OpenTelemetry is a vendor-neutral approach that enables DevOps and developers to collect performance metrics in a standardized manner. Currently a Cloud Native Computing Foundation (CNCF) sandbox project, OpenTelemetry was conceived by merging OpenCensus, Google's open-source method of collecting metrics and traces, and OpenTracing, a vendor-neutral API to collect traces.
I’ve spent most of my career working with tech in various forms, and for the last ten years or so, I’ve focused a lot on building, maintaining, and operating robust, reliable systems. This has led me to put a lot of time into researching, evaluating, and implementing different solutions for automatic failure detection, monitoring, and more recently, observability. Before we get started: What is observability?
We're excited to help our customers further leverage the benefits of OpenTelemetry via integration with AWS Distro for OpenTelemetry (ADOT) for traces 1.0.
Modern application architectures are complex, typically consisting of hundreds of distributed microservices implemented in different languages and by different teams. As a developer, SRE, or DevOps engineer, you are responsible for the reliability and performance of these complex systems. But while you might have metrics that will help you debug when there’s an issue, metrics alone can’t help you narrow down and ultimately identify the root cause.
Our log agent is powerful, efficient, and highly adaptable. Now, with OpenTelemetry setting new standards in the observability space, we wanted to incorporate that collaboration into our log agent and offer our users the ability to take advantage of the OpenTelemetry ecosystem. Starting today, you can upgrade the log agents in your observIQ account to the new Open Telemetry-based observIQ log agent with a single click.
Sampling is the practice of extracting a subset of data from a dataset to make conclusions about that larger dataset. It’s far from a perfect solution, but when it’s implemented with Refinery, Honeycomb’s trace-aware sampling proxy, sampling can help you manage very high volumes of complex event data.
It is no surprise that monitoring workloads are top of mind for many organizations to ensure a successful customer experience. As our applications become more distributed and cloud-native, we find that monitoring can become more complex. A single user transaction fans out to interact with tens or hundreds of microservices, each one requesting data from backend data stores or otherwise interacting with each other and other parts of your infrastructure.
In this tutorial, we will go through a working example of a Ruby application auto-instrumented with OpenTelemetry. To keep things simple, we will create a basic “Hello World” application, instrument it with OpenTelemetry’s Ruby client library to generate trace data and send it to an OpenTelemetry Collector. The Collector will then export the trace data to an external distributed tracing analytics tool of our choice.
Open Telemetry represents an effort to combine distributed tracing, metrics and logging into a single set of system components and language-specific libraries. Recently, OpenTelemetry became a CNCF incubating project, but it already enjoys quite a significant community and vendor support. OpenTelemetry defines itself as “an observability framework for cloud-native software”, although it should be able to cover more than what we know as “cloud-native software”.
Plenty of blogs have answered the very Google-able question, “What is OpenTelemetry?” To keep it short and sweet, OpenTelemetry is a collaborative effort across the observability space to create industry-wide standards that will benefit all cloud service providers and observability customers. Technically speaking, OpenTelemetry is a collection of APIs, SDKs, exporters, and collectors.
There’s no strict definition of a distributed system. But generally speaking, if you have reached a point where you’re running more than five interdependent services at once, that means you’re running a distributed system. It also means you are more than likely experiencing difficulties when troubleshooting using traditional debugging tools. Unfortunately, pulling up multiple tools, each built for a monolithic world, doesn’t help pinpoint the problem.
