How to monitor Containers in Kubernetes using Prometheus & cAdvisor & Grafana? CPU, Memory, Network

▬▬▬▬▬ Experience & Location 💼 ▬▬▬▬▬
► I’m a Senior Software Engineer at Juniper Networks (12+ years of experience)
► Located in San Francisco Bay Area, CA (US citizen)

▬▬▬▬▬▬ Connect with me 👋 ▬▬▬▬▬▬

▬▬▬▬▬▬ Related videos 👨‍🏫 ▬▬▬▬▬▬

▬▬▬▬▬▬▬ Timestamps ⏰ ▬▬▬▬▬▬▬
0:00 Intro
1:28 Monitor CPU
15:47 Monitor Memory
20:39 Monitor Network

▬▬▬▬▬▬▬ Source Code 📚 ▬▬▬▬▬▬▬

#Kubernetes #Prometheus #DevOps

[jorgearaujo]

First of all many thanks, you are a savior. In this tutotial I have fooled allow until 4:51 that query does not work for me... I have changed the values due to my cadvisor exporting different labels.. I don´t have or just "image" or "pod" even so I get always this error "bad_data: 10:1: parse error: unexpected <by> in aggregation", also I have notice in my Prometheus I don not have the "container_spec_cpu_quota" target.

Again Thank you for all the tutorials

[sujithsurendran]

As usual very informative. Subscribed!

[chasim]

Great content as always, can you please create a detailed video on opensource monitoring and alerting solution for kubernetes with Robusta? thanks

[meron]

This is so good. Thank you for sharing.

[mehranmoradi]

Hello great conten if i want add node exporter where can i add node exporter or add alertmanager configs?

[koskoskng]

С шестой минуты вообще всё несходится. У меня совсем другие переменные из Прометея возвращаются.
Жаль, как раз искал дэш который бы измерял потребление ресурсов подами. В галерее Графаны вообще ничего толкового не нашел(.

[George-mklp]

is there an advantage of using Prometheus over cloudwatch ?

[ashutoshpandey]

Hi Anton, How have you been?😊

I am working on a personal project basically prometheus monitoring, hope you can help me in this topic. Basically i want to do k8s node monitoring where i want to calculate the current cpu and memory usage, so that i can get the actual information of a node !!! for memory -- current usage i am using this formula > by (pod, node, namespace))) by (node) / 1024 / 1024 for request memory --> by (node) / 1024 / 1024 for allocatable memory on node --> by (node) / 1024 / 1024 These are the formula i am using to monitor the memory usage but i am not able to align these formulas together where i can get the actual information of my node usage!!

[michaelpege]

Does anyone have the promql query? I'm new to prometheus and when I try typing the promql myself the chart doesn't load up like the video at 6:28 :')

[yacinebmy]

is there a way to retrieve containers metrics (cpu and memory usage) on a specific namespace ?

[PePTo-dxyj]

Приветствую! не тянет метрики Grafan показывает no Data при выборе POD moitoring, как будто метрик не хватает

[wawaji]

How to filter cfs metrics from cadvisor of some specific namespaces? Thx

[_JAGATHEESANV]

Are we able to automate using the data from Prometheus

[wawaji]

Where i can find the source code in the video? Thx