Chris Jahn

The problem is that it will use the event that caused the program to run, and in this case, that last event will be an armed up event. You may be able to accomplish what you want by splitting it into two programs, but its going to be tricky. I think you would need the first program to have only armed conditions and a 10 second wait (similiar to what you have), and have a second program stop the first program within 10 seconds if it doesn't actually arm. Aha, of course you could change it to use the actual area, since you know it is 'Williams', assuming 'Williams' is area 1, the variable would be ${elk.area.1.armedState}

The wiki for the Elk Module: http://wiki.universal-devices.com/index.php?title=Main_Page#ELK_Security Connection problem summary checklist: Elk Firmware [*:33lssmgs]Elk M1 minimum level is either 5.2.4 or 4.6.4 [*:33lssmgs]Elk M1XEP minimum level is 1.3.28NOTE: Most development/test was done using Elk M1 5.2.8, Elk M1XEP 1.3.28 Elk RP2 NOTE: Make sure to Save/Send as you make your changes [*:33lssmgs]Enable the non-secure port for M1XEP [*:33lssmgs]Go to Account Details -> M1XEP setup -> TCP/IP Settings [*:33lssmgs]Select the Enable non-secure port Checkbox[*:33lssmgs]Go to Users [*:33lssmgs]Select the user that you want to use for the ISY connection [*:33lssmgs]Ensure the user has a 4 or 6 digit access code defined [*:33lssmgs]Select the following checkboxes: [*:33lssmgs]Arm [*:33lssmgs]Disarm [*:33lssmgs]Bypass[*:33lssmgs]Make sure the following are not selected: [*:33lssmgs]Access [*:33lssmgs]Go to Globals [*:33lssmgs]Select G29-42 tab [*:33lssmgs]Select all checkboxes in Serial Port 0 Transmit Options ISY Elk Configuration [*:33lssmgs]Enter the access code you have for the user in ELK RP2 (either 4 or 6 digits) [*:33lssmgs]Use IP address and non-secure port defined in the M1XEP setup

Hi killervette, The program will only run when the status changes, similar to that of status for lights etc.. This program will turn 'My Light' on while the zone is violated, and turn it off when it is not. If Elk Zone 'Zone 001' is Violated Then Set 'My Light' On Else Set 'My Light' Off

Hi ryarber, Yes, you should see some or all outputs 1-64. The way to make them show up is to give them a name in Elk RP and select "Show" (in Automation -> Outputs). Make sure to save/send to control. When you are finished, do a "Refresh Topology" from the ELk config page in the ISY Admin console.

Hi Frank, Those cryptic Zone messages are the current voltages for each zone in decivolts. The voltage for a zone is sent whenever a zone voltage query is performed and the voltage actually changed. eg. Sun 11/27/2011 11:54:33 PM : [ELK 0 50 0] Zone : Garage Dr #1 Up : (53/134/0) Zone : Garage Dr #1 Up has a voltage of 13.4 volts (134 decivolts)

Hi Tim, The problem was that device database corruption could occur (rarely) at times with a lot of SD card activiity at the same time you are modifying links. This could result in all the links being removed from the copy of a device link database on the ISY. We think we've finally solved the problem with the upcoming build. The KPL linkage problem description is incorrect; the links are fine. The problem was that some of the changed bytes were not being written to the device. A restore link, or restore device was the work-around.

Hi Teken, This is fixed in the next build. The line separation has been removed and it is also now scrollable.

Hi tim2u, I think you added it as a RemoteLinc, not a RemoteLinc2. Best thing to do would be to remove it, and add it back again using "New Insteon Device". Looking at the dropdown menu, I can see how easily this mistake could be made. I'll have a look and see if we can make this a little nicer in the GUI.

Hi Illusion, Don't worry about the update pending flags that you are seeing. All the writes are being performed, its just the count that gets sent back to the GUI that is incorrect. This will be fixed in the next release. You can also restore a single link by selecting "Restore link between xxxxxxx >" on the right+click menu for a member of a scene. This is useful for IRLincTX because of the large amount of data stored in a single link.

