Remote ST Deployment -- What are YOU Doing?

My ST hub is deployed at a second home over 1500 miles away. My primary motivation is remote climate monitoring: specifically humidity and high temperature. That lets us keep track of heating & a/c, plus intrusion detection while we are away from that location. The system was deployed when we were there in December 2015 and has worked well since then.

[As a side note, here at home I’m anxious to replace my ADT Pulse system with something that looks like it was developed this century: home automation in Pulse is abysmal. I hope to have a SmartThings hub in place later this fall for this location to monitor things here at home while we are at the other place!]

My remote system consists of the Smart Things v2 Hub, a switch on the front door (it is a 2nd story condo in a gated community, so other points of entry aren’t as much of a concern) and an Aeon Multisensor 6 inside. (There are other things the system does while we are in residence, but the ‘monitoring’ part of the system is what makes this remote deployment useful and interesting.)

The Internet connection is a Verizon MiFi 4G LTE device (we don’t have a cable modem yet and it may not be active all year when we do). The cellular connection may eventually become the “fail over” path for a router also connected to a cable modem. The Verizon MiFi box was powered up in December and has been on-line ever since. When I was on-site in February it was still ticking and I didn’t touch a thing!

Two things I had originally hoped for in the Smart Things v2 Hub apparently are not available at present:

  1. I would like to assign a static IP to the hub and was told by support that feature is (apparently) not available. Not a huge issue at present that can probably be solved in the long run as I build out the network on-site and have a ‘better’ router that has more configurable options. The MiFi is pretty simple in that regard!

  2. After inquiring whether the hub had any kind of ‘watchdog’ capability, I was led to believe the answer was, “no!” [From my experience, a hardware watchdog and a software process to monitor network connectivity would go a long way toward long-term stability & the ability to recover from situations that stump the hub!] On a remote deployment such as mine, rebooting the hub isn’t always an option.

So, I installed a Synaccess NP-02 remote power switch – luckily I had one on-hand, as it costs more than the ST hub! The Synaccess switch faithfully reboots the hub every night: it’s rude, but it works and I’ve had no issues. Had the hub been assigned a static IP address, the Synaccess switch could simply ‘ping’ the hub rather than doing a nightly brute-force power cycle reboot. Prior to deploying the equipment, a pushed hub update caused my system to freeze up – luckily I caught it while testing here at home and could put together a workaround.

What are others in the community doing with remotely deployed systems?

The next chapter of my remote deployment story happened a couple of weeks ago. Since I wrote the earlier story in April 2016, there have been no hiccups at all in the monitoring that I have noticed. SHM has notified me (correctly) each time the exterminator has entered the premises to treat for bugs, too. [If I had a camera, I could have watched…]

So, while spending a couple of weeks there in July 2016 painting & fixing up in anticipation of a longer stay this winter, I added a GE Fan Controller to the mix. It has Smart Lighting automatons to run a ceiling fan for an hour at 8AM and again for an hour at 8PM to ‘stir’ the air when the a/c hasn’t seen the humidity rise enough to run.

Alas, the fan automation doesn’t work too reliably: support (and engineering) are involved and indicate they are working on the issues. This is apparently another case of scheduled events not executing that many of you have experienced. FWIW, the environment monitoring (mainly temperature & humidity) seem reliable – but, they are asynchronous Z-Wave events, not ticker!

EDIT: the fan is running again without issue after ST deployed some code fixes. [09/02/2016]

EDIT: another round of instability and missed fan events occurred in late September 2016, but have again been corrected with no changes on my end. Problems in the cloud were fixed by ST. [10/12/2016]

At this time it looks like I will not be replacing the ADT security system here at our primary home: SHM is just not ready for prime time right now.

My second hub is in place to do interior monitoring of temperature & humidity, as well as some lighting automatons to give the place a ‘lived in’ look while we are away, albeit currently not 100% reliable due to the same scheduler issues.

EDIT: as of this date, scheduled lighting is again reliable. I’ve since also integrated monitoring of our wired smoke alarms. [10/12/2016]

It would be very interesting to hear from others doing similar implementations at remote locations!

No remote location yet! Will post when I do. (Starts looking…)


Hi John,

I know this is an old post but I have a similar situation albeit only 600 miles away :slight_smile: and am hoping you can shed some light on your remote location experience.

