DevOps Fundamentals Ideas for the new Domain 5 in the Version 004.

Michael Schmitz

Well-known member
  • Aug 9, 2021
    348
    327
    24,616
    Germany
    www.linkedin.com
    As mentioned in the Sneak Peek for the new version, (in case you missed it, you can watch the recording and download the slides)
    DevOps (why not DevSecOps as now called),
    and most of the participants said, they have not much expierence with the DevOps,
    lets share Ideas, ressources how to teach it.

    If you are training over weeks or months, spend plenty of time here, would be easier, as in teaching it in 5 days to non-dev people, who need to understand the basics.

    are their free ressources available to the different tools, like
    ELK,
    Grafana,
    git,
    jenkins
    and so one, that we can use or share?

    Post it here, or upload it to the Ressources / Media Section in CIN and provide the link here.

    Would be great to have some stuff for the Trainers going into the TTT then later this year..


    Michael
     

    BrianFord

    Well-known member
    Jun 26, 2023
    55
    123
    8,941
    Flagler Beach, FL
    fordsnotes.com
    Great topic Michael!

    A resource that I found very useful was this from Altassian. Agile development and Devops methods are compatible but not the same thing.

    This article walks through both agile development (the agile manifesto) and introduces a couple of the sources of Devops methodology (the Devops three ways).
     
    The meaning of DevOps has changed dramatically since Werner Vogels originally started the idea back in 2006 with "You build it, you ship it."

    It originally meant putting more and more shipping responsibility in the hands of developers - this came in the form of a few frameworks like Ruby on Rails and Heroku, but was really popularized by the Docker container model, which led to the split from the original term.

    As container development practices matured, DevOps became more associated with development types that placed "most" deployment in the hands of developers (this later became known as GitOps), and Site Reliability Engineers (SREs) became the role that handled the stuff that developers really couldn't do in larger projects, such as fancy orchestration (e.g., Jenkins and large scale Kubernetes) as well as monitoring (e.g., Prometheus/Grafana), security (e.g., Stackhawk, IAM), and end-to-end solutions (e.g., OpenShift).

    Today, DevOps is less understood as a term than it original was - but in general, it's used to denote developers who have some experience with some forms of deployment-related activity.

    For Cloud+, I'd personally recommend focusing only on SRE-style topics and from that viewpoint, given the audience for the certification.
    My two cents :)
     

    Michael Schmitz

    Well-known member
  • Aug 9, 2021
    348
    327
    24,616
    Germany
    www.linkedin.com
    @Jason: thank you for the Info, the Proble with this domain is, the Non Dev Trainers, that have no expierence with this Stuff, and probably will never get...
    And then there are the Academy Instructors, who have plenty of time to spend time on this Dev Part with using the Tools to start with Projects..

    So, how to teach it? For me, i will talk about the Tools and where to use them, since my Courses are usally 5 Days or less.
    I hope till the TTT, we will have more Ideas what to include and what we can use to explain..

    Michael
     
    As someone who has spent the last decade working as a developer, devops engineer, SRE, and cloud architect for a local university incubator, I can tell you that there are no shortcuts when it comes to building skills in the area of devops and SRE.

    What I can tell you is what I designed for my cloud program at triOS College, which prepares students for what they'll need to know about devops and SRE today, and in a way that received extreme praise all around from our industry PAC members who employ our grads.

    After getting enough practice in Linux administration (Linux+ topics), Network administration (more=better), and security (CySA+ topics), we get our students comfortable with Proxmox VE in depth - they install it on their own system on Hyper-V with nested virtualization and use it to run both Linux and Windows Server VMs and LXC containers. They practice all Proxmox VE admin concepts (SDN configuration, 2FA for all accounts in the pve Kerberos realm, storage types, backups, monitoring, etc.).

    Then we have them configure their Proxmox VE host OS into an Ansible control node and teach them YAML and Ansible CM (in depth!) to automate the configuration of all of the VMs and containers in their Proxmox infrastructure. They end by assembling their Ansible playbooks and supporting files into Ansible Roles that are pushed to their own GitHub repos.

    Then we install Kubernetes (K3s from Rancher) on one of the VMs and teach Kubernetes in depth for building, deploying, and managing containerized microservices (this is easy after Ansible because it's all YAML). This includes connecting storage to microservices, running supporting sidecar containers, the networking types, configuration of services and load balancers, etc.

    Then we finish by covering Oracle Cloud Infrastructure (their generous free tier) using their OCI-I and II curriculum. By this time, copying SSH keys for administration and setting up the Ansible CM, services (e.g., load balancers), storage, etc. is easy for students to grasp - and more importantly - allows us to finally talk about the different situations where they'd need to do these things.

    Hope this helps!
     
    jason: But you have much more time then 5 Days to go through all this.. i have max 5 days in my Courses..... So less time for the Pratical World.
    Correct, I have a minimum of 160 days to build up prerequisite skills to the material I described, and 40 days to cover the actual material I described. From a learning perspective, I wouldn't recommend doing it in less than that time.