MrJeffreyGee wrote:

bschollnick2 wrote:The only time the username / password is placed into the log, is if Debugging is turned on for Find My iDevice. And it's configured to do that, so that the owner can see if there is an typo....

Debugging is not enabled in my plugin and I'm seeing my password in the event log. The password is showing up in plain text besides the AltMobileMePassword field name. I'm using version 1.52 that has the version typo 1.51.

I just searched the plugin, and AltMobileMePassword is not contained anywhere in the plugin file. Before the rewrite, it certainly was, but not now.

I just went through the indigo.server.log lines, and there is none that refer to the password. The only logging done on the password is with the debug flag set...

I just checked the IWS plugin, and even that doesn't do that...

My first suggestion, is to completely remove the FindMyIDevice.IndigoPlugin, and replace it with a fresh download. If it continues, then please send me a copy of the log file, with your password obscured. I need context to see how this could be happening, when it's *NOT* in the release copy of the plugin. Send it to Benjamin AT schollnick DOT net.

- Benjamin

Good morning -

I'm loving this plugin, but having a problem in execution, as indicated in the log below:

Jun 30, 2012 9:18:10 AM
Reloading plugin "Find My IDevices 1.51"
Stopping plugin "Find My IDevices 1.51" (pid 16282)
Find My IDevices Error exception in deviceStopComm(Adam's Apple ID): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStopComm(Adam's IPhone): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Plugin "Find My IDevices" disconnected
Starting plugin "Find My IDevices 1.51" (pid 38318)
Plugin "Find My IDevices" connected
Plugin "Find My IDevices 1.51" started
Find My IDevices Find My iDevice Plugin Loaded...
Find My IDevices Error exception in deviceStartComm(Adam's Apple ID): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStartComm(Adam's IPhone): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStartComm(Allie's Apple ID): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStartComm(Allie's iPhone): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Retrieving location data
Find My IDevices Locating - Adam's IPhone (Adam Husiks iPhone)
Find My IDevices Done locating - Adam's IPhone
Find My IDevices Locating - Allie's iPhone (Allison Days iPhone)


Any ideas?

Thank you!

Adam

aghnj wrote:Good morning -

I'm loving this plugin, but having a problem in execution, as indicated in the log below:

Jun 30, 2012 9:18:10 AM
Reloading plugin "Find My IDevices 1.51"
Stopping plugin "Find My IDevices 1.51" (pid 16282)
Find My IDevices Error exception in deviceStopComm(Adam's Apple ID): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStopComm(Adam's IPhone): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Plugin "Find My IDevices" disconnected
Starting plugin "Find My IDevices 1.51" (pid 38318)
Plugin "Find My IDevices" connected
Plugin "Find My IDevices 1.51" started
Find My IDevices Find My iDevice Plugin Loaded...
Find My IDevices Error exception in deviceStartComm(Adam's Apple ID): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStartComm(Adam's IPhone): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStartComm(Allie's Apple ID): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Error exception in deviceStartComm(Allie's iPhone): 'DeviceStateChangeTrigger' object has no attribute 'pluginTypeId'
Find My IDevices Retrieving location data
Find My IDevices Locating - Adam's IPhone (Adam Husiks iPhone)
Find My IDevices Done locating - Adam's IPhone
Find My IDevices Locating - Allie's iPhone (Allison Days iPhone)


Any ideas?

Turn on Debug logging, and stop and restart the Find My iDevices plugin again. Cut and paste the log and if you could send it to benjamin at schollnick dot net, I'll take a look and see if I can figure out what is causing the issue...

So I have only been using FMID for a couple days, but I already have some questions, comments and concerns.

First off, this is awesome. My wife doesn't initially like the idea of being "tracked," but I think once she sees the power of location awareness build into an automation system, she might think twice; that's what I'm counting on anyway.

Some background: I am fairly new to automation. I have been playing with it for a few weeks now, but I am pretty comfortable with Indigo, and I have some fairly complex custom logic triggers built on top of the Smartphone Radar plugin to allow more states of presence than comes built in (ok, maybe it's not that complex, but there are 4 triggers per smartphone that set custom variables accordingly based on its presence). I have a few triggers and schedules built on top of this logic, and it basically turns lights and fans on/off based on our location (again, home/away) and the outdoor states (daylight, weather, etc.). My interest in FMID is to further dial in my automation to allow more power savings with the same level of customization. For instance, my porch light comes on automatically when it's dark outside and someone (anyone) is away, while the light above my kitchen sink comes on when it's dark outside no one is home at all. This pretty much keeps us from having to remember to leave a light on. The fans in the house do similar stuff based on our presence and the weather, helping us save power on cooling in the summer. (I'm still waiting on Nest's API to further this customization, as I bought the Nest thermostat before I got into automation.)

