ethonarm_21.03.00 won't finish setup

[eccentricexit]

I flashed my sd card with this image a few times today and the setup never finished.

Final result is an empty home folder and no geth.service running.

It seems cloud-final.service fails to start Execute cloud user/final scripts. (see image).

Output of sudo systemctl

Image: ethonarm_21.03.00.img.zip The hash matches 725359703b7c321f56a0e193be61c1f0102a23463549285e8f286e9fb6cc522f

Hardware

    Raspberry 4 (model B 8GB);
    MicroSD Card (128 GB);
    4 TB SSD USB 3.0 disk or an SSD with an USB to SATA case;
    Connected to internet via cable .

[eccentricexit]

More potentially useful information:

apt update is not working:

[eccentricexit]

I've installed Ubuntu 20.04 manually using Imager and saw similar messages when trying to apt update.

I managed to fix it by changing the clock to my local time following the steps here: https://www.how2shout.com/linux/fix-inrelease-is-not-valid-yet-invalid-for-another-h-min-s-updates-for-this-repository-will-not-be-applied/

• I traveled with my raspberry and changed timezones before trying the image.
• I also noticed when starting the image that the terminal output January 1st 1970

This might be the solution to the problem I faced with the ethonarm image. Perhaps we can let teh user select the timezone during setup?

[diglos]

Hi, never saw this before, we will take a look, sure. Thanks for reporting!!!

[eccentricexit]

Thanks

I decided to give it a go again and I have more information. I noticed the following: The date is synced to some time in 2020

systemctl thinks its been running since the 70s

I went ahead and manually set the date using timedatectl set-time but it was moved back again after rebooting.

[diglos]

Hi,

Could you please run this command and paste here the output?:

timedatectl status

[eccentricexit]

Yes:

After some tinkering I restarted the service and noticed a timeout on the log

Timed out waiting for reply from 200.160.7.193:123 (200.160.7.193).

That is a brazilian NTP server. Don't know why its timing out.

Anyway, I went ahead and changed the /etc/systemd/timesyncd.conf to use pool.ntp.org instead and it synced after restarting the service:

After a reboot it went to another incorrect date because I had not systemctl enable systemd-timesyncd. After doing that I get the correct time on boot.

apt update continues to not work and the setup did not finish (no geth.service)

[diglos]

Seems that only the Grafana repo is affected. Please try:

wget -q -O - https://packages.grafana.com/gpg.key | sudo apt-key add - apt update

Can you try this commands and paste the output:

dpkg -l | grep ethereum
dpkg -l | grep geth

[eccentricexit]

Outputs:

Also, in an attempt at getting things to move I went ahead and ~went nuclear~ deleted /root/first-run.flag hoping that just running the script again would safely produce the output I wanted and now I have geth syncing in the background :)

[diglos]

Yeah, seems you ran into a problem with the NTP server. I'm closing the issue as it is not related with the image. Feel free to reopen if you think otherwise.