Plugin
IntelliJ IDEA Plugin
Others
Observability
Root Cause Analysis of Outliers
Architectural View to Check The Issues at a Glance
Tracking Serverless Transactions End-to-End
Staying Aware of Issues with Alerts
How to Respond Alerts in Thundra
Reducing Data Size with Sampling
Chaos Engineering with Thundra
Three Pillars of Observability
Seamless Monitoring
Digging Observability Data with Thundra Queries
Performance
Zero Overhead with Lambda Extensions
Zero Overhead with Asynchronous Monitoring
Dealing with Cold Starts
Deployment Integrations
AWS SAM
Serverless Framework
Step by Step Migration Guide from Epsagon to Thundra
Powered By GitBook
Tracking Serverless Transactions End-to-End
In today's modern software development, distributed applications have become more popular – and by adopting distributed serverless systems, we can expect the process of building applications will be faster and highly scalable. Despite these benefits, distributed serverless systems do have challenges when it comes to monitoring and troubleshooting. Users working with distributed applications need to display their serverless system with visibility into services and their interactions.
Thundra offers users the capability to track their serverless distributed system from end-to-end with Full Tracing, which provides both distributed tracing and local tracing. Full Tracing allows you to monitor your Lambda functions from a high level, such as interactions with other Lambdas and services, to a deeper level, right down to lines of code.
Distributed tracing is supported with two unique capabilities in Thundra:
  • Multiple upstream transactions - There can be multiple invocations when just one invocation is triggered. For example, a batch of messages can come through multiple invocations and be written to DynamoDB with one transaction. In Thundra, you can display each upstream invocation link to a downstream invocation.
  • Business transaction - Some invocations are related to each other with logical interactions, such as writing a message to a DynamoDB table after approval. You can link logically related flows with Thundra's distributed tracing.
Thundra provides this capability via Traces. You can display your all individual traces on the Traces Page. Using detailed queries, you can filter your traces in terms of:
  • Trigger of the origin
  • Entry point
  • Start time
  • End-to-end duration of the trace
  • Interacted resources
  • Error types, if any are thrown
  • Duration breakdown
When you want to go through a review on a deeper level of your serverless system, filter your traces according to your needs and click on any trace. The Trace map page will be shown as below:
Trace Map
The Trace Map provides you with a view of interactions between Lambda functions in a specific trace, along with metrics and the health of interactions. As seen in the image below, a failed Lambda function is highlighted with a red square. You can click on the failed Lambda function to figure out what's going wrong.
Trace Details
Side view is an easy way to detect the root cause of issues in a problematic node in a specific trace. With one click, you can investigate your CloudWatch logs or details of the invocation of that trace.

How to configure your serverless system for distributed tracing

To use end-to-end tracing, you can easily configure your system by going through our documentation on the following pages:
​
Observability - Previous
Architectural View to Check The Issues at a Glance
Next - Observability
Staying Aware of Issues with Alerts
Last modified 1yr ago
Copy link
Contents
How to configure your serverless system for distributed tracing