apostolakisl

@JTsao Opposite direction. I actually have that and am using it, but that is for ISY to send IR (via Itach transmitter), I need ISY to receive IR. I have a Harmony Elite but also have a Harmony 900. The 900 does not have a network connection so the Harmony node doesn't work as with the Elite. And now Logitech doesn't make them anymore. I could probably get another Elite used, but man, so much work. It would be so much easier if there were an IR receiver for IoP.

I am considering moving ISY to Polisy, but I am using the IR receiver on the 994i unit and don't see that Polisy has any option to receive IR. Ideas?

Success. I moved polisy to the other vlan, it found Roomba, I moved Polisy back to the secure vlan and it was able to maintain connection. Thanks.

With ISY, you don't need to know the Insteon address. It has a discover mode. You just click on the auto discover button on the ISY menu, and then press and hold the discover button on each device. You can do a whole bunch at once. I would start by going through every insteon device in your house and pull the tab out, then press in and hold until the long beep quits. This factory resets. After that, put ISY in discover mode and press the little button on each device and in a second or two ISY will discover it. Label the device in the ISY app. Before you start, I would create a plan for how you will name things and create a folder tree. Life is much easier later on. I like to add an "L" to the end of the name of every device that manages the actual load. Once all of your devices are part of Insteon, then start creating your scenes. You will find that ISY allows the creation of extremely complex scenes, if you like. It can get confusing because there are so many options. But in the end, you will appreciate it. Understand that each scene has one or more controller devices. A device can control only one scene, but respond to as many as you like. Each device will also have local settings. Meaning that when you push that device's switch paddle, the local settings are what happens, any scene to which it is a responder or controller is irrelevant, the local settings is what that device does when you control it directly (locally). Then there are the settings when the device is the responder to a scene. They can be the same or different as the local settings. Lets take an example, you have a 3 way. Each device is set as a controller to the scene. You want all 3 switches to do the same thing. Within the scene, you set each device to respond as you like (all the same). You then need to set each device independently to do the same thing when locally controlled. The beauty of this is that it doesn't have to be that way. You could instead make something different happen when you touch each of the three switches. A single scene with three controllers could do different things when using each of the 3 switches, if you like. Anyway, in short, factory reset, add devices to ISY, and have a clear plan of organizing them before you start.

The router wouldn't have any issues. I have set it to allow com between polisy and roomba bridging the two vlans. The only question is if the node server will try to re-discover roomba when (if) it senses that its subnet has changed after I return it to the more secure vlan.

Would it keep the ip address if I moved polisy to the Roomba subnet and then back again?

I would install a proper security system based smoke detector which can then easily be bridged for "informational purposes only" to your ISY via lots of methods. DSC alarm panels are quite inexpensive and UL listed and all. The smoke detectors themselves will likely cost far more than the alarm panel depending on the quantity. Linking DSC to ISY can be done with polyglot and an IP adapter for the DSC panel. If you already have a standard smoke alarm system that is just a local, interconnected, siren only type system, often they do have models with relays that can bridge to an Insteon, Zwave, or other IP based contact closure device that will link to ISY.

@bpwwer I have an i7. I keep all of my iot stuff on a different subnet that is firewalled from my important stuff. When needed, I open a path between two devices, which I have done between polyisy and roomba. I had the pg2 roomba node working, except it really doesn't work on the i7 very well at all. I also have polyisy talking to my blue iris server using the same firewall hole technique . . . so I am confident there is no firewall problem. I was hoping to use your pg3 node which appears to be expressly made for i7, amongst others. But the autodiscover does not seem to be looking across subnets, the node server log lists it as giving up after several tries and not discovering. Anyway I can tell it where to look or set the ip address manually? Thanks.

OK, getting them now. Sent a note to support.

"long time". Not sure what that means. But I am now at 3 hours.

I don't know. Not getting any email. Checked spam and all. my login name is my email address. Searched my email for that ircodes@. .. email and I see they sent me one 5 years ago. So I must have been playing with it back then. but not working now.

Not that I need any right now since I already manually learned them all on my own, but how does the website work? I already had a user/pass with them. I click to get it, it says "sent" . . . sent where? I didn't get anything in my email, no download, where? And as it turns out, the primary device I use GC for is not in the database. So I'm copying them from where I figured them out into the format GC wants. I am seeing that you did it this way because I assume it is how they come formatted from GC.

Deleted ISY from PG3 and re-added. This was my very first installation of PG3, so I had no nodes installed to screw up by doing that.

