alarmdecoder.com

Poking around in the forums here, I've noticed a bunch of examples of storing a user code somewhere in a configuration file so that an automated process can "type" the code into the panel for arm/disarm purposes. From a recent post:

Code: Select all

device.send("12341")
device.send(AlarmDecoder.KEY_F1)

I'm not comfortable storing a user code in a text file on my server. Is there any reason not to configure an AD2* virtual expander zone as a keyswitch, and arm/disarm the system that way? It seems like the sane way to do things, but since the herd appears headed in the opposite direction, I figure I've overlooked something.

Thanks!

pp6knjoR5puA8eps wrote:Poking around in the forums here, I've noticed a bunch of examples of storing a user code somewhere in a configuration file so that an automated process can "type" the code into the panel for arm/disarm purposes. From a recent post:

Code: Select all

device.send("12341")
device.send(AlarmDecoder.KEY_F1)

I'm not comfortable storing a user code in a text file on my server. Is there any reason not to configure an AD2* virtual expander zone as a keyswitch, and arm/disarm the system that way? It seems like the sane way to do things, but since the herd appears headed in the opposite direction, I figure I've overlooked something.

Thanks!

This was just an example of how to send keys to the device from the AlarmDecoder python library - how you choose to retrieve the code for sending through the library is all up to how you want to do it

how you choose to retrieve the code for sending through the library is all up to how you want to do it

Thanks, Kevin.

I want the system to be able to arm/disarm without my typing a code. I could probably come up with in-memory user code storage, which would satisfy my compulsion to not leave secrets on disk, but I'm not sure it's worth the effort. Does arm/disarm with a code offer any advantage over doing it with a keyswitch?

I know that with LRR support, you can see which user disarmed the system when using a code - I do not know if your proposed method allows for this or not, but that's the only really neat thing I can think of as to why using a code would be beneficial.

Thanks again, Kevin.

I'm planning to leverage the LRR support to know who is coming and going, but it's only interesting when the arm/disarm comes from something other than my AD2Serial.

When the computer is the one arming the system, the computer knows that it did it.

I don't have any real LRR device. I suppose that would be an important distinction. The monitoring company can't tell who's responsible for disarming the system if the event comes from a keyswitch zone, right? That's a good point, thanks for raising it, because it hadn't occurred to me.

I'm just getting started with developing my system, don't want to head down roads that turn out to be dead ends. Thanks for bearing with me.

My plan for the LRR user info is:

- static IP addresses (DHCP reservation) assigned to the WiFi radio on each alarm user's smartphone
- a table that maps each alarm user ID to their smartphone's reserved IP address
- a table that maps each alarm user ID to their smartphone's phone number
- a table that specifies which users get notified about which other users arm/disarm events

Prior to sending an SMS, try to ping that phone on the WiFi. If the phone responds (it's at home), don't bother sending the text.

This way, when I return home with one of my kids, and the kid's code disarms the alarm, I won't get a text about it.

OTOH, if I'm at the office, the ping to my phone will fail, so I will get a text about my kid's arrival/departure.

If only 5800* buttons can be used to arm/disarm I wonder would it be possible for me to spoof an 5800 receiver message that a given RF serial was activated and thus disarm using unique serial numbers of devices.

This would allow for assigning a user to the keyfob serial number and then a simple way to disarm the system with a single command and still know who did it. It would require using an RF zone and some unique serial number and then assigning that button to a user code.

I will have to investigate this in my lab.

Re
Sean M