I just updated all of my ESPHome devices from the HA ESPHome GUI as I always do. After the update, I am still able to control all of them in HA and see their status, but a few of the smart outlets show “offline” in ESPHome, even after I unplugged and plugged them back in. When I try to ping their normal IP address, they seem unpingable at that IP, but they are controllable. An IP scan also doesn’t show these devices on the network, but they have to be to be controllable from HA. Any ideas what is going on here? I’ve never seen anything like this…
Just a few random ideas, based on the possibility that your mDNS and DHCP/DNS environments may not be synchronized:
e.g.,
HA may be finding them by mDNS (Avahi, zero-config) and you are looking for them using conventional DNS, and the DNS/DHCP data in your router may be stale (e.g. if they got different IP addrs after restarting).
Take a look at your DHCP setup to see if it’s assigning them the IP’s you expect (if you care) and/or if they are showing up as having leases at all.
Try ‘nslookup node’ and ‘nslookup node.local’ (where ‘node’ is the node-name of your ESP) and see if the addresses coincide.
This is just a set of random items that relate to the possibly-unsynchronized world of having both DHCP/DNS (e.g. dnsmasq of various ilks on your router) and mDNS which is peer-to-peer and runs without mind to what your router’s doing with DNS/DHCP.
Or, it’s something else entirely 
Thanks for the suggestions! Luckily, it was an easy fix. I went back to the basics… reboot everything. Well, not everything, but I rebooted the VM hosting hassio and my Unifi controller. Everything is working as usual now. Whew!
1 Like
I’m having issues too lately, … the switches are online can connect to them directly and control them, sometimes it seems like the esphome firmware has done something which makes the clients run slow or something not sure. Something is wrong.