MrBill

I confess, you have more details on how GPS works than I do. I have noticed that iOS these days asks repeatedly for awhile (maybe 3 times over 3 days) if you really mean "always" then every few months it asks again. Frankly I'm happy with apple on that... a few years ago apps were taking the "always" way too often and apple was allowing it to happen, where the avg consumer doesn't really realize what they are giving away.

It's been awhile since I experienced a similar issue, but I think the : can't have a space on either side of it--however I don't think it caused the ISY to crash, i think the mail send simply errored out. personally I also need a greater than 1 second timeout. @CMcCarron are you using the "test" button to crash the ISY or is an actual program in use?

I use a comcast account dedicated to my ISY: My redaction ink gets in the way of "From", but it's Home:<userID>@comcast.net where <UserID> is the same thing as entered in the box above and is the portion of the email address before the @ Outgoing messages are not saved in SENT. I actually use the account with a separate email app on my phone as the TO address as well. (I use a different email app so that I can have Notifications turned on. (whereas real email uses a different app and has notifications mostly off) I've found that sending email from and to the same account at the same provider to be very beneficial in terms of speed and reliability.

@randyf the most important thing to understand before upgrading is which z-wave board is installed in your ISY. The easiest method is open the admin console choose Z-wave > Z-wave version. if the answer is 6.81 then all of the 5.x versions are available to you, you should upgrade directly to v5.3.3 If the answer is 4.55.00 then you can not go past ISY version v5.0.16C. If you fall in the the 300 Series version 4.55.00 group then the z-wave board is up-gradable and UDI has special pricing based on the age of your ISY. See here: https://wiki.universal-devices.com/index.php?title=Z-Wave:_Ordering/Assembly_Instructions

It's hard for everyone to grasp how ISY programs work. It's important to remember that when IF is evaluated either Then or Else will always run. I suspect you maybe on to something there. Good luck, let us know if more questions come up as you experiment.

The program being disabled prevents it from running by itself at Sunset and Sunrise. However when ANOTHER PROGRAM tells it to run, it still runs. In this case THEN will run from Sunset to Sunrise.... during the other period, ELSE runs... but it is only triggered to run when the other program tells it to run. The IF statement: $sPBHome is 1 And $sNeighborhood is 1 And From Sunrise + 1 second To Sunset - 1 second (same day) will be evaluated anytime the value of $sPBHome changes AND at anytime the value of $sNeighborhood changes AND at SUNRISE AND at Sunset.... at any of those events/times Let say we arrive in the neighborhood after sunset. $sPBHome is 1 And $sNeighborhood is 1 And From Sunrise + 1 second To Sunset - 1 second (same day) will be seen as False and True and False when we consolidate False AND True AND False to a single value.... we get False.... because we get false ELSE runs so now when PBHome goes to one we get: True AND True AND False which reduces to False so Else runs again Now lets change the time we are driving in to NOON.... first neighboorhood becomes 1....... False and True and True.... Else runs then PBHome becomes 1..... True and True and True... Then Runs.... I never even pointed out in my initial post what happans when the timers time out..... When neighborhood goes to 3 (assume phome is still 1) True and False and True... Else runs again..... The important distinction between State and Integer variables is that CHANGING THE VALUE of a State Variable causes the program to RUN.... If a state Variable changes the IF statement will be re-evaluated AND either THEN or ELSE is going to be executed. Integer Variables on the other hand don't have that magic property of causing the IF of a program to be evaluated so we can think of them as a filter, rather than a trigger. so grabbing the new IF from above: If '#Geofence / Lake House / PBLakeHouse' Occupied is True AND $iNeighborhood = 1 Nothing happens when $iNeighborhood becomes 1..... this program does not run. when Locative tells the portal nodeserver to become true the program runs... now we use the value. True AND True reduces to True..... so THEN runs.

If Send Notification to 'Default' is timing out there should be an error in the error log. But if that's the case you wouldn't change the delay here, you would want to change the SMTP timeout.... Configuration tab -> Emails/Notifications, try 2500 or 3000. Changing 3 to 4 seconds in this program won't do anything that helps.