Hi IndyMike, I'm assuming program XXXX is the program itself containing the status conditions. The program you have will run every time a status changes, regardless of the program true/false state. Therefore with each status change, it will alternate running the then/else unless the status of all the devices in the if are Off. As for the case where the then clause doesn't run, could you post the exact programs you used to test this. I cannot recreate that problem here.

Thanks Tim, I managed to recreate the problem and it looks like you actually discovered a bug that has been hidden in the code for quite awhile. The problem occurs if you have more than one close parentheses in a row with no condition between them. I have a tentative fix that will likely be in the next build. Until then, the temporary work-around is to ensure you have a condition between any two closing parentheses. For example, you could change your program from this: ... And-( | $sAlarmSys_Z05_SlidingDoor is 1 | Or $sAlarmSys_Z08_KitchenWindow is 1 | Or-( | | $sAlarmSys_Z06_MasterWindow is 1 | | And $sAlarmSys_Z26_MasterBed_Door is 1 | -) -) ... To this: ... And-( | $sAlarmSys_Z05_SlidingDoor is 1 | Or-( | | $sAlarmSys_Z06_MasterWindow is 1 | | And $sAlarmSys_Z26_MasterBed_Door is 1 | -) | Or $sAlarmSys_Z08_KitchenWindow is 1 -) ...

Hi Tim, What build do you have installed? There was parentheses bug when using variables in conditions in build 3.1.1 and earlier.

Hi ergodic, We don't consider this a change because all variables are initialized at startup, therefore every program referencing a state variable in its conditions would be run at startup. In other words, we do not know if the value of the variable actually changed from what it was when the ISY was last powered off and/or restarted. Is there a specific problem this is causing?

kingwr's assessment of the issues using a tiny wait interval in a loop are correct. There should be no issues with this approach as long as you keep the wait interval as high as possible. I wouldn't worry about it at all if the interval is above 15 seconds +/-, depending on the number of programs you have, and how many are running. For both accuracy and efficiency, the rule of thumb is to make the wait interval as high as possible.

Hi Clarence, There was a bug in using And(), Or() in a previous beta release, so make sure you are using the most recent beta firmware. As for your specific problem, I'll have to test this out because If xxxxx And yyyyy And zzzzz Then Should behave exactly the same as: If xxxxx And ( yyyyy And zzzzz ) Then

Here's the startup sequence: Step 1. Variables are loaded/initialized Step 2. All programs are initialized to false Step 3. All conditions are initialized based on the current status of the devices, programs, variables etc. (the programs do not actually run). Step 4. 'Run at Startup' programs are run Step 5. Programs run as usual

Hi Vipola, Lou, There is no technical issue in doing this, just prioritizing. The most likely way we'd do this is to add tabs on the program details screen that would allow you to alternate between viewing the program content (if/then/else) and the summary of the programs within the folder. All of this would be done on the details page, using the same tree.

It would take about the same amount of memory. The point of the post really was that you can do this now, although more direct support for 'while' may be added in the future.

We may add that at some point, but you do have most of that functionality now by having a program do a 'Run If' on itself, essentially using the conditions in the 'If' as the 'while' condition. Pgm: 'Bathroom Timer' If $BathRoomLightCount > 0 Then Wait 5 minutes $BathRoomLightCount -= 1 Run Program 'Bathroom Timer' (If) Else - No Actions - (To add one, press 'Action')

Hi Nick, Yes, this is intended and is part of the initialization pass for programs.

Thanks Rod, This is a bug, and is fixed in the next build. The problem was that the selection box was not being cleared before being repopulated.

Hi tim2u, Thanks for testing out the new variables feature. Could you please post your programs that are using variables as there should be no performance degradation at all.

Hi Illusion, One of the existing rules is that if the body contains no data, the email is not sent. In your example, the value for ${alert.control} turns out to be empty, and since there is nothing else in the body, the entire body contains no data, and therefore the email is not sent out.

Don't worry about these messages, its logging a little more info than it should.

Lets move the questions to the variables forum: http://forum.universal-devices.com/viewforum.php?f=68 kingwr - http://forum.universal-devices.com/viewtopic.php?p=45801#45801 bpwwer - http://forum.universal-devices.com/viewtopic.php?p=45800#45800