De-commingling (?) LAN equipment: It’s all in what you call it

A welcome career transition (and employer-responsibility expansion) begs for a hardware-plus-software evolution. Hold his beer; this engineer’s got this.

As some of you may have already noticed (assuming you even care about such things), my relationship with EDN recently (and happily) re-deepened. After being a full-time as a (senior, eventually) technical editor from 1997 to 2011, I returned beginning a year later, this time as a content contributor. And now I’ve added associate editor to my EDN repertoire.

“Wait,” you might be asking, “isn’t Aalyia Shaukat the associate editor at EDN?” You’re part-right; for nearly four years, she was. And for a couple of recent months, she (somehow) worked a double shift of jobs. But she’s now the full-time editor-in-chief at Power Electronics News, where she’s already rockin’ the house with her talent abundance. And I’m grateful to follow in her EDN associate editor footsteps, along with continuing my own frequent content-contribution cadence.

What’s this all got to do with “de-commingling (or if you prefer simpler vocabulary, “separating”) LAN equipment”? An excellent question. Now that I’m more intimately interacting with the EDN website and other publication (and publisher, and corporate owner) resources and services, I needed to set up a standalone computer so that nothing attacking my home office LAN could make its way to the corporate network and other facilities, too. That said, I remained heavily broadband-reliant. And I wasn’t up for setting up a completely separate Comcast service connection just for a single (albeit also a singularly important) computer. What to do?

Just call me “guest”

That last part was actually the easiest part to solve, it turns out. My home LAN, as mentioned before, is based on a multi-node mesh implemented using multiple Google Nest Wifi routers, with the primary one connected to the cable modem in the furnace room.

One nifty nuance of the Google Nest Wifi system (shared by not only other Google LAN equipment generations and gear from other suppliers, mind you) is that you can set up a distinct “guest” network that by default (which I’ve left unchanged in my case) is packet-isolated from the main LAN beyond their shared WAN connection.

The computer I’m dedicating to my EDN associate editor work is one you’ve seen before; a Microsoft Surface Pro 7+ (SP7+):

along with my longstanding tech-gear companion, a Kensington Dock:

mated as so:

LAN-migrating the SP7+ was easy-peasy. I disconnected the wired Ethernet cable from the back of the Kensington Dock, switched the computer from my main “RockyMountainBri” wireless network to “RockyMountainBri-guest”, and…that was it. And since my Brother multifunction laser printer was right next to the computer, I didn’t even need to bother migrating the wireless network that the MFC was connected to, foregoing printing support for the rest of my LAN in the process. I just ran a USB cable from the Kensington Dock to it, and…I was done. Perhaps obviously, by the way, any real guests are no longer able to use my “guest” wireless network.

Split personality

How do I handle the fact that, still acting as a contributing editor along with my other contributor colleagues, I’m now in effect submitting content to myself for subsequent publication, now wearing my associate editor hat? My contributing editor workflow is unchanged, actually. The only thing that’s different is the email address I now send my stuff to.

It used to be that I’d submit content from my personal email account to Aalyia’s corporate email address. Now, instead, it’s my corporate email address that the goods go to. I’m still using one of my other systems for initial writing—typically but not always a Mac. But, to maintain “firewall” purity between my newly transformed associate editor work system and the rest, I exclusively receive corporate email (and don’t send or receive personal email) on the SP7+.

Going loc(al, not o)

And what about backing up and archiving all this content I’m now receiving? Regular readers may remember that I’ve long been a fan, along with a frequent implementer and upgrader, of network-attached storage (NAS) for such (and other) purposes. That said, unless I wanted to dedicate a NAS solely to my “guest” network and connect it exclusively over slow Wi-Fi, I was going to need to transition to some other solution.

Therein lies the admittedly and intentionally somewhat obscure title of this piece. Instead of network-attached storage, I wanted something locally tethered. It had to be at least dual drive configuration, with RAID 1 support so I didn’t lose everything if a hard drive died. And ideally it’d run hardware RAID to avoid bogging down the computer. Yes, I know, if the RAID controller fails, you’re dead in the water, too, which is why I also wanted something that was reasonably popular. So that I could, if necessary, find a replacement to slot the HDDs into without too much trouble.

