Kevin Connolly

I'm with ya. I built my first Linux box back when Fedora was Redhat V6. Can't stand Windoze. I was a little disappointed when Ubuntu didn't come to market with their phone and OS in any real way. In fact, I bought a Nexus 7 v2 with the intent to put Ubuntu on it, but that didn't really pan out. Now I use the device, mounted to the kitchen wall, as a home message board.

I can agree. Google is just about everywhere, but than so was Microsoft not so long ago. The interesting thing about Google products, at least for the the ones I use, is that they are never quite complete, but they offer more openness than either Apple or Microsoft. (It wasn't so long ago that Apple shutdown the Siri proxy hack whereas AutoVoice is offered on the Play store) Google's "Works with Nest" deployment is a good example. I mean really. What can you really do with the thing? I get the warm your home prior to arrival part (although I have yet to see it actually work with my stats), but the set temperature to a specific degree part? That part I don't get at all. I'm a "it could be a little warmer (or a little cooler)" kind of guy. I could care less what specific temperature it is. My guess is that they just threw something into the market ahead of CES.

Thanks for the reply. Nest is just a momentary popular device. I agree that it's not something that should abruptly change UDI's design direction. It's readily apparent, at least to me, that you guys put a lot of hard work into bringing a really great device to market. I'm not particularly worried about it. I have the Nest plugin for Tasker (which just got updated btw so that the variables actually work). Sure, it's rudimentary at best, but it works better than Google's own implementation through "Okay Google" so I guess for the time being I'm ahead of the game...and, since I have an ISY with a Zwave expansion board, I also have the best of all worlds

I saw that the "Works with Nest" website now includes Insteon's hub. Does that mean that the ISY is soon to follow?...Maybe?

In theory, it may work. It's a BT transmitter/Receiver, but not sure, if you got it working what kind of range it'd have. Also, with Linux hardware you always have to first do research on what's compatible and what's not. My BT dongle has a CSR chipset and I believe that's compatible out of the box. Heck, give it a try...it may work. On another note, when did you graduate? I'm BSCE 80.

I'm sure that would be a nice gesture. I really like the program. It's simple. It's elegant. It has a lot of things already in it, but it's also expandable...and the best part is it can utilize the Mac address of any defined BT device.

My initial request to flores/github didn't include the required user name and password part of the REST command string,but that should be easy enough to adjust. I do thank Carlo for everything.

I've been researching a few options for presence detection. My main criteria were that I wanted something relatively secure and something where I didn't have to lug my phone around everywhere. This morning, while waiting for the bank to open I looked down and saw my fitbit in a whole new light. It's light, I wear it almost all the time, and it's Bluetooth! Long story short, I did a web search on finding a devices Bluetooth MAC address and came across a ruby program by Carlo Flores on GitHub. I wrote to Carlo and asked if he could thin down his program for an ISY and below is what he wrote back about an hour later (I did ask his permission to post his response and I'll also send him a link): If you're doing multiple devices and zones (like many R-Pis or computers scanning bluetooth) it's easier to have it call to a webservice. In this case each of those devices would run bin/tracker-bt.rb and then send their information to the central webservice (webservice.rb). The central webservice then uses lib/location-by-bluetooth.rb to figure things out by zone. But your use case sounds possibly simpler than that, so I put this together. https://github.com/flores/lotomation/blob/7d7cd904cdf434ce1871393c72784180ed8fc028/bin/tracker-bt-standalone.rb In that script, you would replace "tracker_mac" with your FitBit's tracker and "device_endpoint" with your Universal Devices endpoint. It should let you control a single device by zone really easily, and you could run a bunch of these on a bunch of R-Pis or something per node. If you want some help controlling multiple devices (which would just be iterating over an array of those devices) if in a zone, let me know and I can expand it. If you like this I might just break up these different bits of lotomation into so Oh, and to run it you'll just need Ruby and hcidump. Something like this should install all the dependenices on a Raspberry-Pi that is already running Raspbian/etc. https://gist.github.com/flores/bc823e5ff2eb213a6a53

That makes sense from what I observed. I guess what confused me is that I was associating the programic ON/OFF with a paddle push at the device.

