I was feeling generous when I wrote that.
Any updates on when we can expect the new app?
My guess is the second Tuesday of next week.
Is that the Tuesday before or after Wednesday?
Yes… (10 char)
And honestly, all this talk about secret codes and questions is a little silly. If someone wants in your home, a BRICK will get them into your home without having to buy spy gear.
Exactly. This is like the XKCD comic about how geeks think about security…
It was self invented, and recognized that theory and practical do not always have a contemporary clean meeting place and sometimes there are dis-contiguous solutions.
It is not a concept that escapes security theorists…
Multiple vectors, you mitigate the vectors you can, you accept the rest.
But ya… tis funny. I have this and several other Security related XKCDs on display in my office.
However, outside of the humor of the rubber hose attacks on security - just because someone can nuke the whitehouse does not mean they don’t need metal detectors at the door. Otherwise, I might suggest one should take the locks off their doors and see how that goes.
I’m curious how simple text based polling can overwhelm their servers. Firstly, they should have the polling servers separate from authentication and setup. And if they cannot afford the infrastructure then they shouldn’t offer it to begin with. If they do not want less than X amount of time in refreshes then they need to put a caching server such as varnish in place. Sucks that companies shut down their services to customers because their IT staff doesn’t know how to fully support their servers.
Well, to be fair, they never explicitly offered the ability to have their servers polled at all. And I don’t necessarily blame them - polling a server every few minutes to see whether or a not a door has opened or closed is not very efficient. But on the other hand, I really wish they’d come up with some better option (webhook or callback support, maybe?) rather than just shut it off.
I am not going to speculate as to Chamberlain’s server resources, but can say that when they turned off access we were getting errors in the hundreds of thousands every day.
A properly architected application strategy would make getting all of these poll requests no big deal. I’m sure they have built some monolithic solution to opening and closing garage doors instead of going the microservice approach. Do I blame them? Yes I do.
They are not a tech company. I am 95% sure they outsourced the app and the servers to manage this.
Just because they outsourced the work and have little pride in their tech is no reason to not blame them. This is likely easily fixable and they just don’t want to.
Everything is fixable, but the people writing the checks don’t understand and don’t care.
So I’m interested in this, but can live with the myq app in the interim. Did smart hub integration still require you to purchase the myq gateway if your garage door opener isn’t one with built in wifi? Thanks.
Yes - you need the MyQ gateway to make it work. It works great, just have to spend the $$
And to get around the lack of polling, I just added the pollster app and have it refresh the garage doors every five minutes. It add to my own hub, but at least I can use the monitoring and notifications for when we leave…or leave the door open for greater than 10 minutes!
Is SmartThings and MyQ still working on this? I would love to have some type of update as to when this might take place.
It’s not likely going to happen, at least anytime soon. MyQ wants to severely limit what ST would be able to call their servers for. The way ST is built it isn’t easy for them to limit what the devices brought in are called to do.
As an example(from what I remember about this topic): MyQ doesn’t want to allow Automation of their doors, only remote control. There isn’t a way for ST to limit calls to MyQ for Only remote control, and It would kind of defeat the purpose of integrating it.
So TLDR unless MyQ gets their head out of you know where, an official integration isn’t really beneficial to any of us. We would be Much better off pushing for custom smartapps and device handlers to run locally so that there isn’t a difference.
- Be able to send a command to lock it
Our garage door opener broke down two days ago, and we got a new one installed yesterday. The company only carried MyQ devices, so I figured I’d deal with it. I had been using a Linear GD00Z-4 on our old opener. I figured the tilt sensor on the door would still work fine, so I’d at least still be able to know when it was opened and closed. We ended up getting a Liftmaster 8500.
I see a post above about hacking the wall switch to use legacy controllers. No need to mess with that relatively expensive piece (Edit: The installer used an 888LM wall switch, which can be had for under $30. So it’s really not that expensive either way.) The installer gave us the remote that came with it, and a second Liftmaster universal remote, an 893LM. The buttons on the remote are just open/close switches, and you can get them for under $20 so if you mess something up it’s not that expensive an experiment. I soldered a couple of wires to the appropriate pins on the remote, and the GD00Z works perfectly fine with it. If you don’t want to ever have to deal with replacing the battery, you could also get a 3v power supply and solder a DC barrel jack to the battery contacts. I will probably do that eventually, after we get our second garage door opener replaced and I have to do this again for that one.
This doesn’t get you all the MyQ features, honestly I have no idea what they are. Access to the motion sensor on our wall switch would be nice, as would controlling the light, but it at least lets me get open/close status and open or close the garage door from SmartThings for now, and was relatively easy.