First off, very happy to have the node server. Not complaining, trying to be constructive. So many different situations with node servers because of the vast differences in node specific needs, so my comments are more to the general construct of PG, not to this node. I get it, PG is this generic template. The whole concept of the manually entered key is just not a user friendly situation. It isn't the end of the world, but it isn't really a GUI and everyone wants GUI's. Lots of directions to read and then you must manually type in "keys" precisely as expected and wondering "was the colon part of the key?" type stuff. The general format for these sorts of things are that all the "keys" are already there. Makes it obvious what info needs to be entered for values. With this particular node server, the value has the potential to get out of hand. For example, lets say you have 25 ir codes for a device, trying to input that into the format of "key" / "value", well that "value" line will be insane. You'd have to do it in a text editor and copy/paste it over. Lots of opportunity for screw ups, getting all your commas, quotes, and spaces right. Which is where I am now as I am adding my 5th "button" to the same device and plan to add about 15 to that device, with others to follow. Within PG, a system where you have radio buttons to push "add device", then "add code" would be more typical GUI stuff. Obviously, with iTach, you have these crazy *** codes that are generated for each and every ir message in the world. And unlike every other universal remote, which keeps all that code internal, itach keeps none of it onboard, instead requiring that you send the crazy code with each request. So, yeah, tough. Now I don't know anything about "control tower", I have actually been using my itach for a good 7 or 8 years as is without need to change anything, so maybe this is something new they added to simplify the ir code retrieval. All my codes I had to generate myself, which was a PITA since often times they didn't work. A lot of trial and error. What I have done here is just copy/paste the codes over from where I saved them on ISY network resources. I totally didn't need to install this node server, but I prefer using nodes to network resources, and I am also a glutton for punishment . . . screwing with things that work perfectly fine and breaking the golden rule of "if it ain't broke, don't fix it". Oh well.

The screen shot was the most helpful, but you can't put that into the Node server display, I'm sure. Where things get unclear is whether things like "" symbols, or , symbols, or spaces are part of the language of explanation or part of the actual parameters. Often times people use the <> symbols to make that clear or if you could change text color or similar to make it clear what actually goes there. Finally, just a general polyglot comment. For this thing to really go mainstream, need to get rid of all this manual configuration of "key' and "value". Needs to have custom menus for each node server with pre-labeled slots, drop down's when appropriate, and then once configured, have it displayed in a more user friendly format. Basically, something like the polyglot ISY config page where it has pre-labeled slots for all the parameters like user, password, ip, port, etc.

Just spent some time with some trial and error. Anyone setting it up might find my experience helpful. The basic theme is that each device/ virtual remote occupies a single custom parameter. So if your TV has 20 buttons to control, it is just a single line with the "key" being the name of the remote (ie TV) and the "value" part being a very long single line of 20 commands one after the next. The format allows for 2 codes per button. If you put nothing for the second code, it just does the same as the first code. You can not eliminate code 2 from the drop down menu by not including "code2, hexcode2" in the preamble part of the "value" section. Here is an example of 2 buttons I installed for a single remote. Note the double space between the two buttons. The precursor part of <function, code1, hexcode1, code2, hexcode2> is just entered once at the very start of the "value" section of the custom parameter, from there on, a double space between each button setup. function, code1, hexcode1, code2, hexcode2 "whole house on","sendir,1:1,1,38580,1,1,19,177,19,177,19,177,19,178,19,178,19,178,19,274,19,178,19,178,19,178,19,274,19,3858" "whole house off","sendir,1:1,1,38343,1,1,15,180,14,181,19,176,19,176,19,176,19,272,19,272,19,176,19,176,19,177,19,272,19,4907"

Never mind. I deleted it and re-added and now it shows the nodes.

I did put the isy info, ip, username, password, and it says "connected", since I don't know what else it would connect to, I assume this means ISY. It also lists ISY as 5.3.3 which is correct along with the correct UUID, neither of which were entered by me but rather discovered.

I just installed and ran PG3 on polisy. It appears to have connected to my ISY, it says "connected" and it lists the uuid of my ISY in the settings. However, on the dashboard, it lists no nodes, unmanaged or otherwise. When I try to add a node , it asks me what slot and the drop down makes any number I want available. Should it not see that I already have nodes (from PG3 and IoGuy) occupying many of those slots?

I have been having a similar issue with my Husqvarna mower node.

What you are asking doesn't make any sense. Are you asking if there is an IP address to control your switch through ISY? If so, then yes, it is all documented in the REST command instructions in the wiki.

If you are asking if you can know the Insteon address of a switch that is not already added to ISY from ISY, the answer is no. If the automatic discovery method isn't working, then I suspect your switch has comm issues. Of course there is always the sticker on the switch, but that can fall off. Anyway, in short, if using the auto discovery doesn't work, odds are good that manually entering the address isn't going to work either.

I don't get a whole lot of snow here. I would speculate that snow at night would be the same issue as rain at night with a camera that "shines" IR light. Maybe with enough snow it would all just be white and not trigger anything though.

Not much ambient light where I am. Unless the moon is bright. I have decorative landscape lights only. Putting out general area lighting doesn't appeal to me from an aesthetic. I suppose a real IR camera, a thermal camera in other words, would do the trick.

And another nice feature of BI is that the same camera can be used multiple times with different programming to create different triggers.