Roger that. Another thing that I noticed is that: if I turn on a light via ek ISY the light switches on but the other switches tied to it via the ISY statuses don't change? If I do it through the scene than everything changes correctly. I guess it's kind of like that old saying: Hey Doc, my arm hurts when I move it like this...haha. Just curious though why that would be.

Thanks for the home depot tip! Is there a lot of line noise with Halogen? I have some under cabinet lighting (that I inherited with the house) that's Halogen. I haven't put it on an Insteon switch as of yet.

Very good point.

Whew! Good to know. I have two line filters on some solid state equipment.Amazing how many times the filters led switches from green to yellow to red when the generator is running.

Has anyone had experience with dual band devices running under generator power? I know much is said about line noise and I also know that generators don't produce the same sine wave as my electric company. I live in an area that has outages maybe once a year (Sandy aside). It'd just add insult to injury if I could run the generator but not turn on the lights.

I'll check out the LEDs OFF Topic: I live in Sparta in Nothern Jersey.

Are LEDs any better...line noisewise. I live in northern NJ and I don't think you can even buy a incandescent light bulb any more.

The loads are either Halogen or Florescent with a dimming ballast. Where I live the electric company has a tendency to throttle so a dimming ballast is a must.

Thanks Lee. I thought I'd ask since the ISY writes to the device. It wouldn't be an unheard of thing to have a device take a write incorrectly or to have a device get stuck in a routine. I had a iPhone get stuck in a routine it was trying to execute with an Exchange server. Happened out of the blue and after several months of use, but it drained the battery so quickly that you could watch the numbers tick down and it physically got hot. I did a little internet research, wiped and reinstalled the Exchange account, and never had the problem again. But, on the other hand, if it's not possibly a write issue from the ISY, the comm values look good and consistent, and the wiring is tight than...? As far as communications is concerned, there's a difference between a consistent signal and a strong signal. Any way to test the signal strength and/or use an external antenna on the ISY?

Just to re-summarize the symptoms. I notice a device that's showing the wrong status. A button push makes the LEDS flutter and the device locks. If the affected device is a primary device it is still accessible through secondary devices. A hard reset (so far) seems to correct the issue. (Or at least I haven't had to reset a device twice, so far) Communication in the event viewer indicates for most cases max hops=3 hops left=2. (I say most because there was a section for the KPL that said max hops=1 hops left=0) The solutions thus far are: Factory reset all devices as you install them and make sure your connections are tight. Where there any issues with 4.0.5 that could have caused such issues that may have been fixed in some iteration of the current RCs?

I think so too. That's why I started with the event viewer after the reset. The KPL looks like it is consistently a one hop device.

So far, any affected device has been corrected by a hard factory reset. That said, I like to know the why's of things. I've had some good advice so far. Check communications to the device by the event viewer at level 3, Factory reset devices as you install them, and Check to makes sure connections are tight. Since the device(s) showed a misaligned status just prior to a flutter lock, is there anything else to check that could cause that?

Thanks Stu. If the neutral were loose though wouldn't resetting it than have zero effect? Ie. Wouldn't it still flutter? Personally, I thought that a status on the KPL got misaligned with the ISY or something along those lines. When it occurred with the other two Switchlincs the only reason I noticed it was the misaligned status (ie. The button said it was off when it was on or vice versa). That what happened with the KPL today. One of its buttons "C" said a light was on when it was off. When I pushed the button it fluttered and the device locked. The KPLs on/off is the controlling device to an overhead light and even though it wouldn't let me control anything at the KPL I was able to turn the light on and off from a secondary Switchlinc.

Let me clarify one thing. When I say the lights fluttered what I mean is on the device. NOT the load.

Ok. So l'll factory reset all my devices that havent already been reset. Other than a loose connection, could anything else cause the flutter lock symptoms. I want to use physically taking things apart as a last resort. Also, the Switchlincs I reset initially a few weeks ago haven't acted up since. One would think that a loose connection would make it fail regardless of a reset. I also checked the neutral bar. All connections are tight to both breakers and bar (I know they can have a tendency to loosen IP over time due to heat.)

The KPL chattered (as in chattered and lights fluttered)than locked. I've had two other switchlincs do the same thing. I haven't had a reaccurance with the other two Switchlincs since factory resetting them. I guess its been a few weeks now. The KPL acted up today. Although it could always be a possibility, I tried to make sure all connections were tight as I installed devices.