I tried to update home assistant from 103.5 to 103.6 today through hassio as I normally do. I am running on ubuntu 18.04 latest patch.
I can now no longer access home assistant, but I can ssh and some of my other containers are running. I’ve seen a few people mention stuff about the google backup addon so I’ve tried killing that. I’ve also read about the ip addresses in the containers, so I’ve tried stopping all containers, removing all containers and rebooting the server.
I notice in the images I have both home assistant 103.5 and 103.6. I also have hassio supervisor 193 and latest (the same image I think).
The containers that are being run are:
homeassistant/qemux86-64-homeassistant:0.103.5 and homeassistant/amd64-hassio-supervisor
Looking at the logs for both I’ve seen nothing alarming. I’ve also looked at the ifconfig by exec in to each and checking against the hassio dns hosts:
My supervisor logs after killing containers and restarting are:
20-01-08 22:56:08 INFO (MainThread) [hassio.snapshots] Found 4 snapshot files
20-01-08 22:56:08 INFO (MainThread) [hassio.discovery] Load 2 messages
20-01-08 22:56:08 INFO (MainThread) [hassio.ingress] Load 3 ingress session
20-01-08 22:56:08 INFO (MainThread) [hassio.secrets] Load Home Assistant secrets: 22
20-01-08 22:56:08 INFO (MainThread) [main] Run Hass.io
20-01-08 22:56:08 INFO (MainThread) [hassio.api] Start API on 172.30.32.2
20-01-08 22:56:08 INFO (MainThread) [hassio.addons] Phase ‘initialize’ start 0 add-ons
20-01-08 22:56:08 INFO (MainThread) [hassio.core] Hass.io reboot detected
20-01-08 22:56:08 INFO (MainThread) [hassio.tasks] All core tasks are scheduled
20-01-08 22:56:08 INFO (MainThread) [hassio.core] Hass.io is up and running
20-01-08 22:56:23 WARNING (MainThread) [hassio.tasks] Watchdog found a problem with Home Assistant Docker!
20-01-08 22:56:23 INFO (SyncWorker_5) [hassio.docker.interface] Start homeassistant/qemux86-64-homeassistant
Well I had a similar issue.
I was on the beta channel.
Last night system unresponsive.
rebooted
Woke up this morning and still the same, cpu % going crazy.
Checked docker ps
Only listed hassio_supervisor and hassio_dns
Paused the both and cpu % 0
Loaded portainer
Attempts to start the same issue
Found dns spiking cpu
Supervisor in a restart loop
Tried to start things some did others didnt
no samba, not much of anything
Checked images
Saw supervisor had two
194 and latest. I’m sure I was only a on 193
so it trying to upgrade itself and failing.
Remove the latest.
Restarted supervisor
and hooray
Slowly it fixed itself, updated and update home assistant to 104.b0 supervisor back to 193. I think.
will have to check.
Turned off beta channel.
Now my hassio is wanting to upgrade to 103.6 from 104.b0
well at least it all working
For now
Same issue on docker not hassio when I upgraded to 0.104.0.0b0 dropped back to as far as 0.103.3 same issue.
Hopped back onto the dev channel now at 0.105.0.dev… same issue.
I think the two images for supervisor e.g. 194 and latest are the same image just with two tags, as in 194 is the latest (at least that’s what I do at work anyway). Whether this is a new process or it should only grab the version tag I’m not sure.
So am I correct in saying by deleting the latest supervisor image and restarting supervisor, seemed to get things back up for you? I’m not entirely sure what to do with two images for ha 0.103.5 and 0.103.6. I’m wondering if mine grabbed the new image for ha, perhaps the supervisor failed because of the latest tag like you’re saying, then failed and reverted to the 0.103.5 ha image.
I’m not going to be able to try anything till later tonight for me now, but I’ll post back if removing that latest image helps. As far as I could tell my supervisor was on 193 not 194 and I was on normal release channel.
if you on linux
install portainer its saved me
you can see whats happening, what is running, cpu, memory you can see the logs, containers, images
if i tried to do this from the cli i would have been lost as to what was happening
Removed latest image, now the one I have left is with tag 194, rebooted but can’t get in still. Webui starts when I use “docker start homeassistant” but hassio menu won’t load or deconz, z-wave loads fine…
dmesg:
[ 3464.892424] hassio: port 3(veth38b2ef3) entered forwarding state
[ 3464.896851] IPv6: ADDRCONF(NETDEV_UP): veth711baa6: link is not ready
[ 3464.897171] IPv6: ADDRCONF(NETDEV_CHANGE): veth38b2ef3: link becomes ready
[ 3465.118981] docker0: port 1(veth8344869) entered disabled state
[ 3465.119385] eth0: renamed from veth3d7ded7
[ 3465.146828] docker0: port 1(veth8344869) entered blocking state
[ 3465.146833] docker0: port 1(veth8344869) entered forwarding state
[ 3465.271202] hassio: port 3(veth38b2ef3) entered disabled state
[ 3465.271462] eth1: renamed from veth711baa6
[ 3465.291011] hassio: port 3(veth38b2ef3) entered blocking state
[ 3465.291016] hassio: port 3(veth38b2ef3) entered forwarding state
[ 3465.376742] udevd[6]: starting version 3.2.8
[ 3465.379238] udevd[7]: starting eudev-3.2.8
[ 3466.635458] audit: type=1400 audit(1578589107.460:33): apparmor=“STATUS” operation=“profile_replace” info=“same as current profile, skipping” profile=“unconfined” name=“hassio-supervisor” pid=25133 comm=“apparmor_parser”
[ 3466.635463] audit: type=1400 audit(1578589107.460:34): apparmor=“STATUS” operation=“profile_replace” info=“same as current profile, skipping” profile=“unconfined” name=“hassio-supervisor///usr/bin/gdbus” pid=25133 comm=“apparmor_parser”
[ 3466.635466] audit: type=1400 audit(1578589107.460:35): apparmor=“STATUS” operation=“profile_replace” info=“same as current profile, skipping” profile=“unconfined” name=“hassio-supervisor///usr/bin/git” pid=25133 comm=“apparmor_parser”
[ 3466.635469] audit: type=1400 audit(1578589107.460:36): apparmor=“STATUS” operation=“profile_replace” info=“same as current profile, skipping” profile=“unconfined” name=“hassio-supervisor///usr/bin/socat” pid=25133 comm=“apparmor_parser”
[ 3466.824913] hassio: port 2(veth560d3d2) entered disabled state
I’m back on looking at this now, something seems to have cause a problem in this last release… seeing a fair few people with similar issues in discord and elsewhere too. I’ll report back shortly…
So I had a play last night, stopped and removed all the docker containers and then rebooted the host. This resulted in no docker containers coming back up.
I then tried following frenks install instructions
sudu su
curl -sL https://raw.githubusercontent.com/home-assistant/hassio-installer/master/hassio_install.sh | bash -s
To no avail sadly.
Back to the drawing board… I did have a go at recreating the service under systemctl but no joy there either.