I’m currently a Wink user and am thinking of switching to ST for a number of reasons. One thing Wink has going for it is that I receive notifications about updates but they don’t happen automatically and I decide when they’re applied. I usually wait until I’m on-site to do them–that way if it all goes to crap I can power cycle the hub. No such option for the ST hub so I’m looking at a network controllable power switch or strip and another ST user suggested I look through project_secondhome for similar solutions.

I have an always on PC with BIOS settings to automatically power on after a power outage. We have a full time broadband connection and I can VPN into the computer using RealVNC. I can pickup a used Synaccess netBooter NP-02 on ebay for relatively cheaply. My thought was that if the hub were unresponsive I could VPN in to the PC and control the NP-02 using Putty to telnet into the thing. Since you’re already doing something similar I thought I’d reach out.

Any info, experience and ideas you can provide would be greatly appreciated :smiley:

Hi, @SteveV — the hub rebooting has worked fine for me. No problems have occurred with updates anyway, so it is really a fallback I haven’t needed. Yet. Environmental monitoring has been pretty solid and I get the occasional high humidity warning, so it is doing what I needed. Before we left last spring, in April ‘17, I added two more GE fan controllers, adding both bedroom ceiling fans to the mix.

That being said, once this last summer my Verizon MiFi 4G LTE internet connection needed a power cycle and I persuaded my homewatch guy to do it for me. Other than that, it has been solid for the entire deployment. Note that if I’d had a reboot-router-on-Internet-failure setup, it might have corrected itself.

The only tense time was when the power outage with Hurricane Irma happened. Within a few days my hub came back online, reporting high temperatures & humidity inside the condo. I watched as both went down to ‘normal’ levels (adding additional fan cycles temporarily), which proved the A/C system also survived. When my homewatch guy got back to town & checked, all was well…

Like my setup, your weak point might be the Internet connection. But knowing that, you may be able to plan accordingly…

Good Luck!

Thanks for the info. I was able to pick up an NP-02 on ebay for under $40 shipped. How are you using the NP-02 to reboot the hub nightly–is it something that the device can do on it’s own or do you have a computer running a scheduled task, batch file, etc?

Yes, the loss of internet and the resulting loss of knowing whether the heat or AC is working is the thing that has be the most worried. Unfortunately, I don’t have a backup plan. I’m with Cox in a single ISP market and am paying through the nose; $80/mo for 50Mbps. Both my wife an I work remotely when we’re down there and need a relatively fast connection. The slower tiers aren’t significant cheaper and with no other ISPs we’re stuck with Cox. From a failure standpoint it’s pretty unlikely Cox will be down for an extended period of time. The more likely mode of failure is a hung modem or router the requires a power cycle to cure–not really sure how to deal with that without going to some kind of expensive 4G failover.

The NP-02 will do it on a timed schedule, which is what I’m using. (In the past I used a number of these at work to manually control equipment at remote locations over the corporate network that I was using while developing firmware,) This is the first time I used the timer functionality and it seems stable. Log in to the web interface of the switch and you can set up a schedule…

The NP-02 is supposed to be able to set its clock via NTP, but I wasn’t easily able to get that working and, well, just gave up at the time.

Originally, I had hoped to be able to control the NP-02 remotely, but Verizon uses double NAT and I was not able to easily set up any kind of dynamic DNS so that I can get at the switch from the outside. Hence, the brute-force nightly power cycle.

The unit claims to have “Auto ping and power reboot for each power outlet.” I should probably try to get that working and put the Verizon Internet box on that outlet (keeping the first outlet for the nightly reboot of the SmartThings hub). Perhaps I’ll give that a whirl this winter.

There are a number of dedicated “modem reset” plugs out there that perform the same functionality. Reset Plug is one (see; there are others available on Amazon and elsewhere. YMMV.

Keep me in the loop, Steve, as I’m interested in fine-tuning my setup!

Thanks for the info–will definitely keep you posted on my progress.

1 Like

Update on the NP-02 Synaccess power switch:

  1. it now is getting time via NTP and
  2. it now ‘pings’ the outside world occasionally and reboots the Verizon MiFi 4G LTE device if it fails for some length of time.

So far, the NP-02 was a valuable addition to the remote configuration!