rHomelab

Most people pick a sharing protocol by accident. Whatever the NAS web UI defaulted to, or whatever worked the first time they tried mapping a drive on Windows, becomes the permanent choice, usually without anyone asking whether it actually fits the workload. For a media library that mostly gets read by one Jellyfin box, that accident rarely bites. For a Proxmox cluster sharing storage across nodes, or a mixed household of Windows laptops and Linux boxes hitting the same array, the wrong protocol choice shows up later as corrupted files, mysterious permission errors, or a performance ceiling nobody can explain. The three real options, NFS, SMB, and iSCSI, solve different problems, and the right call depends on what’s connecting, not what’s easiest to click through first. ...

The managed-switch and 10GbE articles on this site both talk around one thing without ever explaining it: the actual port you plug 10GbE gear into usually isn’t RJ45. It’s SFP+, a small rectangular slot that takes a separate module, and that module is where most people buying their first piece of 10GbE gear get stuck. This is the piece that fills that gap: what SFP+ actually is, the three ways to fill one, and the specific mistakes that turn a cheap used switch into a pile of ports that won’t link up. ...

Every homelab eventually runs into the same wall: you’ve got more VMs you want to run than RAM to dedicate to them if each one reserves its full allocation up front. Proxmox gives you two tools that let you push past that wall, memory ballooning and KSM, and both get treated by a lot of guides as a free lunch you just flip on. They’re not. They buy you real headroom, but only if you understand what each one is actually doing underneath, because the failure mode when you push either one too far isn’t a warning, it’s an OOM kill or a host that grinds to a crawl. ...

Most homelab stacks start with the fun stuff, media servers, dashboards, maybe a VPN. The boring problem that never goes away is paper: tax documents, insurance paperwork, warranty receipts, kids’ school forms, the occasional letter that actually matters. Paperless-ngx exists to solve exactly that, and it’s one of the few self-hosted apps where the payoff shows up the first week instead of months later. Scan something in, and from then on you can find it by typing a word that appears anywhere in it, not by remembering which folder you filed it under three years ago. ...

Every layer of protection covered elsewhere on this site has a specific job. RAID survives a dead drive (see the RAID redundancy guide). Checksums and scrubs catch silent corruption (see bit rot and checksumming). The 3-2-1 strategy (see the backup guide) gets a copy off-site. None of them protect you from the most common actual cause of data loss in a homelab: you, or something acting with your credentials, deleting or overwriting a file that was perfectly healthy a minute ago. ...

The out-of-band management article on this site covers IPMI, iDRAC, and iLO, the built-in remote console that real server motherboards ship with. That’s the right answer for a lot of homelab hardware. It is not the answer for all of it. Mini PCs, most consumer desktop motherboards, a lot of budget NAS boards, and the occasional used server where the BMC itself has quietly died, all have zero out-of-band path. When one of those needs a monitor and keyboard in front of it, which eventually happens to every piece of hardware you own, you need a different plan. ...

Clustering gets you live migration, moving a running guest from one node to another with no downtime, but “it just works” is doing a lot of hiding in that sentence. Live migration has real mechanics underneath it, real prerequisites that aren’t optional, and a short list of device types and configurations that simply cannot migrate live no matter how the storage is set up. Knowing which category a given VM or container falls into before you click migrate saves you from finding out mid-migration, which is the worse time to learn it. ...

RSS feeds never went away. Most blogs, news sites, forums, and even a lot of YouTube channels and subreddits still publish one, whether or not they advertise it on the page. What disappeared was the easy on-ramp, Google Reader shut down over a decade ago and nothing free and centralized ever fully replaced it as the default. The gap got filled by a handful of paid hosted services (Feedly, Inoreader) and by self-hosting, and for a homelab operator the self-hosted path has a real advantage beyond just avoiding a subscription: your reading list, read/unread state, and starred articles stay on hardware you control, not in a third party’s analytics pipeline. FreshRSS, Miniflux, and Tiny Tiny RSS (TT-RSS) are the three tools that consistently come up, and they land in genuinely different places on the simplicity-versus-features spectrum. ...

If you’re already on ZFS, whether that’s TrueNAS, a Proxmox host, or a DIY OpenZFS box, the obvious tool for moving a data hoard to a second machine is rsync. It works, but it has a cost that gets worse as the hoard grows: every run has to walk the entire directory tree, stat every file, and figure out what changed before it transfers a single byte. On a pool with a few hundred thousand media files, that walk alone can take longer than the actual transfer. ZFS has a better answer built in, and it’s been there the whole time: zfs send and zfs receive. ...

Every build guide tells you to “get a good power supply” and leaves it there, as if that settles the question. It doesn’t. A PSU is the one component that touches every watt your homelab ever draws, and the difference between a cheap unit and a good one shows up as heat, noise, and a power bill you can actually measure, not just a line on a spec sheet. This is the PSU side of the picture: what the efficiency ratings mean, how server PSUs differ from the ATX unit in a tower build, and what redundant power supplies actually buy you once you start running used enterprise gear. ...