Agree that just ignoring the etcd error is the wrong thing to do here. If we can't determine whether or not we're still the master, we should exit. In most cases, this will move the crash loop up a level (systemd continually respawning the daemon), but I think that's preferable as there's additional monitoring for such cases.

Firefox has signaled they're now also going to support Manifest V3, by the end of 2022: https://blog.mozilla.org/addons/2022/05/18/manifest-v3-in-firefox-recap-next-steps/

I've uploaded a new image, incorporating some changes from the two weeks:

Also, there's a new image which resolves the Geckodriver/libc issue.

In T309041#8016565, @larissagaulia wrote:

@dpifke Could you share the link for the documentation, please?

In T309041#8008320, @Peter wrote:

It seems like geckodriver do not work:

wmf@rpi4-20220616:~$ geckodriver --version
geckodriver: /lib/aarch64-linux-gnu/libc.so.6: version `GLIBC_2.33' not found (required by geckodriver)
geckodriver: /lib/aarch64-linux-gnu/libc.so.6: version `GLIBC_2.32' not found (required by geckodriver)

New image is uploading now, with the following fixes:

Debian by default turns off the green and red LEDs to indicate booting has completed. So the Raspberry Pi itself is likely still on, just the LEDs are off.

Glad that helped. I'm going to leave this open and assigned to myself to make the error message more helpful.

Can you try running:

Also, from a practical standpoint, I think we're much more likely to get approval to add these if we DON'T require a bunch of custom network setup and ongoing maintenance/complexity burden.

We should talk through what we would hope to accomplish by not having them connected to the rest of the DC.

systemd in bullseye (and up?) appears to object if User=nobody, see: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=969329. Logspam like the following is emitted:

Old hosts in deployment-prep have been deleted.

I think we're good to go.

Curious why we're using Ubuntu here instead of Debian?

Some questions:

No objections from xhgui owners, the service is low traffic and should automatically reconnect.

Thanks for the advance notice! Please keep us updated on the timeline for this being ready; we will keep it on our radar for next quarter's planning.

We (performance team) need to do some evaluation of the impact of this, and are hesitant to make such a change before the holidays.

I've dropped a bullseye virtual machine image in my home directory on integration-agent-qemu-1003 and updated the instructions at https://www.mediawiki.org/wiki/Continuous_integration/Qemu#Runbook to reflect how I created it.