Random emails - tried HomeSeer forums and failed.

[Gadgets]

Hi there everyone,

So I think I had a "duh" moment the other day, and posted the below message in another forum here at HomeSeer forums instead of this one. I had some help, but nothing has changed. Then I realized this is a INSTEON module I'm dealing with, so maybe I should try here.

I have been running HomeSeer4 on a PI since insteon went down. Everything works great, accept I am getting random email notifications sent to me in the middle of the night about one of my devices. It is not always the same device, and only one email is ever sent in the night.
I have attached a snipet of my log where the action takes place, to me (who's new to this) there seems to be no reason for it. Below is the description of what I have configured:

In the log I have posted, it is my pool heater that is in question this time, I am using the Insteon Hub, with a Insteon appliance module to control the heater. I have it set to turn off at 11pm, and upon device state change it emails me to let me know, sort of as confirmation, that it has turned off. This works just fine. However in the log at 1:56:40am the system decides the module has changed state and sends me an email, even though it is already off, and was turned off at 11pm and sent me the email at that time. It only sends the one rogue email per night, and not every night either, although I did receive one during the day a few days ago. It's not really a huge deal, but I'd like to stop it if anyone has any ideas.

I have no other events that run after 11pm.

HS4 on a PI3
Insteon Hub
Everything but the hub is on a UPS. (ie. all network, routers, and PI)

I'm puzzled.

[mnsandler]

whats the insteon module

[Gadgets]

Sorry Mark, should have included that. It is a ApplianceLinc module 2456S3.

[Burrington]

Try using the trigger "Changes and Becomes" instead of "Had its value set". Had its value set can trigger for various reasons even though the value didn't change. Changes and Becomes only happens if the status actually changes.

[Gadgets]

Sadly, that's what I have it using already.
Last night, while playing around in HomeSeer, I discovered the "communications" page within th Insteon plugin. I've attached the screen capture of it. Notice the only item on the list with a "Good" rating is the pool heater module, all others are excellent. I also notice the protocol used is i1 vs i2 for all the others. I haven't had time today to check, but could that be an older version of the appliance module, and it is reporting random status changes even if they are not happening? I'm thinking maybe I should swap out the module. I have the pool pump module plugged into the same power strip, and it reports excellent communications.

**EDIT** I have replaced the module, and it is now communicating i2 as the protocol, and reporting excellent! Time will tell if this was the problem.

[Burrington]

You are polling every device once a minute which is causing the huge number of responses in your comm results. Polling makes the interface ask a device for its status and the device responds. This is over and above the normal command and response process that takes place when something changes. Polling is only needed if you have a device that regularly fails to get it's message through due to poor comm. In that case, polling on a regular basis, maybe once every 5 or 10 minutes, can help retrieve any missed messages. When your Insteon network gets super busy like yours is, you will have crashes of multiple devices trying to respond at the same time which can cause messages to be confused. Normally the protocol handles it with a retry, but if it's too busy that can get messed up too. This is especially true for the older single-band devices like your 2456S3 which communicate by powerline signals only.

I think if you change all your polling numbers for all devices to zero, your problem will go away. You will likely find your response times are better too with the greatly reduced traffic.

You could also post your event here so we can check your triggers for problems too.

As an aside for anyone who does require polling turned on for more than one device, poll your devices on a staggered schedule rather than all at the same moment. For example device1 every 10 minutes, device2 every 11 minutes, etc. This will reduce the volume of traffic at any given moment and speed up the network response time.

[Gadgets]

Ok then, I have mis-understood the polling option then. I didn't even think of increased traffic (but should have), I thought the polling time would speed up response times, guess I was big time wrong with that thought.

I have changed all the polling times to zero, which if I remember right is the default anyway. Reset the communication stats, and re-tested them all. All work, all reporting "Excellent" still. I also think you are correct on the reponse time, the items I have tested do seem to be responding faster.

I have attached a screen shot of both my event that turns the heater off, and the once that sends the email to me when the state changes. Feel free to let me know if anything is wrong there.
Thank you very much for taking the time to help, I very much appreciate it. Hopefully, these changes will resolve the issue, but either way I will post here to let everyone know. I won't know for a couple days probabaly. I receive a rogue email almost everyday, but sometimes it skips a day.

[Burrington]

Your events look fine to me. Assuming there are no other things that change the device's value (Other events, group links, Alexa, etc.), it must just be the heavy Insteon traffic. I think you'll be good now.

[Gadgets]

I made the changes July 3rd, as you had suggested, today is July 5th, and so far no un-expected emails! I will report back in another couple days just to be safe, and at that time I think we can consider the problem solved.
Again, really appreciate your help on this one.

[Burrington]

I made the changes July 3rd, as you had suggested, today is July 5th, and so far no un-expected emails! I will report back in another couple days just to be safe, and at that time I think we can consider the problem solved.
Again, really appreciate your help on this one.

You're welcome. Glad to hear it is working out so far. We'll keep our fingers crossed.

[Gadgets]

Well it has been several days now, and no random emails, only what should be being sent, so I suppose this topic is solved!!

Everything is working smoothly now. Thank you again for the kind assistance!

[Burrington]

You’re welcome - happy to help. Thanks for reporting back.