🧑‍💻🔥 Coroot 1.18.0 is here with custom, granular alerts from built-in log patterns, #PromQL conditions, and health checks that cover your entire system! https://github.com/coroot/coroot/releases/tag/v1.18.0

#Coroot #observability #opensource #softwarelibre #ai #cloud #devops #sre #monitoring #aws #freesoftware

@ any time series query language: let me facet or where with the value from another query.

For example: show me all containers on host where host has memory pressure, facet by container termination reason.

AFAIK this cannot be accomplished with #dataDog or #promql - strong RTFM maybe on promql though.

Quick promql tip: promql _hates_ duplicate data in joins and will throw an absolute fit if you ever have them.

Unfortunately sometimes you end up with overlapping data points, because, e.g., kube-state-metrics is rolling out and there are temporarily two copies running, or because Prometheus also keeps stale data for up to five minutes after a datapoint is no longer seen.

The way I've worked around this in the past is to throw my query inside `max(metric) by(labels I care about)`, but I also learned/realized today that you could instead do `topk(1, metric)`, which is like `max`, except that it keeps all the labels on that datapoint.

Obviously you should think carefully about whether this is correct or not for your use case. If your metric is binary (only ever 0 or 1) the max is fine. If you use topk and you depend on the labels output, be aware that it's just going to pick one arbitrarily.

It would be best if you are able to clean up/select out the duplicate metrics, but sometimes that's.... not possible or more annoying than it's worth.

#kubernetes #prometheus #promql

"We’re writing to remind you that as of October 22, 2024, Monitoring Query Language (MQL) is no longer a recommended query language for Cloud Monitoring. We made this change to focus on PromQL, the open-source standard for time series querying, which offers similar functionality, a wider user base, and more community resources."

#GoogleCloud #Prometheus #PromQL

Как правильно использовать rate() в Grafana: от мониторинга до расчёта SLO

Функция rate() в PromQL необходима для вычисления средней скорости изменения метрики в секунду за определённый период времени. Она часто используется для мониторинга таких показателей, как: Как подружить rate() и Grafana

https://habr.com/ru/articles/905124/

#prometheus #grafana #promql

«База» по метрикам в Prometheus

Привет, Хабр! Меня зовут Глеб Типсин, я являюсь ведущим разработчиком продукта «Системы персонализации и геймификации» в SM Lab. В этой статье мы разберём, как устроены метрики в Prometheus, как происходит их сбор, а также что с ними можно делать. Кроме того, обсудим подходы в мониторинге, архитектуру Prometheus и её ограничения. Материал подойдёт тем, кто только знакомится с мониторингом и хочет освоить «базу», а также тем, кто уже работает с Prometheus , но хочет "освежить" свои знания и узнать что-то новое.

https://habr.com/ru/companies/sportmaster_lab/articles/872204/

#prometheus #мониторинг #мониторинг_производительности #prometheus_grafana #monitoring #promql

Have you ever created an awesome #PromQL query, but then get a “No data” response when you run it? If so, PromQL vector matching might be the culprit. In this blog post, a #Grafana community member explains what you need to know.

https://grafana.com/blog/2024/12/13/promql-vector-matching-what-it-is-and-how-it-affects-your-prometheus-queries