I figured I’d start my search using the term “DAS”, for direct-attached storage. Interface technologies I’d used in the past—Firewire, Thunderbolt, and eSATA among them—weren’t relevant to this particular hardware configuration, so I settled on USB 3.x, as fast a flavor as possible, over USB-C. My (perhaps imperfect) search yielded exactly one result, QNAP’s TR-002, which ironically is primarily intended to capacity-expand the company’s NASs but can also find use as a standalone storage peripheral.

Tomato, tomahto

At this point, I reset my lingo-options list, expanding beyond “DAS” to also include “enclosure”. That change helped a lot from a results-options list length standpoint. What I’ve ended up with is the Mercury Elite Pro Dual from a company I’ve mentioned multiple times before, Other World Computing (aka, OWC) and bought open-box (with 1-year warranty) for $167.50.

It’s hardware RAID-based, supporting four different operating modes (albeit only one at a time):

  • RAID 0 “Drive Striping”
  • RAID 1 “Drive Mirroring” (the mode I’m using)
  • Span, and
  • Independent Drives

Its interface to the computer is 10 GBps USB 3.2; perhaps obviously, I’m direct-connecting it to the SP7+ versus going through the Kensington Dock intermediary. It also embeds a three-port hub, a particularly attractive proposition given the SP7+’s dearth of integrated connections. And here’s a rarity (as I’ve written about before); the hub’s USB-C and dual USB-A ports are all 10 Gbps peak bandwidth-capable, too.

Why, you might be asking, did I go with HDDs instead of SSDs? I’ll turn around and ask you a question in response to yours: have you priced SSDs lately? That said, HDD price tags are also skyrocketing lately, although they still hold a tangible edge over solid-state alternatives especially at higher capacities. And in my case, I thankfully was able to repurpose a couple of spare 3TB HDDs I’d already bought in the “before times” and still had sitting around unused (I’ll have more to say here in an already-planned upcoming follow-up post).

Software completes the magic trick

The last, but not the least, question: how to integrate it with my computer for mirroring and broader backup purposes? I planned on consistently using the SP7+’s upgraded-by-me 1 TByte SSD as primary storage of in-process and completed associated editor work, so one-way mirroring (versus two-way syncing) that portion of the SSD to external storage would be fine.

But I wanted that mirroring to be file-by-file, not lumped together into some unified-file or otherwise nonstandard format (Apple’s Time Machine, for example) that is be difficult to resurrect the contents if primary storage in the computer failed, say, or if I needed to physically pass the external storage device to someone else. And, of course, I’m also looking for cheaper solutions, so open source or another free source would be best.

I found my solution in a two-part open-source program suite, developed and maintained by the FreeFileSync project and supporting Linux, MacOS and Windows platforms. FreeFileSync itself does the sync-and-mirror heavy lifting for both files and the folders containing them. And the closely related RealTimeSync monitors directories for content changes, which then kick off FreeFileSync (or any other operation more broadly).

This discussion thread was very helpful when I was setting up RealTimeSync and FreeFileSync on my system. And ever since then, it’s run like a charm; the only time it pauses is when it detects an abnormally large number of changes (multiple directories-and-files moved at once) and wants my OK before it proceeds.

Oh, and by the way…since I’ve got plenty of empty capacity available, at least at this early stage in my associate editor career, I’m also using the OWC Mercury Elite Pro Dual more broadly as a successor to the NAS for my ongoing computer-wide backup purposes using Windows’ built-in File History and (deprecated but still functional) Backup and Restore facilities that I’ve mentioned before. With that, I’ll wrap up for today. I hope what I’ve shared will be of help to at least some of you in similar configuration situations either now or in the future. As always, please share your thoughts on what’s worked (or not) for you in the comments!

—Brian Dipert is the associate editor, as well as a contributing editor, at EDN.

Related Content

The post De-commingling (?) LAN equipment: It’s all in what you call it appeared first on EDN.