I am currently using FMID to send myself emails when a trigger is pulled, so I know it's abilities. So far, considering the 15-minute delay I have configured, it works great. But as I mentioned in another thread, that's one of the problems. My neighborhood is rather small, and I can get through it on foot in 15 minutes; in my car it takes two. The ability to force checkins more often based on a schedule is nice, but still not as dynamic as I'd like to see. That's why a real geofence feature in Indigo Touch would be so beneficial.

So, onto the comments and such...

• I have noticed some interesting behavior surrounding the web interface. It appears as if there is a 15 minute delay in my position reported by Apple, completely separate from the 15 minute delay of FMID. That is, say I am at home at 8:00AM. By 8:15AM I am dropping my son off at daycare. At the 8:15 check-in FMID will report that I am still at home. By 8:30AM I am at work, but the 8:30AM check-in reports me to be at my son's daycare. Then around 8:45AM my position shows me to be at work. This is interesting behavior, to say the least, when you consider the following...
• My triggers get pulled at the proper times, and based on my ACTUAL location at the time of the trigger. So although the FMID web page shows my position 15 minutes ago, the emails I have set up based on my distance from home get sent at the proper time based on my actual position. So if I am within .5 miles of my house, and I have a trigger to send me an email when I am within that distance, I get the email even though the web page shows me at the location I was at 15 minutes ago. So it would seem my actual position is not being reflected in the web page, but the triggers that are fired at the time of check-in are fired based on my actual position at that time. (Confused yet?) It's weird.
• FMID chews through my battery like crazy. Not much that can be done for this, I imagine, aside from extending the checking time to 30 mins, an hour, etc. but it's an observation nonetheless. Of course, this would limit the tool's accuracy and keep events from firing at the proper times.

This is about all I have so far that's of any real concern. I still have some more testing to do, and I am excited to see the ability to add more "Centers" you mentioned in the Indigo Touch thread I started. That might be better than how I have it configured now since we can set FMID to trigger actions based on us leaving work or our parents' house, versus arriving at home - that might be slightly more reliable anyway, although I am still concerned about the battery drain on the devices being tracked. Smartphone Radar just uses SNMP to monitor your network traffic on a router, which causes minimal battery drain. (The dirtiest part of this setup is that I had to add a POP email account to both phones and force it to pull mail every 15 minutes for absolute reliability and minimum delays in its decision making. Not 100% ideal, but better than nothing I guess.)

Thanks a lot for all you've done thus far - I can tell you've worked really hard on FMID to make it the best it can be. I really do appreciate your effort.

EDIT: I'd like to point out that my custom presence states are somewhat duplicates of the existing states (or a combination of existing states, rather) that come built into Smartphone Radar. The reason I created new ones, however, is because I cannot set up a Trigger to act upon a combination of variable changes. I can only use one variable, as far as I can tell, to trigger a response. So, my SomeoneHome and SomeoneAway variables are a combination of Smartphone Radar's AllPhonesHome and AllPhonesAway variables, with slightly different conditions. This way I can use the same state to trigger an action instead of, say, depending on a 15-minute Schedule to accomplish the same thing. I guess it's a matter of whether you want your action based on a time increment, or a change in state... Whatever the case, it works for me. Just FYI, in case you were wondering.

richy240 wrote:

• I have noticed some interesting behavior surrounding the web interface. It appears as if there is a 15 minute delay in my position reported by Apple, completely separate from the 15 minute delay of FMID. That is, say I am at home at 8:00AM. By 8:15AM I am dropping my son off at daycare. At the 8:15 check-in FMID will report that I am still at home. By 8:30AM I am at work, but the 8:30AM check-in reports me to be at my son's daycare. Then around 8:45AM my position shows me to be at work. This is interesting behavior, to say the least, when you consider the following...
  

I haven't been able to find any cause for this, nor duplicate it....

I just tested with my current work build, and it's working as expected. Are there any errors in the Indigo log?

When this occurs, does the "This was last update on *date*" match the Last updated for the devices?

richy240 wrote:

My triggers get pulled at the proper times, and based on my ACTUAL location at the time of the trigger. So although the FMID web page shows my position 15 minutes ago, the emails I have set up based on my distance from home get sent at the proper time based on my actual position. So if I am within .5 miles of my house, and I have a trigger to send me an email when I am within that distance, I get the email even though the web page shows me at the location I was at 15 minutes ago. So it would seem my actual position is not being reflected in the web page, but the triggers that are fired at the time of check-in are fired based on my actual position at that time. (Confused yet?) It's weird.

Quite confused. I've been looking at this from a few different angles, and I just don't see how this disconnect can occur.... The triggers are evaluated at the same time the web data is written. In fact, both are processed at the same time as the location fetch occurs...

Actually, I just had a thought... Are you manually triggering any location updates? For example, a timed schedule within a minute or two of each other?