First since the neighborhood geofence is a filter to prevent things from happening as opposed to a trigger (something that makes things happen) it needs to be an Integer variable rather than a State variable. State Variable cause programs to run. Integer variables don't trigger them to run, but can be used as a filter. So move $sNeighborhood to $iNeighborhood. Also $sPBHome isn't needed at all. Neighborhood - [ID 0057][Parent 002B] If '#Geofence / Neighborhood / PBNeighborhood' Occupied is True Then $iNeighborhood = 1 Wait 10 minutes $iNeighborhood = 3 Else (nothing) In the above program I changed the variable from state to integer, deleted the Else clause, and removed the [Run at Startup] attribute which won't have any effect on how the program works. ------------- PBHome - [ID 0056][Parent 002B] If '#Geofence / Lake House / PBLakeHouse' Occupied is True AND $iNeighborhood = 1 Then Run Program 'Arriving Home' (If) Else (Nothing) ----------------- Arriving Home - [ID 002A][Parent 002B][Run At Startup] [Disabled] If From Sunset To Sunrise (next day) Then Send Notification to 'Lake House' content 'PB Night Geofence Ran' Set '#House / Arriving Home' On Wait 2 minutes Set 'Security / Security' Off Wait 10 minutes Set 'Back door / Backdoor' Off Set 'Hallway / Hallway' Off Else Send Notification to 'Lake House' content 'PB Day Geofence Ran' Set 'Garage / Garage 2 / Garage 2' Open Wait 3 minutes Set 'Garage / Garage 2 / Garage 2' Close Geofences don't work because of Cell Towers or Wifi. It's all based on the GPS coordinates of your phone. A change of cell tower or wifi on/off may cause Locative: to check the GPS values. I've noticed my wife's phone take much longer to realize we have left (if we leave in the same car) than mine. I assume that's because hers in buried deeper in obstacles (purse and purse junk) that make it harder to get the low power signals from GPS satellites. Now if you want that feature, it's doable, i just wouldn't rely on GPS/Geofence. Instead maybe a driveway sensor that detects you driving in, that triggers a transmitter in the house to "ask" the car if it's the correct car to open the door for.... when the car has answered with a secure message "yep this is PBsCar" then open the door.... I still would keep the manual button in the car tho because Automation Can always fail.

@jblackst5000 I'd even suggest going all the way to 5.3.3

Your programs could be written in simpler form. Because ISY is event based there are some things happening that you may not realize. For example, $sPBHome and $sNeighborhood are both state variables meaning that anytime the value changes they fire the programs. What does that mean lets trace what happens as written. You enter the neighborhood and $sNeighborhood becomes 1 --> that fact alone causes Arriving Home at Night PB AND Arriving Home Daylight PB to run Then because Arriving Home at Night PB ran and the If was false the ELSE of that program ran, Which again caused Arriving Home Daylight PB to run. now you get close to home and $sPBHome becomes 1 --> that fact causes Arriving Home at Night PB AND Arriving Home Daylight PB to run again. depending on the time of day we might run Arriving Home Daylight PB yet again. Also the way the programs are written each attempts to run at Sunrise AND Sunset which will RUN the nighttime ELSE clause, causing the Daytime programs to run. which will then RUN again at Sunrise+1 second and Sunset+1 second. I guess I see the point of the dual geofences if for some reason the inner one trips in an out, but I would look for the root cause of that problem, because it's certainly not one I've seen myself or heard others complaining about. I wonder why your instance of locative acts up? Also while it's very James Bond like to have your door open automatically upon arrival I would never have my geoFence actually do that step. For security I'd prefer to press the button myself to open and close the door. I do in fact use geofencing to turn on/off AWAY mode but I also get notification of it were to misfire for some reason (it's never happened). Another reason I would never choose to have the door automatically open is what if someone had cased the place and was standing waiting for that to happen. I'd rather the door never open until I'm in front of it and ready to watch what happens. I'll think about your programs for a bit and post my suggestions. I won't include the line opens the door, If you want to put that back in that's on you.

With locative you should be using Occupancy. The other one Occupancy V2.0 is when you are using UDmobile. We've had very good luck with it. I had the ISY firing lots of Pushover notices in early days, but I've silenced all except the two that say "Away Mode is On" or "Away mode is off". Away mode sets further away from the house than I'd prefer but I think its because my wife's phone is in her purse and it takes longer for GPS. Returning home Away mode turns off right as we reach the drive. Response might vary with phone hardware, we do use iPhones. With the IFTTT portal integration you need to make two entries per phone, and each phone needs to have one state variable. the first entry is "arrive home", the variable associated with the person, and set the Value that you want to use "at home", the second entry is "leave home", the variable associated with the person, and set the Value that you want to use "away from home". Repeat those two entries with another variable for another person. and so on...... For ISY Programs use AND to determine everyone is AWAY and use OR to determine if someone has arrived home. If sPB11 = 2 AND sPB12 = 2 AND sPB13= 2 (etc) then Whatever we do when everyone has left else (none) ------------- If sPB11 = 1 OR sPB12 = 1 OR sPB13= 1 (etc) then Whatever we do when at least one person has returned home. else (none)

upgrade to 5.3.3 before you start! It's stable.

