On May 27-29, we are hosting the fourth Slurm: Kubernetes Intensive.
Bonus: Docker, Ansible, Ceph Online Courses
We have deduced topics from Slurm that are important for working with Kubernetes, but do not directly relate to k8s. How, why and what happened - under the cut.
All Slurm-4 participants will have access to these courses.
Full moneyback on the first day
At St. Petersburg Slurm, two participants left . How I regretted that it was impossible to return to the past and part with them without mutual claims.
If you find that you absolutely don't like it on Slurm, first day write to any of the organizers. We will disable access and refund the full participation price.
Technical director's consultations
If anyone knows (he formed a club of technical directors), we invited him to Slurm (to study, not to perform). He promised to advise everyone. It is unlikely that this will be interesting for administrators and developers, but for IT managers it will be very interesting.
What is Slurm
Slurm-4: basic course (May 27-29)
Designed for those who see Kubernetes for the first time or want to systematize knowledge.
Each participant will create their own cluster in the Selectel cloud and deploy the application there.
Price: 25 thousand
Program
Topic #1: Introducing Kubernetes, Core Components
• Introduction to k8s technology. Description, application, concepts
• Pod, ReplicaSet, Deployment, Service, Ingress, PV, PVC, ConfigMap, Secret
• Practice
Topic #2: Cluster design, main components, fault tolerance, k8s network
• Cluster device, main components, fault tolerance
• Network k8s
Topic #3: Kubespray, Tuning and Tuning a Kubernetes Cluster
• Kubespray, configuring and tuning a Kubernetes cluster
• Practice
Topic #4: Ceph, cluster setup and features of working in production
• Ceph, cluster setup and features of work in production
• Practice: configuring ceph
Topic #5: Advanced Kubernetes Abstractions
• DaemonSet, StatefulSet, RBAC, Job, CronJob, Pod Scheduling, InitContainer
Topic #6: Introduction to Helm
• Introduction to Helm
• Practice
Topic #7: Publishing Services and Applications
• Overview of service publishing methods: NodePort vs LoadBalancer vs Ingress
• Ingress controller (Nginx): balancing incoming traffic
• Сert-manager: automatically obtain SSL/TLS certificates
• Practice
Topic #8: Logging and Monitoring
• Cluster monitoring, Prometheus
• Cluster logging, Fluentd/Elastic/Kibana
• Practice
Topic #9: CI/CD, building a deployment to a cluster from scratch
Topic #10: Practical work, application dockerization and launch in a cluster
MegaSlurm: Advanced Course (May 31 - June 2)
Designed for Kubernetes engineers and architects, as well as graduates of the basic course.
We configure the cluster in such a way as to simultaneously launch the update of the cluster components and the deployment to the cluster.
Price: 60 thousand (45 thousand for Slurm-4 participants)
Program
Topic No. 1: The process of creating a failover cluster from the inside
• Working with Kubespray
• Installation of additional components
• Cluster testing and troubleshooting
• Practice
Topic #2: Authorization in a cluster using an external provider
• LDAP (Nginx + Python)
• OIDC (Dex + Gangway)
• Practice
Topic #3: Network policy
• Introduction to CNI
• Network Security Policy
• Practice
Topic #4: Secure and Highly Available Applications in a Cluster
• PodSecurityPolicy
• PodDisruptionBudget
Topic #5: Kubernetes. Looking under the hood
• Controller structure
• Operators and CRDs
• Practice
Topic #6: Stateful Applications in a Cluster
• Running a database cluster using PostgreSQL as an example
• Starting a RabbitMQ Cluster
• Practice
Theme #7: Keeping Secrets
• Secret management in Kubernetes
• Vault
Topic #8: Horizontal Pod Autoscaler
• Theory
• Practice
Topic #9: Backup and disaster recovery
• Cluster backup and recovery using Heptio Velero (formerly Ark) and etcd
• Practice
Topic #10: Application Deployment
•Lint
• Template and deployment tools
• Deployment strategies
Topic #11: Practical work
• Building CI/CD for application deployment
• Cluster upgrade
Docker, Ansible and Ceph
Patrimony
The first Slurm was an experiment. The speakers completed the presentations literally on the stage, and in the audience there were administrators of such a level that it is just right to invite them as speakers.
The real basic course turned out on the second Slurm: 80% of the participants saw Kubernetes for the first time, and a third had never worked with Docker.
It was clear how hard it is for people to listen to a lecture on Docker in the morning, and work with it in combat mode in the evening.
Ceph caused a lot of difficulties. Moreover, there were 20 people in the audience who definitely need to explain Ceph, and another 60 who do not need Ceph at all.
For the third Slurm, we brought Docker and Ansible into separate webinars, freeing up more time for Kubernetes. The solution turned out to be practical in essence and underdeveloped in implementation: the lecture was uninteresting for experienced guys, and the discussion for beginners.
By the fourth Slurm, we made online courses on Docker, Ansible and Ceph. The idea is simple: those who need it will thoughtfully take the course, those who don't need it will calmly ignore it. Judging by the group of testers, the Docker course takes 6-8 hours. Ansible and Ceph haven't clocked yet.
Disclaimer:
- experimental course. Some solutions will surely fail.
- the platform (Stepik.org) is rather crude and we haven't worked with it before. Surely there will be slips and blunts.
- The course has only been tested on Southbridge employees. You will probably have to finish something along the way.
Just the other day, in the chat room of the first Slurm, they recalled how cool and fun it was, despite all the organizational horrors. The first to get the most vivid impressions. Let's see what happens to the first students of online courses. 🙂
Source: habr.com