Hey all, just wanted to reach out to let y’all know that orbit 1.54.0 rolled out to our linux endpoints and it broke the connection to our (admittedly very outdated) fleet 4.19.0 server.

Hi Max, thanks for bringing this to our attention! What distro are you running when you see the crash?

The client that i posted the logs from is running Flatcar container linux. But it’s also the same behavior on some of our old ubuntu bionic systems too.

Is there any chance you could upgrade to the latest Fleet server version?

It’s on our to-do list. It wasn’t a high priority before now since it was still working.

Were you able to roll back orbit?

I was struggling to find out how to do that since we don’t use fleetd and instead run orbit directly.

Any chance you can use agent configurations/update_channels to downgrade the hosts? https://fleetdm.com/docs/configuration/agent-configuration#update-channels

I was looking into it, i think by the time we try downgrading the individual hosts, we might as well upgrade the server side. I just wanted to make you guys aware of the incompatibility.

Understandable, thanks for letting us know!