Havent had that specific problem in a week now, but i often get messages that the supervisor could not open grafana or node-red… then when i try a second time, or at the third time it finally works…
the superwisor seems to talk to the other services with a 172-something-address… so still, if there is a problem with the network driver it can also impact the entire TCP/IP-Stack.
but currently i have less time for that, so i started to document some stuff for myself, before i plan to do a greenfield-installation
I got the same problem 3 times last week, whitout any action, the Pi is not in my network anymore (even if it’s LAN connected to router and the LED is on). after a Pi reboot everything works fine! until now, i did not succeed how to pass over this issue
I agree that this appears to be on the HASS side. However I do have a fix that works. Simply pull the LAN cable out once a day, wait 10 seconds, then plug it back it.
So, being an automation guy, I automated this :). I put a switch between the rPi running HASS and the hub it is on. I have the switch automatically disconnect the rPi from the network every 8 hours, wait 10 seconds, then reconnect it. That fixed the problem for me.
jim
ISY with Z-wave dongle and a serial Insteon PLM
Over 100 Insteon devices
About 20 z-wave devices
A few TP-Link devices
A handful of ESP8266 node servers
Experimenting with Lutron - it looks really great. I got the picos working to control my Insteon devices (nice now that Insteon is out of business)
I had this problem and could not figure it out. I could ping but could not connect to the home page. IP was asssigned and static, checked the power supply and everything. I also had a link light for connectivity.
After spending a lot of time trying to troubleshoot, i decided to log into my router page. Even though the device was active, no response. I was able to click on the device and under advanced settings, my router had the device setup as a network device. I changed the device type from network device to generic PC. Immediately i did this, i was able to connect. I am guessing the network has a different set of rules and connection types for network devices versus generic computers. Lets hope it holds up. Router type is a netgear orbi
Please note this only happened after i performed my most recent upgrade.
I have to newest HASSIO and HA Version installed on my Pi4 (white/red case).
Some in here are are saying that downgrading to hassio 5.4 is the solution. so, if so…how do i do that without loosing configurations? actual version is 7.0. this is a big gap.
24 hours later, my Raspi didn’t lose its Wifi connection even once - so, i’d call that one a success…
Also, Homematic integration is still connected whereas i had to manually restart / reconnect at least 4-5 times a day (after noticing it has gone offline of course…)
So, lets hope that whatever is borked in OS 7.0 gets fixed soon…
After being on 6.6 for a long time, i’ve updated to 7.4 a few days ago (and more recently to 7.5) and only had constant connection issues again.
To add insult to the injury, also at least daily full lockups - the Pi again completely vanished from the network.
I suppose i’ll be downgrading to 6.6 once again - i really just wonder how such an issue doesn’t seem to be more widespread and at the same time being so persistent for that many OS-versions?
My internet goes down about once a day and starts working if kill the power to and restart my HA Raspberry Pi that is connected via ethernet. I’ve made sure I’ve set HA to a static IP and reserved an IP on my router. I’m about ready to give up on HA and try something else because I can only tell my gf I’m trying to fix it for so long lol
I had the same problem. In my case the dhcp client was overflowed by many docker containers all requesting a lease. It can be recognized by seeing “route socket overflowed” in the journal of the dhcp daemon (journalctl -u dhcpcd).
A workaround is to add the following line at the end of /etc/dhcpcd.conf:
denyinterfaces veth*
Afterwards the service needs to be restarted (sudo systemctl daemon-reload followed by sudo systemctl restart dhcpcd)
I had the same issue after upgrading HAOS to the latest. After a while of not getting things to work, I reluctantly moved to openHAB which I don’t like. openHAB is buggy as heck. Nowhere close to the stability of Home Assistant. Anyhow, Interesting enough i had to install Ubuntu first in order to install openHAB and the issue started with Ubuntu. The exact issue I was having with HAOS. Turns out, there’s a bug with Realtek drivers in the Ubuntu OS.
With that said, is HAOS built on top of the Ubuntu framework and if so, is there access to default Linux commands?
I ran sudo apt-get install r8168-dkms rebooted and my wire network popped up.
Its completely annoing!