Feature Request - Add a Sleep condition/setting

[mmstano]

Is there an existing issue for this?

• I have searched the existing open and closed issues

Is your feature request related to a problem? Please describe

This feature request is for fine tuning devices that go into deep sleep. Currently, netalertx behavior allows setting a down condition with a time setting before the device is reported down in a notification. While this works good, it's not visually pleasing to see devices appear as down before the next scan or two happen to report the device as online. This is especially prevalent for battery powered devices, even more especially when the battery starts getting lower in charge. Some devices can deep sleep upwards to 20 minutes.

Describe the solution you'd like

I would like to propose another condition that will work similar to "down", but called "sleeping". It should have it's own time setting, but can work hand-in-hand with the down condition/setting. When the sleep condition is checked (selected) for a device, instead of appearing as "offline" or "down", it will appear as "sleeping". Should the device retain this sleeping condition past it's time setting, only then should it be reported as "offline" or "down" if the down condition is selected. If the down condition is selected in conjunction with the sleep condition, once the sleep time has past, the device becomes subject to the down time setting before it's reported as down in a notification.

Describe alternatives you've considered

I'm using OpenWrt with LUCIRCP and ARP for reporting. I've tried unselecting the "only reachable" option for LUCIRCP, but unfortunately, OpenWrt's neighbor tables can be janky and have devices appear as online on the network, even though the device is no longer connected. The only way to have the device be correctly reported in these situations is to restart the network interface in OpenWrt or, in some cases, restart the router. Having "only reachable" selected appears to be the better option, as devices are reported on the up-and-up, but deep sleep devices can be annoying. I've tried setting cron jobs in OpenWrt to refresh the neighbor tables, but I haven't been able to get it to work.

Anything else?

I use NetAlertX App in Home Assistant and I have minimal experience with Docker. I'm willing to run a docker container on my PC to test the feature.

Am I willing to test this? 🧪

• I will do my best to test this feature on the netlertx-dev image when requested within 48h and report bugs to help deliver a great user experience for everyone and not to break existing installations.

Can I help implement this? 👩‍💻👨‍💻

• Yes
• No

[jokob-sk]

Thanks again for the detailed write-up.

I’m trying to better understand where the current configuration falls short. Since NetAlertX already allows tuning of timing, could you clarify a couple of points?

Would increasing the scan interval help in your setup?
(e.g. scanning every 5–10 minutes instead of very frequently, so devices don’t briefly appear as down between scans)

Or alternatively, have you tried extending the global down delay? (Setting NTFPRCS_alert_down_time)
Since this is already configurable, it can be set to cover longer sleep cycles (e.g. 20 minutes)

From my perspective, both options already provide a way to handle devices that go offline temporarily without introducing additional states or logic.

The main trade-off I see is:

• Increasing the global delay helps deep-sleep devices
• But may delay detection of actual outages for always-on devices

Is that the core limitation you're running into — needing different timing behavior per device rather than a single global setting?

If so, it sounds like the requirement is less about a new “sleeping” state and more about per-device timing overrides, which might be a simpler and more flexible solution.

Curious to hear your thoughts 👍

[mmstano]

Thank you @jokob-sk for responding so quickly. You are right in your assertions regarding the main trade-offs. I contemplated increasing the scan interval and/or the global delay, but the idea seems silly to punish the majority of devices that are working well due to the behavior of a few rouge deep sleep devices. I'd much rather deal with the annoyance of deep sleep devices than sacrifice time with a delayed response when a device legitimately loses connection from the network.

Your idea of having a per-device time overide sounds better, too, and would most certainly curb the annoyance. It would definitely make NetAlertX more flexible because each device can be tailored to their particular behavioral sleep patterns, as opposed to an additional logic providing a blanket setting.