The RHLS labs are good quality. I'm learning a lot from them. But I think it's important for me to also get to grips with my own learning environment. I think a bit different when I am not being guided through an exercise, and that seems to be making some knowledge stick a little more.
Anyone else set up their own home lab to learn Red Hat products like the OS, Ansible, or Openshift?
If so, are you using something on-premise like a PC with VMs, or are you going off-premise with a cloud-hosted solution?
I prepared for RHCSA, RHCE, Ansible, OpenShift Administration and even some others by simply building the needed VM's from scratch on a beefy Lenovo laptop, so yes, that suffices.
Although I must say, that after a year of RHLS and preparing for 5 exams doing only the included labs, it should work quite well, as long as you don't look at the answers of the end of chapter labs and comprehensive reviews beforehand But mileage may vary of course,
I have a 32 cores, and 64GB of RAM in a PC using RHV (self supported) using VMs for various technologies. I can take advantage of snapshots for my virtual machines to facilitate the experimentation, learning, restore process. And I can take advantage of Ansible on my Mac to further facilitate automation.
Don't let installation of virtual machines inhibit your learning, take advantage of virtualization, containers, cloud technologies where possible.
Use Ansible Automation, and snapshots to your advantage.
And use the Red Hat Developers Subscription!
If you're satisfied with the solutions provided please mark the solution as ACCEPTED.
Don't forget to thank those who helped you out with kudos!
Two votes for using a workstation, with lots of grunt for running virtual machines.
I thought some might use AWS, GCP or Azure.
I'm setting up a workstation. Anyone set up Openshift in VMs? Is that even a good idea?
I set myself up with a RHEL workstation. I bought an old Lenovo W520 laptop. Plenty of CPU cores. I replaced the memory with 16GB and the disk with SSD. I signed up for a developer subscription and installed RHEL7.
I failed to get the Nvidia card to play nice so classroom video playback was choppy. I dont want to waste revision time troubleshoting that, so it's back to Win10 and Oracle's Virtualbox.
Next I install Openshift,. Or perhaps OKD, or Minishift. Or even Kubernetes and Minikube.
I'm in the same boat really, I've got a couple of laptops that I use to run Proxmox/KVM. Meets all my needs at this moment in time.
I've also got an AWS account that I use for testing (e.g. EKS), however, it costs money, so I tend to default to my homelab if possible.
I'm setting up a home lab environment as well. One question though - are the lab files used in the courses available externally someplace? I started to use a local system for the courses since I was burning through my 400 lab hours fairly quickly.
Some labs were easy enough to recreate, but other ones have been more difficult. Right now I'm going through DO180 and am reaching the point where the custom lab files are necessary.
I would advise everyone to follow Ricardo's suggestion of using the Red Hat Developer Subscription. Having real product bits makes a big difference when you think about certification and real-word jobs, compared to using Fedora, CentOS, and other Linux distros. They are valuable, but I was hurt more than one by small differences in package names, configuration defaults, and software releases.
I would also advise everyone to jump in all Red Hat: install RHEL on your lab machine, use KVM virtualization, use nested virtualization to setup OpenStack, install OpenShift, Satellite, RHEV, and play with Ansible. Even when installing a product is not among the objectives of an exam, the knowledge and experience you get are invaluable. Even if your short term goal is only RHCE, you will move up and want specialist certifications.
Red Hat certifications are performance-based. The more experience you get, the easier they become.
I checked out the price of a laptop with 32GB and 8 cores, so I can run Openshift in VMs. The price made me cry. I'll have to wait.
I stood up an openshift master, infrastructure node and compute node on AWS (3 EC2 machines with 8GB memory, 2 CPUs, 30GB disk and a 10GB partition for docker). But I used CentOS, and like @flozano says, the devil is in the details.
My head scratching continues.
Yes, those machines can be quite expensive (and they are much more where I live, compared to the US) and the cloud provider bills may also surprise you. Sometimes you can make it work with smaller machines:
- Many times a 1GB or 1.5GB RHEL VM can do the work, you will not have lots of users nor requests in a lab environment, and you can fire up more of them to simulare multi-server scenarios
- Minishift (and the soon to be released CodeReady Containers product) allow you to run a single-worker OpenShift cluster confortably inside a 12GB or 16GB workstation.
- That spare/older machine can work well with only SSH access, no GUI desktop, saving more RAM for your VMs
Yes, it all requires more work, and won't provide for any kind of lab environment, compared to a beefy 32GB or 64GB computer. But they can help you land the next better job that allows you to buy the more expensive machine, or where the employee gives you one. :-)