Show use the specific webhooks you created, redacting authentication strings. I haven't done it this way for a long time. Also are you using the portal IFTTT integration or are you using direct API calls? What did you find wrong with locative? it seems to work like a champ for me. Were you using it with the portal node server? I did the reverse of the path you're following, I was using IFTTT and webhooks and it never worked correctly and then discovered locative and the portal node server... works much better, and is more secure.

right on schedule! Worked perfectly! many thanks to @kclenden!!

if you have Polisy/polyglot check the nodeserver store. here's the link from the store: https://github.com/jimboca/udi-poly-hue-emu

if there are 10 zones, it sounds like you would need 10 programs, however it seems that those 10 programs would work to brighten and dim, and not turn on or off. Something like this should work: If Motion zone1 then Set Zone1_Scene to 100% $sTimer_Zone1 = 10 repeat while $sTimer_Zone1 > 0 wait 1 min $sTimer_Zone1 -= 1 Set Zone1_Scene to 10% else (nothing) The above assumes the lights are already turned on to 10%, if there's motion it turns them up to 100%, starts a timer counting down, if there is more motion the timer will be interrupted and start over again, only turning the lights down again if the timer actually runs out. Edit to add: after thinking about this a few minutes there's an even simpler version of the same program: If Motion zone1 then Set Zone1_Scene to 100% wait 10 min Set Zone1_Scene to 10% else (nothing) No variable actually required. The wait will restart with each new motion detection.

I found that email in my Junk folder Not sure why, I've received have plenty of UDI mail in my inbox. I've now added sales@universal-devices.com to the trusted senders list.

With the first experiment I want to follow the directions in the wiki and find the failure points. The point being there should be instructions that can be followed without guessing what to do next, or guessing what additional parameters need to be set or adjusted after restoring a backup. After that I can experiment further with ideas such as yours. I do actually have two new microSD cards. I suspect the reason that you did NOT have the problem with missing parameters is exactly because you placed an image on the card first.

I noted that in my Feb 6th post in this thread. I really also am still wondering what's included vs not included in the restored backup. If all else fails I'll find a block of time to set up a test... I have another new memory card, and a second ISY. This time I can be more methodical and record more observations since I know what to watch for.

I've had this problem... I bet the 0 is a D I've also had one older device that just flat out had an incorrect address label on it. It was a device with a separate address and Rev label.

Thank you Michel, Are the NLS files then installed during format? Because format was what cured that problem. Thanks you for the additinal information about logging, On smaller memory cards it does seem that would add accelerated failure, especially since nodeservers cause much more information to be logged. I don't have a small card anymore tho, so how would I determine the best value to change this to? Is timezone and mail server information included in the ISY backup that is restored? I suspect I didn't have this problem with the first memory card I robbed from the 2nd ISY because those values where already set on that system. When I repeated the process 24 hours later with a brand new card, those values where not set when the backup was restored. If that's the case are there other things in the configuration we should set or check after a memory card swap?

@Teken I agree, even the failing memory card showed no bad blocks, that must be a feature not yet implemented. I have no problem with scheduling memory card replacement. My specific complaint was and is that there is no prescribed recommendations published. 1) how to determine the frequency for replacement (I suspect this varies by system size and now number of nodeservers and how much data they write to the LOG) 2) a list of recommended memory cards to purchase. UDI doesn't sell memory cards in the store and not all memory cards are created equal. A simple list of suggestions should be published with the proceedure. 3) Lack of procedure. I admire Michel greatly, yet in another thread he is currently talking about what is "not a sustainable business model" but guess what... having him help every single user that's headed down this path also isn't a sustainable business model. He shouldn't have to get involved because I need to systematically replace my memory card, or worse I need his immediate attention because Poof it crashed and I need help putting it back online. The written instructions don't go far enough, certainly I had the basic idea... install new card, upload firmware, restore backup. The wiki is also wrong... it says the max mem size is 16gb, and we know from your experience that it's at least 32gb. It's harder to find smaller memory cards these days, bigger is actually less expensive than older smaller sized cards. Also on the topic of lack of procedure, I believe the following is missing a) check and reset times zone b) check and reset mail server information c) check and reset ?????? The only reason I had a better experience the first day was because the memory card I installed came out of another working ISY, there for time and mail sever didn't need to be adjusted. Adding more space to the file doesn't add more writes to the card. What added more writes to the card was node servers! It used to be the main LOG would hold months worth of Insteon switch traffic and state variable changes. Then came nodeservers! now my 64,000 (or whatever number) rows of logfile are used up and re-written in less than 24 hours. I don't know how sectors are allocated and/or reallocated on that card but if the files stays in the same spot on the card then increasing its size would actually decrease the number of time specific bytes that are being constantly re-written. If that card is as smart as they say it is tho, then it may no matter because the whole card should be used as part of the wear leveling algorithm. AND if it is that smart that wear leveling is using up all parts of the card more or less equally then we should buy the biggest card the hardware can handle. We however can't make any of these decisions because there isn't enough information available (without asking Michel to hold our individual hands, which is not a sustainable business model). to add to your list: Log file out of order date-wise, or possibly it was in the write order with wrong dates (no way to tell which) A program clearing displaying in the admin console IF condition_1 AND condition_2 was clearly running as IF condition_1 OR condition_2. in the finally hour the program tree vanished entirely from the admin console however program were clearly still running [LIB_403] errors when trying to start the admin console. An error not mentioned in the UDI wiki, reading threads in the forum it seems to boil down to replacing or reformatting the SD card. To sum it up, for the price of an SD card I really don't mind having to schedule annual preventive replacement, the problem is lack of guidance and procedure unless we request Michel to come hold our individual hand, a business practice that is not sustainable.

