Can't access local or remote after a random time

Hey,
After update to the 23.10, I startet to have problem with loosing access both local and remote. It can work normal a day or maybe 2, but then, all dead. Need to do a hardware reset to get in back online.

It did get better now after 23.11 update, almost 3 days online.
I can’t see to find out what’s the problem is, I have it connected to a screen so here a photo of the screen before I reboot.

Please help.

Running HAOS, on a Dell Optiplex with SSD.

  • Core 2023.11.1
  • Supervisor 2023.10.1
  • Operating System 11.1
  • Frontend 20231030.1

When it not accessible can you ssh or input command line command? What is physical condition when offline?

Only command on screen, I use “ha host reboot” command to reboot when it locked. Can’t access from anything else.

Did take some logs photos, somebody understand something?



Have disabled wireguard now to see if that helps…

There was recently a “brakin” change (to some extend) … Renaming of certain network interfaces ! , this should/would be done peer automatic … HOWEVER, if you have/had any configurations anywhere, where you have hard-coded the name of your “now old” network interface, you had to change this. !
Now Did You Do This ?

New name eth0 !!! you have a " vethbdo774f " And another " vethbdo " … you know your " Setup " best , so make sure you change everywhere to eth0 … it seems to “keeps changing” in the last log

PS: Read the release-note 2023.10.0 or was it 11.0 :wink:

type: net inf

then: net -h

read

:slight_smile:

That make sense, but I cant find how to change my interface to eth0, know how?

[core-ssh -]S ha net info
docker:
address: 172.30.32.0/23
dns: 172.30.32.3
gateway: 172.30.32.1
interface: hassio
host_internet: true
interfaces:
connected: true
enabled: true
interface: enpOs31r6
1pv4:
address:
192.168.1.99/24
gateway: 192.168.1.1
method: auto
nameservers:
192.168.1.1
ready: true
1pv6:
address:
fe80::F627:c556:2c1b:dre8/64
gateway: null
method: auto
nameservers:
ready: false
mac: 18:66:DA:40:2A:E8
primary: true
type: ethernet
vlan: null
wif1: null
connected: false
enabled: false
interface: wlp2so
1pv4:
address:
gateway: null
method: disabled
nameservers:
ready: false
ipv6:
address:
gateway: null
method: disabled
nameservers:
ready: false
mac: B8:8A:60:7C:44:A3
primary: false
type: wireless
vlan: null
wifi: null
supervisor_internet: true

hmmm weird , there it says something else

interface: enpOs31r6
1pv4:

Seems like the interface-name in the log maybe are some “virtual” interfaces (veth******" )

but the "net info " also shows something “strange” " 1pv4 " should be " ipv4 " , like below ( only the last interface “disabled” says “ipv6”, but for that same interface the ipv4 says " 1pv4 "

Im actually in doubt now ( not only about this " 1pv " ) but also whether you actually needs to change the “interface: enpOs31r6” name, “enp” don’t seems to be among the ones that had to be changed in the previous update.( i run also on an old Dell Latutitude (with enp* net-interface) no need too change ( However i don’t have any hard-coded, or suspicious virtual network interfaces configured in any places or ADD-ONs/integrations

Tho it was these “kernel infos” i was suspicious about

Some path to /desktopmanager, but i also noticed you are running on a “Ubuntu”, beside the “Warnings” to various ROG-Laptop_network_card_count / Windows machine ( d_not_available ‘offline’ this is not expected.Please correct your configuration) etc … Very hard to tell what you actually see(can read from your screenshots) , so , i think a trow in my “Towel” here

Read your logs, try to get your instance up and running, start to disable whatever you see in your logs which seems to be related

BTW, how are you able to get Errors related to Ubuntu, in your HAOS Installation ?

I believe Ubuntu error is browser connection error

Honestly you troubleshooting this many days. I would reinstall OS >> verify it run as default >> then restore with my backup config. This presumes you have technical ability to restore with ease, if not then it could just be another mess.

I havent troubleshooting anything for days, i had a quick look at his log-screen shot, yesterday, and saw “info” i suggested he should “look at”
And my second commend, was answer to his answer/respond to me

And as OP seems like a “New” user, i would not just suggest “reinstall all”, And i have no idea how many hours he has spend on this, in middle of a week he could just be working, and picking up kids , for eventually find time to look at his failing Installation
But I tend to agree (with the limited (screenshot-info) and/or performed actions, system.setup etc. )
Restoring however, might not even be best, as it could be caused by “Configurations” which will then just be applied during a restore, from backup

but the "net info " also shows something “strange” " 1pv4 " should be " ipv4 " , like below ( only the last interface “disabled” says “ipv6”, but for that same interface the ipv4 says " 1pv4 "

This may be from screen copy with phone, so is sees 1 and not i, so that is nothing wrong, sory about that.

i run also on an old Dell Latutitude (with enp* net-interface) no need too change ( However i don’t have any hard-coded, or suspicious virtual network interfaces configured in any places or ADD-ONs/integrations

I run a Optiplex at my parents and been problem free for over 1 year.

but i also noticed you are running on a “Ubuntu” (…)
BTW, how are you able to get Errors related to Ubuntu, in your HAOS Installation ?

May be that a did use a Live ubuntu to transfer haos from usb to ssd.

ROG-Laptop_network_card_count / Windows machine

Thats was a sensor on my laptop, I have fixed that.

Honestly you troubleshooting this many days. I would reinstall OS >> verify it run as default >> then restore with my backup config. This presumes you have technical ability to restore with ease, if not then it could just be another mess.

Think its best to reinstall and now direct to the ssd (got a usb to ssd dongle now), so I can get a fresh install .

1 Like

Ok, sounds like a good idea, If you install a fresh, you can retrieve alot configurations, Dashboards .jason and yaml-files etc, from your backup-filies ( So save some of these, if you have alot, you would like to restore) (and fail to restore fully from a backup), and don’t want to start on a total “fresh”,

1 Like

I have now installed fresh HAOS directly on the SSD, so hoping my problem is solved.