Grafana invests a lot into observability. Observability has 4+1 areas: logs, alarms, metrics, traces and correlation. Tempo is the tracing puzzle of the Grafana observability portfolio. This article investigates how it works in my test environment. This article uses environment of below articles: https://faun.pub/multi-hop-tracing-with-opentelemetry-in-golang-792df5feb37c https://faun.pub/istio-tracing-with-jaeger-756ed9872e73

Microservice architecture is famous nowadays, but it has some drawbacks, see my earlier article: Microservice architecture is not a silver bullet. One of the painful complexity is handling the component version dependency. The component version dependency is more than component dependency, because it specifies which version is used on the…

Nowadays most of self-respecting architects prefers microservice architecture. But it’s not a silver bullet. I’ve already worked in the era of Service-Oriented Architecture and Layered Architecture, so the Microservice Architecture is only a today famous concept for me, which will be precipitated by a new concept in the future. …

Correlating several events and logs is necessary in distributed systems (not only in microservice architecture). OpenTelemetry tracing gives a standardized solution for this expectation. OpenTelemetry supports several observability solutions, this article focuses only on tracing. OpenTelemetry is not only a specification, it’ s an SDK for many programming languages (OTel…

Golang throws panic on closing an already closed channel. There are several alternatives to avoid this kind of panic. A signal channel in Golang is used to trigger event handling in several goroutines parallel, for example handling a shutdown signal. The type of this kind of channel is struct{}, because…

The world is changing. Four decades ago one developer could develop a killer application from scratch with a pure compiler, see Jonathan Sachs's interview about Lotus 1-2-3: https://conservancy.umn.edu/bitstream/handle/11299/107619/oh388js.pdf . At that time the creation meant coding. Nowadays, when the traditional coding time is decreasing, because developers write automatic tests, use daemons and services, make deployments, using the AI in software development is just following the long-term trends, which increase the productivity to deliver services, instead of standalone applications.

Not only value copy speeds up Go. Slice is faster on filling, map is faster on seeking. Preallocation is faster than dynamic allocation. Sometime slice is faster for seeking. Let’s see the figures. Golang is a garbage collection (GC) language, similar to Java and Python. The GC is an expensive…

ChatGPT and AI are famous buzzwords in the last few months. Journalists, analysts, writers, and historians are scared about their future. This kind of profession will be changed in the next years. Shall the IT engineers care about their jobs? The short answer is: partly yes. …

The message-based communication between distributed services is not an easy ride, because the response target is not evident. The idiomatic way for sending the response is asynchronous request/response with a 2nd queue. The 2nd queue can be a fan-out topic, where all request publisher subscribe to, or can be a…