hi, I am currently reading and documenting a book about Kubernetes. I am glad to see that I can understand and remember most of the stuff you are talking about in this video. Really great for the basics of Kubernetes! Awesome work, Anton!
Большое спасибо, сейчас прохожу разные курсы для становления DevOps инженером. Там дают много информации, но что бы вот такую крутую и понятную выжимку про основы, не видел ни в одном.
Content is good but it feels like video is missing empathy it is like robots reading things out. Give some pause in between and it will give more realistic touch
Given the pace at which you go through the concepts, one would have to be familiar with these concepts and just view your videos as a refresher. Too fast paced for me and relations between PVC and PV shown in diagram is not clear what that is about (many other things like these).
for the hpa with `request-per-second` do i need Prometheus to get this metric ? ... i don't think so because it's type: object,,, i think i need only Ingress
You're correct, you don't need prometheus for that only for "external". kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale-walkthrough/#autoscaling-on-more-specific-metrics
Hello Anton, I am one of the subscribers to your channel and your videos are quite informative and very helpful in understanding the technology. -:) I have couple of doubts related to tools that are used. Could you please help clarify these doubts please? In REAL WORLD projects, tools such as Jenkins, ELK stack, Databases, RabbitMQ/ActiveMQ, Kafka etc. are generally provisioned on EC2s (or) Vendor provided/managed cloud environments (SaaS) (or) Cloud Provider(AWS/Azure etc.) specific provided/managed environments but not on Kubernetes/EKS? And, Only microservices are deployed on K8s clusters and they connect to these tools via K8s environment configurations? It is very confusing and struggling to understand as most of the tutorials/examples in Internet are showcasing above mentioned tools to provision on K8s along with microservices. Kindly help clarfying the doubt! Thank you.. :-)
In the real world, mostly startups and small companies try to deploy everything to Kubernetes and use as many managed resources as possible. Generally, when they achieve some level of profitability, they start to think about infrastructure more seriously. For example, if you are a startup and your services are down for a few hours, it's okay or even expected, but if you are a large company such as Facebook, it's a disaster. :) So yes, microservices go to Kubernetes, and Kafka, ELS, Cassandra, etc., are deployed and managed independently on VMs (e.g., EC2). Those microservices then communicate with Kafka, etc., inside the private cloud (VPC) using private subnets and private DNS hosted zones.
Thank you for the video!!! Could you please create a video that delves into the topic of stateful applications on Kubernetes and provides insights on effectively managing persistent storage in this context
Kubernetes nodes are a global Kubernetes concept; they cannot belong to a particular namespace. For example, if you want to run only specific applications on those nodes, you would add taints to those nodes, and in the YAML deployment, you would use a tolerations block. doc - kubernetes.io/docs/concepts/scheduling-eviction/taint-and-toleration/
