RHCSA lab setup — what VM specs are people actually using for practice?
Starting my RHCSA prep and trying to figure out the right lab environment. I've got a decent laptop with 16GB RAM and an i7, but I'm not sure how many VMs I should be running simultaneously or whether RHEL 9 is the right version to practice on. I keep seeing conflicting info depending on how old the source is.
My Linux background is intermediate — I use it daily at work but mostly for scripted deployments, not deep system administration. I'm planning on 10 weeks of prep, about 90 minutes of hands-on lab time daily plus 30 minutes of reading. That's roughly 140 hours total which feels right based on what I've read from people who've passed recently.
I've been doing RHCSA practice test questions to identify weak spots and the storage and logical volume management sections are clearly where I need more reps. My written scores are around 70% but my hands-on confidence is lower than that, which worries me since the exam is entirely performance-based.
Does anyone run nested VMs or is a flat setup with 2-3 machines sufficient? I want the lab to mirror the actual exam environment as closely as possible without overcomplicating the setup.
16GB RAM is more than enough for a 2-3 VM setup. I ran my whole lab on 8GB with no performance problems. Allocate 2GB per VM and keep the rest for your host OS. Nested VMs add complexity that isn't worth it for RHCSA prep.
The LVM questions are no joke. I'd say spend at least 20% of your lab time just on storage — creating VGs, extending LVs, mounting persistently. Do it until you can do it in 4 minutes without notes because the exam clock pressure is real.
RHEL 9 is the current exam version so that's definitely what you should be practicing on. Two VMs is enough — one main server and one client for the NFS and SSH tasks. I used 2GB RAM each and had no issues at all.
Performance-based format means your hands-on time is worth more than any amount of reading. The people who struggle are almost always the ones who prepped heavily on theory and lightly on actual CLI practice. Reverse that ratio if you can.
Failed my first attempt last spring and honestly the lab setup was part of why. I was trying to run three VMs at once on 12GB RAM and everything was crawling, so I'd skip the hands-on parts and just read docs instead. Big mistake. Second time I scaled back to two VMs max, gave each 4GB, and actually did every task until my hands remembered it. RHEL 9 is the right call since that's what the exam uses now.
The thing that helped most second time was drilling storage stuff obsessively because it's where I lost the most points. I found a good rhcsa lvm storage management resource and worked through it until creating and resizing logical volumes was just muscle memory. With your i7 and 16GB you're in a better spot than I was, so don't waste it the way I did — keep two VMs running, practice breaking things and fixing them, and you'll be fine.
With 16GB RAM and an i7 you're honestly in good shape. I ran three VMs simultaneously during my prep -- one "server" node, one "client" node for practicing stuff like NFS mounts and autofs, and a third I'd snapshot constantly as a clean slate to reset when I broke things. RHEL 9 is definitely the right version, don't let anyone talk you into practicing on 8 unless you've got a specific reason.
The thing that actually helped me wasn't just doing the tasks, it was breaking them on purpose and figuring out why. Like, I didn't just practice setting SELinux booleans correctly -- I'd set them wrong first, watch what broke, read the audit logs, then fix it. That way when an exam task asks you to troubleshoot something you haven't seen before, you're not just pattern-matching from memory, you actually understand what's happening underneath. A lot of people fail because they memorized the right commands but have no idea what to do when the wrong answer error message shows up.