Go to Smart presence app settings and Create diagnostic report after false Away detection, describing your issue, eg. which device was detected wrongly as away.
I made one last night after posting the message. I have two devices, and both have had a change in behavior. Thanks for the quick response! Let me know if you want me to make another or post you the ID of another report. Love this app, and hope the report is of help
Well it’s hard to say what is specific for you, because so far it’s installed by several thousands of users and I don’t see yet any wave of false triggers reports but maybe it will come 
However Test version has been in place for several weeks without any issues for ~100 users. Can you maybe as first step, use recommended settings if it makes any difference for you and submit another log WHEN false trigger happens again - also use your nickname, so I can link it with you ? Based on the log I just see that device M** is misbehaving somehow but detection on SmartPresence side works as it should unless that specific settings you are using triggers some problems on detection side.
'Away Delay': '90 seconds', ---> you are using 220 and 300sec, shall be okay but to rule it out
'Normal Mode Check Interval': '15000 ms', --> you have 5000ms, which might be too frequent
'Normal Mode Timeout': '2 seconds', --> you are using 2&3sec, shall be fine but same as above
'Stress Period': '60 seconds',
'Stress Mode Check Interval': '2500 ms', --> you are using 1500ms, might be too low
'Stress Host Timeout': '2 seconds'
Thanks for the follow-up, and sorry for not following up sooner. I submitted another report today, with my nick. I updated the values to your suggestions, but still I’m having some false triggers. To me it really felt like the update started honoring a setting or treating something differently, given how it changed literally with the update. If there’s nothing else to indicate that to be the case, I’m sure it’s somehow a coincidence.
Just figured I’d check as I tweak values further. I don’t mind checking frequently, if that in turn enables me to have a shorter away delay. Not knowing the implementation, I still figure it doesn’t generate more than a few packets of traffic. But since I don’t know the implementation, I figured I could try understanding better how to reach my goal.
If I wanted to try to find my phone for 50% of the time, what’s the conceptual difference between having a check interval of 2000 ms and timeout of 1000ms, compared to 20.000ms and 10.000ms? I’m guessing timeout is the time it will keep trying to send packets with no reply before flagging it as no response, but if the phone is online, the roundtrip time is really low, so it shouldn’t have a chance of not being able to reply, which makes it feel like the two settings should be functionally identical?
My other idea is that launching a check might have some scheduling or otherwise delay with Homey, and that that may warrant longer intervals?
There hasn’t been a real change in detection mechanism, just fixes for some type of devices and first triggers.
Which kind of devices do you have ? Maybe then someone here for several thousands of users can share their experience. I assume you have also set correctly Wifi, ruled out Wifi sleep and while eg. checking availability of your device, eg. when running PING in parallel to the device and to your Homey in parallel, you get no timeout ? Can you try that and share with me the screenshot over mail that I responded, so I know it’s worth of deeper investigation ? Given the frequency of mis-detections it shouldn’t be time consuming.
- Interval: How frequently the system performs a check (e.g., scanning or pinging a device).
- Timeout: How long the system waits for a response before declaring the device “offline” or “unreachable.”
I really suggest to stick to default values, setting timeout to 10sec really doesn’t make sense, you would need to have in such case huge connectivity problems on your devices, which I don’t believe it’s your case.
I never had such issues myself but it could be workaround, eg. running a delay from Away fired and fire action only in case “No household member at home” is True (as example) after that delay.
But first I would like to understand the root cause of the mis-detection honestly.
We both have iPhone 15 Plus
That’s what I figured, and it’s good to know. I’ll run a continuous ping of both devices to see if I can spot any patterns that might prove helpful to know about.
Thanks for your engagement! Let me know if I can do anything else to help! I’ll submit new diagnostics if I see any new information
This works well for my android devices. But which port should i use for an iphone? I cannot seem to get it to work for any of my apples. They do have a fixed IP address.
Most users stays with the default settings and it works for them. Unfortunately can’t comment more as I’m an Android user. Maybe someone with Apples can comment.
Hello everyone,
I would like to know what the difference is between the flow cards
A specific user has arrived or
A guest has arrived.
Is the difference, since I have to query a logic (see screenshot), that the name already exists for the specific user and it assigns it?
I don’t understand the differences here!
Do I even have to have a logic query for the top scenario?
The aim is to integrate this into the shutter control, if a certain person is a guest who smokes, then certain shutters should stay up or move up as an exit 
Regards
Pierre
First is related to any household member including kids except of guest, and will be fired when any household member arrive
2nd one is just for guests and will be fired only when guest device arrive
