Last night, about an hour after I finished working a longer than average Monday shift, I opened up Plex to finish watching an episode of Night Court that I had started the day before, and I waited. I waited some more. Then I waited a tiny bit longer.

After a while, I realized that Plex wasn’t going to load my video, and I went into troubleshooting mode- it took me less than a minute to realize that I couldn’t connect into my little NAS in any way- not through a web browser or a shell program. A NAS, for you non-techie folks, is network-attached storage. It’s basically a little computer with a bunch of disks that holds a bunch of my crap. In my case, I had been using it as a storage location off my main computer for everything that needed long-term storage- photographs, important documents, and more.

A moment later and I’d gone to investigate the machine physically- all the lights were solid except one drive light which was rapidly blinking. I let it sit that way for a few minutes, and when I saw no change at all, I tried the usual thing- turning it off and on again. The fans spun up for a second, then the power light started flashing blue and the drive lights all went amber.

Cue the frantic Google research to figure out what the hell just happened, and a short while later I was certain that it was completely dead. There was a design flaw in the Atom chips used in this particular model, and they burn up and stop working over time. Like Roy Batty, my little data robot had always had a finite life-span.

It is not lost on me that my NAS died the night before World Backup Day.

It’s my own fault, I suppose- I had been contemplating upgrading to a newer model for a while, and I guess my little NAS got jealous. It pulled a full Ophelia and basically just jumped in the nearest river.

I am, of course, kidding. This demise was foretold years ago– I just didn’t know it until yesterday because Synology is garbage at notifying their customers of fatal design flaws. The frustrating part of all this for me is that because of the world’s Pandemical issues, it’s a little bit harder to get the replacements that I need to get back up and running. I’ve ordered the main piece that I need- a new host body for my disks. With a little luck, I’ll have it within a week.

My data, on the other hand- that’s a big question for me. It should be ok- the drives were RAIDed so that they protect against data loss. In theory, I should just be able to plug them right into the new console when it arrives and pick up where I left off. The settings, user profiles, and the like will probably need to be set up from scratch, but my data should be safe.

If, for some reason, the drives are scrambled and the drives are unreadable, then I’m still not completely dead because I back up EVERYTHING. I’ve been burned before. In September of 2003, my Windows XP machine crashed and burned. It was bad, people, really bad. I lost nearly sixty gig of personal data. Old photographs, e-mail correspondence- things I will never recover. That crash is why I don’t use Windows for a personal machine anymore. (Well, one of the reasons, at any rate.)

Ever since then, I’ve been fastidious (and perhaps even a little paranoid) about backing my shit up. Each of the volumes on my NAS is synced to a cloud backup service- all my personal data, the photos from trips to 28 countries, the video files stored in my Plex server, and most importantly the music.

Most especially the music.

I lost some of my music collection in The Great Crash of ’03 and ever since then I’ve been hypervigilant. My music collection is, unarguably, the most important data on my machine, and it’s stored in no less than five different places around the world. I will not lose the music again.

So that’s where I am now- waiting patiently for the new shell to put my drives in. The Ezri to host my symbiont now that Jadzia is gone. (I’m kidding again; I did not name my Synology after a Trill, but now that I’ve made the joke, I’m madly tempted to name the new one Ezri.)

Have you ever lost important data to a computer crash? What did you do?

14/52

---

Let’s call this an Appendix, for the technically minded folks who may be curious about some of the technology discussed in this post. Here are two quick notes:

• The failed NAS is a Synology DS415+. The flaw was in the Atom chips used by that model. I’ve chosen to replace it with a DS918+. While the DS918+ has the same number of drive bays, it will have eight times as much RAM and built-in support for a virtual machine, which I’m excited to play with. And before you ask why I’m not going to FreeNAS and playing with ZFS – while ZFS is undeniably awesome, building my own kit is expensive and time-consuming and it has no interest for me now. I used to assemble Frankenstein-computers from pieces-parts when I was younger. Nowadays, I slave over a hot server all day at work, and when I’m off work I don’t want to tinker. I just don’t.
  
  
• I use a variety of backup systems for my data, but the two most important ones are designated as most important because they’re off-site. If my apartment were to fall into a sinkhole right now, my data would still exist in the cloud. (And let’s face it, that happens a lot in Florida for some reason, almost like we’re on a Hellmouth.) There are several great backup services out there, so do your research. I personally use Backblaze for my main computer and Backblaze’s B2 Buckets for the individual volumes of my NAS. I’ve had Crashplan in the past, and the Java app was unwieldy. I find Backblaze to be simpler to use and leaner on my computer.