Mastodawn

Just discovered this very nice piece of software. Very plug and play, nice and easy to use. If you work with #OpenTelemetry, I would definitely recommend
https://github.com/mesaglio/otel-front

GitHub - mesaglio/otel-front: Lightweight OpenTelemetry viewer for local development. View traces, logs & metrics instantly.

Lightweight OpenTelemetry viewer for local development. View traces, logs & metrics instantly. - mesaglio/otel-front

GitHub

setevoy 20h ago

VictoriaTraces: Recording Rules, Metrics, and Alerts from Trace Spans
https://rtfm.co.ua/en/victoriatraces-recording-rules-metrics-and-alerts-from-trace-spans/
VictoriaTraces - just like Vic
#Grafana #OpenTelemetry #VictoriaMetrics

VictoriaTraces: Recording Rules, Metrics, and Alerts from Trace Spans

Example of creating Recording Rules for metrics and alerts from OpenTelemetry traces in VictoriaTraces - latency and errors in a Backend API, AWS ALB, and PostgreSQL

RTFM: Linux, DevOps, and system administration | DevOps-engineering, and system administration. Cases from practice.

Dash0 21h ago

You added #vLLM to your stack. Now try figuring out why inference suddenly got slow. Traditional #APM won’t show you KV cache pressure, queueing, prefill, or decode bottlenecks. New walkthrough from Julia Furst Morgado on observing vLLM with #OpenTelemetry and Dash0: eu1.hubs.ly/H0vjfj40

Observing vLLM with OpenTeleme...

Observing vLLM with OpenTelemetry and Dash0 · Dash0

How to observe vLLM using OpenTelemetry and Dash0. Covers the traces vLLM actually emits, which Prometheus metrics matter, OTel Collector configuration, and what you'll see once data is flowing.

sayzard 1d ago

Agents will make your telemetry explode. You are not ready

에이전트 기반 자동화가 급증하면서 기존의 텔레메트리 및 관측성 시스템이 감당하지 못하는 데이터 폭증과 복잡한 상태 조합 문제가 발생하고 있다. 기존의 메트릭, 로그, 트레이스 분리 방식은 에이전트 속도에 맞춰 확장되지 못하며, 컨텍스트 손실로 인해 문제 원인 추적이 어려워진다. OpenTelemetry의 올바른 활용과 W3C 트레이스 컨텍스트 표준을 통한 신호 연계가 필수적이며, 에이전트의 의사결정 경로를 관측 가능하게 설계하는 것이 관건이다. AI 에이전트 시대에 맞는 관측성 아키텍처 재설계가 시급하다.

https://shippingbytes.com/2026/05/21/agents-will-make-your-telemetry-explod/

#observability #telemetry #aiagents #opentelemetry #monitoring

Agents will make your telemetry explode. You are not ready. – shipping bytes

I have operated systems at scale long enough to know that the scariest failures are not the ones that blow up immediately. They are the ones that accumulate quietly, in the background, until the night you realize you cannot see what is happening anymore. Agents are going to do that to your observability stack. Vercel […]

robrich 1d ago

https://www.linkedin.com/posts/juliocasal_something-broke-in-prod-you-have-no-idea-share-7454601371544264704-Nt1t - "Something broke in prod and I don't know what." That's an observability fail. This is what #OpenTelemetry solves.

Something broke in prod. You have no idea where. That's not a bug. That's a missing observability layer. Many engineering teams reach for logs first. And logs help. But logs alone won't tell you… | Julio Casal | 12 comments

Something broke in prod. You have no idea where. That's not a bug. That's a missing observability layer. Many engineering teams reach for logs first. And logs help. But logs alone won't tell you why a request took 4 seconds, which service caused the cascade, or whether the problem is CPU, memory, or a slow database call. That's what OpenTelemetry solves. It's a vendor-neutral, open standard for collecting the three signals you actually need: 𝐓𝐫𝐚𝐜𝐞𝐬 tell you what happened to a request across every service it touched. You see exactly where latency was introduced and where errors occurred. 𝐌𝐞𝐭𝐫𝐢𝐜𝐬 give you the numbers over time. Request rates, error rates, GC pressure, thread pool saturation. The stuff that tells you when a system is about to break, not just after it does. 𝐋𝐨𝐠𝐬 still matter, but structured logs correlated with trace IDs are a completely different tool than raw text files. Here's what makes OpenTelemetry different from everything before it: → One SDK. Any backend. No vendor lock-in. → Works with Jaeger, Prometheus, Grafana, Aspire, Azure Monitor, Datadog, whatever you use. → Auto-instrumentation for ASP .NET Core, HttpClient, EF Core, and more. → The industry has aligned on this. It's not going away. In ASP .NET Core, setup is a few NuGet packages and a handful of lines in Program.cs. The image has the full code pattern. The real unlock is running an OTLP Collector in production. You batch, filter, and route telemetry without touching your app. Change your backend without a redeploy. Here's a full blueprint to start your next .NET project with all the OpenTelemetry goodness wired up 👇 https://lnkd.in/gnQhKDDC | 12 comments on LinkedIn

OpenTelemetry 1d ago

#OpenTelemetry is a #CNCF graduated project

From OpenCensus and OpenTracing to today, this project has been a community effort from the beginning.

Thank you to everyone who contributed code, docs, specs, reviews, feedback, support, and end-user experiences. This milestone belongs to you.

https://opentelemetry.io/blog/2026/otel-graduates/

Show thread

VictoriaMetrics 1d ago

👇 Learn why #Spotify Engineering relied on #VictoriaMetrics for extreme #Scale observability in this case study.
https://bit.ly/49e5P6w

#Prometheus #OpenTelemetry #Grafana #Engineering #OTel

Spotify’s performance & control across large monitoring environments with VictoriaMetrics

Spotify needed to replace its legacy in-house time series database to overcome stability and performance limitations, which would bring about query delays and timeouts. The Spotify observability team chose VictoriaMetrics to support efficient metric ingestion, querying, and alerting at scale.

VictoriaMetrics

sayzard 2d ago

PromptVC, version control for prompts in production

PromptVC는 프로덕션 환경에서 LLM 프롬프트 변경 이력을 체계적으로 관리하는 버전 관리 도구입니다. 프롬프트 변경에 따른 성능, 비용, 응답 품질 변화를 추적하고, 텍스트 차이뿐 아니라 행동 차이까지 비교해 운영상의 불확실성을 줄입니다. OpenTelemetry와 OpenInference를 활용해 자동으로 LLM 호출을 계측하며, Python과 TypeScript SDK를 지원해 다양한 환경에 쉽게 통합할 수 있습니다. 통계적 벤치마크 기능도 제공해 프롬프트 변경의 효과를 객관적으로 평가할 수 있습니다.

https://www.promptvc.io/blog/introducing-promptvc

#llm #promptengineering #observability #versioncontrol #opentelemetry

Introducing PromptVC: Prompt Version Control for Production LLM Apps

An LLM feature can regress without a deploy. PromptVC gives LLM applications version history, traces, diffs, metrics, alerts, and enough evidence to say whether a change helped or hurt.

Geeking Out Podcast 2d ago

Have you heard of #OpenTelemetry Arrow, but wasn't sure what it was all about? This blog post by Geeking Out host @adrianamvillela demystifies Arrow: What is it? What problem does it solve? How does it work?

Learn more: https://open.substack.com/pub/geekingoutpodcast/p/lets-learn-about-otel-arrow-together?r=27vnzy&utm_campaign=post&utm_medium=web

And don't forget to subscribe to our Substack!