I should perhaps add that I really don't have problems with manual procedures during these once in a few years events, but I find it difficult that the wiki page doesn't better document the steps to follow, including the needs to reset certain parameters and there isn't a better recommendation for what memory card to buy. I also wonder and will ask support in the days to come about increasing the log size since I see that appears to be a telnet selectable parameter and i have a much bigger memory card. (and it's usage doesn't seem to be discussed anywhere).

@hart2hart Today my comments are less rosey. Yesterday the new 16GB SanDisk Ultra cards arrived (I didn't even know about the Industrial line linked by @Teken and I thought I had done my homework, as this has been a topic RE: a Pi that I have running Home Assistant). After getting the ISY back up rather quickly the day before I realized that LOGS are an item being written to the card. This absolutely hammers the card in my case--there was a time that Tools > Log covered literally months, however since Nodeservers the log recycles quickly... 18-24 hours depending. So that file is just constantly being re-written. Because of this I decided to install the new SanDisk Ultra card in the main ISY. Based on my experience the day before, I thought it would be easy.... however the new experience is not as straight forward. Leading me to wonder about your idea of copying the card. Since @Teken states that you can unzip a backup on to a card, but then he later restored it from a backup also I'm wondering if there is a better way. Anyway comments from my experience yesterday: The wiki states that if the microSD card is new, that it's possible to skip the formatting step. I found this to be false. I made a brand new fresh backup and shut the ISY down, I popped the new 16GB card into the ISY and powered back up. Opening the Admin console produced LIB-403 error's, I couldn't get to the point of being able to reload the firmware and restore a new backup. I followed the standard instructions to clear the java cache and re-install The Launcher. No change. Searching and reading around the forum, I was about to decide the new card was bad. I had another new one to try but before I did that I decided to give formatting a try (the step the wiki says to skip if the card is new). Telenet into the ISY and issued the FS command at which point you have to answer the "are you sure?" prompt with an upper case Y as documented in the wiki. The wiki then states "You will see the "Goodbye" prompt confirming that the format is complete and that the has ISY rebooted". That never happens, upon typing the upper Case Y the terminal session appears to lock up, after a period of several minutes the window simply goes to the (inactive) state, your cursor is still blinking next to the capital Y that was typed. No goodbye and No indication the ISY is rebooting.... The ISY did reboot and I tried to fire up the admin console. This time success. First place I went was Tools > Diagnostics > System Status Interesting! the ISY is recognizing the card as 16gb. Seems like I've read plenty of places in the forum that the ISY only used 4gb of a memory card. I proceeded to restore the backup.... all appeared to go well at that point. and before too long we were back in business. BUT Not so fast...... The first thing I noticed was that a scheduled event at 1PM didn't occur.... Check time... Hmmm, the time zone parameters didn't restore. Looked up the custom Longitude and Latitude values I use and set that up again. all good? Nope... next thing I notice is a missing notification..... investigate... OH there's no email server settings.... reset those.... ("Use default" was checked, and I thought that the default mail server was fixed but apparently not.... "use default" didn't work either.) Put the settings back to the mail server I normally use, test, back in business... Until..... something else didn't happen. I quickly discover a program that should always "running then" isn't running. (i.e has turned to , when it should always be Solid Green). Scope out my heartbeat programs and they aren't running either... After some sleuthing I realize that this is likely because I reset the time (two paragraphs ago). That's all I've found and so far we appear to be back in business. I'll be certain to report back if there are other items that didn't properly restore. On another note: I have popped the cards that came out of the ISY's into a card reader and opened them via a Win10 file explorer. There is readable directory structure, making it appear that it might be possible to format a new card (I'd use the ISY to do that, that way the ISY can lay down whatever boot record it wants). and then move the card to a computer and copy the file structure..... I don't know tho, I'm just guessing.... I don't know where the things that were not restored via backup are stored.

<<-- Red means the IF last evaluated to False and the Else section last ran. Green means the IF last evaluated to True and the Then section ran last. and mean the same as above for color, but when the color is solid it means the program is currently running (most likely because there is a wait that is waiting).