If so, can you change it to a scheduled task of every 5 minutes? I'm wondering if that might have an affect?

richy240 wrote:

FMID chews through my battery like crazy. Not much that can be done for this, I imagine, aside from extending the checking time to 30 mins, an hour, etc. but it's an observation nonetheless. Of course, this would limit the tool's accuracy and keep events from firing at the proper times.

That's an issue with GPS in general.... But what I have seen is that 15 minute intervals usually has very little effect on the battery life. Especially if you are running iOS v5.11.

I'm concerned that I have been unable to duplicate the issue you are mentioning... Regarding the 15 minute delay....

But I am working on v1.54, so if we can isolate this, I should be able to add that to v1.54...

- Ben

Should the fenced areas be working in the latest version?
Not working for me. The closest address seems to work if it's exact but in some
cases I see it changing while still in the same location.
Not sure of the best way to handle that.

Thanks,

Carl

I have found that using the "between latitude" and "between longitude" as a condition seems to work well.

Carl

ckeyes888 wrote:Should the fenced areas be working in the latest version?
Not working for me. The closest address seems to work if it's exact but in some
cases I see it changing while still in the same location.
Not sure of the best way to handle that.

Fenced Areas are depreciated and have been removed in the latest version.

Closest address, or setting a trigger by latitude and longitude will work...

In the next version, I'll be adding "Map points", which will be lat/long points, which would can trigger against if you are within a set radius of...
(I.e. if you are within 1 mile of the point)

- Ben

Maps points sounds perfect for me. Look forward to that feature.

Thanks,

Carl

Hi Benjamin,

Thanks for this great plug-in.

Would you mind either
a) changing the distance-from field (when setting up a trigger) to miles/km instead of lat/long

or

b) giving us some quick calculations for
1/4 mi
1/2 mi
1 mi

etc?

it seems to be a somewhat difficult calculation and i imagine this would make the plug-in much more user-friendly

pantalones wrote:a) changing the distance-from field (when setting up a trigger) to miles/km instead of lat/long

Good question, and I just did some checking. The default answer is that the distance is in Kilometers. (I had thought it was in miles... That's an oversight on my part)

The distance is calculated using the Haversine formula, which uses Great-Circle distance calculations...
See http://en.wikipedia.org/wiki/Great-circle_distance

I'll see about making a setting for converting that to Miles....
(Or adding another field for miles...)

- Ben

Hi,

Thanks for the quick reply!

If I understand you correctly, the plugin currently works as I was hoping, in that it does the conversion from KM to lat/long internally? So I can just enter 0.5 if i want 0.5 KM as the threshold?

pantalones wrote:Hi,

Thanks for the quick reply!

If I understand you correctly, the plugin currently works as I was hoping, in that it does the conversion from KM to lat/long internally? So I can just enter 0.5 if i want 0.5 KM as the threshold?

That should be correct. The DistanceAway state is measured in KM currently...

bschollnick2 wrote:I'm concerned that I have been unable to duplicate the issue you are mentioning... Regarding the 15 minute delay....

But I am working on v1.54, so if we can isolate this, I should be able to add that to v1.54...

- Ben

Dude, I know it sounds crazy, but that's what I observed over a two-day (well, 38 hour) period. And regarding battery life, I am often running low by evening, maybe around 15-20%, and the added drain of FMID was enough to make my phone die before I ended up in bed.

I ultimately stuck with Smartphone Radar because it requires much less drain, but I'd still like a more elegant solution. I almost bought Witness for Mac (http://www.orbicule.com/witness/index.php) because it let's you trigger AppleScript based on geofences in iOS, but it can't handle more than one phone reliably for my purposes, so that plan was scrapped (although I will be watching it closely for developments).

richy240 wrote:Dude, I know it sounds crazy, but that's what I observed over a two-day (well, 38 hour) period. And regarding battery life, I am often running low by evening, maybe around 15-20%, and the added drain of FMID was enough to make my phone die before I ended up in bed.

I ultimately stuck with Smartphone Radar because it requires much less drain, but I'd still like a more elegant solution. I almost bought Witness for Mac (http://www.orbicule.com/witness/index.php) because it let's you trigger AppleScript based on geofences in iOS, but it can't handle more than one phone reliably for my purposes, so that plan was scrapped (although I will be watching it closely for developments).

I understand that your seeing it... But I don't understand how this is occuring for you, and not being seen elsewhere....

Force an update, with the log file open, and when it finishes, check the last_updated state for each device. Is it accurate?

On the web page, does the last_updated field for each device match what is in the device's last_updated state?
For that matter, does the web data match what's in the device's states?

In the/Library/application support/perceptive automation/indigo 5/IndigoWebServer/plugins/Find-My-iDevices folder, does the time/date stamp of the "pickled_data" file match the time/date of the last update from Find My iPhone?

(The pickled_data file contains a snapshot of the Find My iPhone data... It's how the IndigoServer Plugin exchanges data with the Indigo Web Service, since they can't currently communicate directly...)

This information should help isolate the issue....

- Benjamin