Почему ваши логи бесполезны и как это починить за полчаса

Когда продакшен падает в три часа ночи, строка ERROR Something went wrong не помогает никому. В статье разбираем, почему привычные текстовые логи быстро превращаются в шум при реальной нагрузке, как перейти на structured logging, зачем каждому запросу нужен request_id и как настроить нормальные JSON‑логи в Python и Go без лишней инфраструктуры.

https://habr.com/ru/companies/otus/articles/1034388/

#structured_logging #структурированные_логи #JSONлоги #логирование #request_id #observability #OpenTelemetry #structlog #slog #диагностика_ошибок

Почему ваши логи бесполезны и как это починить за полчаса

Три часа ночи, алерт, сервис отдаёт 500. Открываете логи и видите: 2026-05-12 03:14:22 ERROR Something went wrong 2026-05-12 03:14:22 ERROR Failed to process request 2026-05-12 03:14:23 ERROR...

Хабр

New post from my blog...

In case you missed it earlier...

Stir Trek 2026 - When Logs Tell the Whole Story
https://barretblake.dev/posts/conferences/2026/05/stir-trek-2026/ #stirtrek #opentelemetry #otel

New post from my blog...

Stir Trek 2026 - When Logs Tell the Whole Story
https://barretblake.dev/posts/conferences/2026/05/stir-trek-2026/ #stirtrek #opentelemetry #otel

Mercure 0.24.1 is out, riding on Caddy 2.11.3.

We contributed native OTLP metrics push to Caddy upstream. Mercure hubs can now ship metrics straight to your OpenTelemetry collector. No Prometheus scrape job needed. Drop

metrics {
otlp
}

into your Caddyfile and the standard OTEL_* env vars do the rest.

Pairs with the OTel tracing landed in 0.24, so traces + metrics work end-to-end against any OTLP backend.

https://github.com/dunglas/mercure/releases/tag/v0.24.1

#opensource #mercure #caddy #opentelemetry

Release v0.24.1 · dunglas/mercure

Community Mercure 0.24.1 picks up Caddy 2.11.3, including our upstream contribution that adds OTLP metrics push to Caddy. The Helm chart now also surfaces a JSON values schema and a signed .prov pr...

GitHub

🫳 #GrafanaCON 2026 talks are available on demand! Just in time for some weekend viewing.

Don't miss talks like:
• GrafanaCON 2026 Keynote
• New #OpenTelemetry engine
• k6 2.0
• Planet-scale dashboards at Google

And more that won't fit in this toot: https://grafana.com/events/grafanacon/

🔭 Amsterdam — OTel Night is coming!

🎟️ Free community event — open to all!

OTel Night Amsterdam v2026.05 — May 20 at ING Cedar (106 Bijlmerdreef). Engineers from ING + Albert Heijn on running OpenTelemetry at enterprise scale. Real war stories, no vendor fluff.

📅 May 20, 2026 — 6 PM CEST
📍 ING Cedar, Amsterdam
🎟️ Free — 24 spots left (of 50)

Register: https://ocgroups.dev/cncf/group/v62fy27/event/g58n983

#OpenTelemetry #Observability #OTelNight #SRE #Amsterdam #CNCF

— Rodrigo Marins Piaba

Open Community Groups

Where Open Source Communities Thrive

Instrument application code, you say? Developers caring about #Observability and #OpenTelemetry, you say? Yes, we do say!

Geeking Out host @adrianamvillela and past guest Diana Todea talk about the importance of instrumenting application code, what should be instrumented, and local desktop tools that developers can use to view OpenTelemetry traces, logs, and metrics.

https://open.substack.com/pub/geekingoutpodcast/p/the-lazy-developers-guide-to-observing?r=27vnzy&utm_campaign=post&utm_medium=web

🏨 #Airbnb now has a 100M samples/sec pipeline powered by #vmagent + #OpenTelemetry
🪜 VictoriaMetrics v1.142.0 provides critical fixes: #OTEL ingestion + query bugs. Please upgrade as soon as possible
🪵 VictoriaLogs v1.50 adds Splunk HEC, Datadog native paths, file log collection
🆔Multiple improvements in VictoriaLogs #LogsQL: multi-field search, stddev(), global filters, CSV exports.
Learn more in our latest article 👇
https://bit.ly/4ue33GM

#Monitoring #logs

VictoriaMetrics April 2026 Ecosystem Updates

VictoriaMetrics April 2026 release round‑up: heads up about critical VictoriaMetrics bugfixes in v1.141.0–v1.142.0, and explore new VictoriaLogs features including Splunk ingestion and advanced LogsQL tooling.

VictoriaMetrics

Your AI agent just took 45 seconds to answer a simple question. Was it the
model? A slow tool call? A retry loop?

#OpenTelemetry is making it easier to find the answers with semantic conventions for generative AI. Learn more in today's post!

https://opentelemetry.io/blog/2026/genai-observability/

🔴 I'm live now with Tiffany Jernigan and Christian Simon to talk about profiling, the fourth pillar, and how to use it with OpenTelemetry! Join us in chat to ask your questions about profiling, Pyroscope, and OTel. :) https://www.youtube.com/watch?v=N47wy9pp7Lo

@grafana
#observability #opentelemetry

Continuous profiling, the OpenTelemetry way (Grafana ❤️‍🔥 OpenTelemetry Community Call)

YouTube