Adventures in home networking

I apparently wrote this up about four years ago but never got around to finishing/posting it. Was interesting to re-read this, especially as I am currently running Windows Server Essentials 2012 R2 with StableBit DrivePool, but am planning to switch to unRAID (a Linux-based OS) in the near future.

Lots of the stuff I mention here is no longer relevant/in use at my house, but whatever.

I’ve been running a home server for some years now. I initially went with an OEM copy of Windows Home Server (v1) after a brief trial and found it pretty impressive. I particularly liked the simplicity and ease of Drive Extender. This worked out pretty well for some time. A few years in, however, v1 was starting to show its age… it was based on Windows Server 2003 after all.

Enter Windows Home Server 2011. Now based on Windows Server 2008, this version supported 64-bit hardware and as such could take advantage of additional RAM. The remote web access was upgraded to use Silverlight. And they killed Drive Extender. I still don’t really understand why (possibly to make room for Storage Spaces, which would debut in Windows 8 / Windows Server 2012?). Thankfully, a few third-party developers stepped in to fill the gap. I ended up using StableBit DrivePool and was fairly impressed/happy with it.

I use the server to host basically all of our digital media: photos, music (ripped to FLAC for archiving, which can then be converted to other formats as needed), and movies (right now mostly Blu-Rays ripped to MKV, though I’m considering redoing this and ripping to ISO). I also use it as a file server to make file access from different computers stupid easy.

The centralized media storage was step one in my eventual goal: effortless access to ALL our (my?) digital media from any room in the house. Potentially even remotely, in theory. After we moved in I even hired an electrician to run Cat6 ethernet cables to most of the rooms (including an additional pair of wires for HDMI over Cat6… which will probably not be as useful as I’d envisioned now thanks to Dish Network’s adoption of MoCA). Now that we have a Hopper, those cables aren’t getting much use. Oh well. Anyway, the wiring was step two. Now I have centralized media and a gigabit network to carry it to each room.

Up next are the streaming devices. These are connected to AV receivers and/or TVs for output. I’ve got a WDTV Live device in the bedroom which does a good job EXCEPT for its inability to decode DTS. In the living room, one of my old home server platforms, an AsRock E350M1 AMD Fusion APU system, was initially running the Windows 8 Release Preview, since it’s sort of free, and XBMC. Since the release of XBMC 12 includes AudioEngine, I can bitstream HD Audio from my Blu-Ray rips to my receiver, which is pretty awesome.

I tried using Windows Server 2012 Essentials for the home server, but using an OS which requires the establishment of a domain is a little tough to deal with. I went back to WHS 2011 and have since upgraded that device to Windows 8. I’m actually trying to come to a decision on how to proceed; what I’d really like to do is move to a Linux or other free software platform like FreeNAS or OpenMediaVault; two things are holding me back right now — CrashPlan support (for offsite backup) and disk resiliency.

It’s important to me that any important files are not only stored on the server with redundancy, but are backed up somewhere else (if the house burns down, storage on multiple drives doesn’t do me much good). Unfortunately, it seems like CrashPlan doesn’t support FreeBSD (on which FreeNAS is based). OpenMediaVault is based on Debian Linux, though, so that seems promising.

With any Windows-based system, I can use StableBit DrivePool and pretty much be assured that an individual drive failing won’t result in any lost files, AND the rest of the pool will still be intact even if multiple drives fail. As far as I can tell, I can’t do that with RAID, even with RAID 6, since the striping means that if more than two drives failed, the rest of the volume would be fucked. So far it seems like any free software solutions are going to be RAID-based.

On the subject of Linux, I’d really love to move to it full-time, at least on my laptop; games keep me on Windows on my desktop, but with Steam available for Linux, hopefully that will change in the years to come. There are really only a few things that I can’t do on Linux (with respect to my laptop, which I use for work) are use Microsoft Office and my company’s software. However, these are both kind of a big deal for me. Maybe if we move to Google Apps in the future, and ever manage to release a Linux version of our software I can.

Rural broadband (or the lack thereof)

I’ve had this domain for a while and recently decided I might as well get off my ass and start using it.

So yeah. I live in what might be classified as “rural” California. In reality, I live a few miles outside of an actual city, and unfortunately in my neck of the woods that means I’m out of luck when it comes to broadband.

Charter is nearby in Redding, with 60-100 Mbps products, but they aren’t allowed into my town because of outdated rural territory restrictions or something like that. Our local ISP (Frontier) basically has an unchecked monopoly in the area. Thanks to this lack of competition, they offer a whopping “up to 6 Mbps” DSL product here, which is actually provisioned, at least for me, at 1.3 Mbps. I try to talk to some of the reps every time we have a town festival, where there’s usually a Frontier booth. Every time, I’m told they aren’t planning to do any upgrades, fiber is under the main road but won’t be brought out to subdivisions, etc. Basically they’re content to offer crap service and be done with it.

This is really irritating because, frankly, it’s 2016. The FCC has reclassified broadband as 25+ Mbps, which means we’re living in the dark ages here.

Over the last few years I’ve tried various options, including a solid local WISP, but settled on a Verizon grandfathered unlimited data plan (UDP). When Verizon purchased their 700 MHz spectrum, the FCC mandated they allow any activated SIM to work on any approved device. The practical upshot of this is that an activated UDP SIM will work in a USB modem or hotspot. Of course, you’re stuck paying for minutes and texts you most likely won’t be using (or have access to), but given the alternatives, it’s a pretty compelling solution. So compelling that I’ve been using one for two years now, with (mostly) great results.

If I get around to it, I’ll try to post some more about the various setups I’ve tried, upgrades I’m considering, and probably stuff about my home network setup as well.