Updating in the past week or so is not proceeding

Hi
I haven’t seen this issue before. Occasionally I have either supervisor or core updates take some time. Sometimes I get them successful by unticking the do backup before install. Usually I do a restart of either HA or in extreme cases (where I have tried 3 times) I do a hard reboot of the Intel NUC where HA is running.
For context I have the following

  • Core2024.7.4
  • Supervisor2024.06.2
  • Operating System12.4
  • Frontend20240710.0

I have both core update 2024.8.0 and supervisor 2024.08.0 staring at me in the update area of settings.
Trying to update using the gui button just leads to spinning of the icon - this can go on for hours (tried an overnight one). Happens on core and supervisor.
I thought I would then try in the CLI.
The core update process returned the following
Post “http://supervisor/core/update”: context deadine exceeded (Client.Timeout exceeded while awaiting headers)

I looked at LOGS but the only thing in them is the list of HACS integrations that I have been running for at least 2 years without them causing this issue.

At the time of writing this the CLI command to update supervisor is still in the running phase (started about two hours ago)

Any ideas?

Pat

Is the NUC busy on other VMs too?
There is a breaking change first in the core update about a database change that will take time and you should not interrupt it.

Check how much available disk space you have left on your HA installation

Wally,
NUC dedicated to HA.
Pat

ShadowFist
disk is 6% used 390Gb free.
Pat

does the supervisor logs tell you something?

Wally,
logs for supervisor don’t show anything critical or indicating why no updates occur.
I have mention of a system autofix complete and then system updated Home assistant API token that was this morning??
Pat

What ISP?
There have been some ISPs that block GitHub downloads, which HA relies on.

I’m using Starlink (rural australia)

Just a data point, no problems here with Starlink (NZ).

Yeah I can still acces github here. I also wondered if I went to my instance through Nabucasa to see if there was any chance - nope.
I’ll keep looking

1 Like

Can you access ghcr.io though?

Yeah that link takes me straight there.

Does it resolve from your ha machine

Nick
now thats a question I don’t have an answer for. If i’m running HA OS and the machine has no other interface - how can I check that? Do i need to use CLI?

Do you have one of the ssh add-ons (everyone should)

Can you plug a keyboard and monitor in to the machine?

Yeah i’ve got ssh. Will take 15 min to set up the keyboard and monitor.

If you have ssh you won’t need the monitor and keyboard.

Ssh into the machine from your desktop and your

nslookup ghcr.io

NIck
provides answer of
Server: 172.30.32.3
Address: 172.30.32.3#53

Non-authoritative answer:
Name: ghcr.io
Address: 4.237.22.32