James Falcon

@catmsred and @aciba90 , where are we at on this one?

No problem; there's no rush on this one. I just wanted to make sure nobody was blocked or waiting for somebody else.

@blackboxsw , looks like the last comment on that bug is from 2023-10-20 . Do we expect that to be SRUed anytime soon?

@ddstreetmicrosoft Thanks for this PR. Does this mean that the Mariner datasource is no longer relevant? Should we remove it at some point? If you check the CI runs, you'll...

> Alternately, I could open a separate PR to remove 'Mariner', either soon or closer to when it goes EOL (which will be late this year at the earliest, more...

The 3.13 test failure is unrelated and being addressed in https://github.com/canonical/cloud-init/pull/5052 . If everything else passes I'll merge as-is.

Cloud-init isn't complaining about the empty user data. It's complaining about the network configuration. It looks like your network configuration contains an invalid key. `ipv6` isn't a key in [network...

This network data comes directly from the cloud, which would be your Openstack fabric in this case. It has likely always contained the invalid key. What is new is that...

@quatre , thanks for the additional context here. This is a bug in cloud-init. It looks like the lines on 661-664 simply shouldn't exist. They don't get used in the...

My last comment was incorrect. I did find a place in the sysconfig renderer where this is used. Fixed via https://github.com/canonical/cloud-init/pull/5191