Machines create successfully but get permanently stuck in created state on ORD

[rnkn]

Machines create successfully but get permanently stuck in created state. They never transition to started. This happens with both fly deploy and fly machines run via the CLI. I have a critical data volume in ORD that I cannot lose. I also have ~12 orphan volumes from a previous rolling-deploy loop that are stuck in pending-destroy state. I’ve tried destroying those volumes, it says they are destroyed in CLI, but the dashboard still shows them there. Not sure what’s happening since I’ve been using Fly for over a month now.

I need help getting a machine to actually start in ORD (or migrate my volume to a working region). Also force-deletion of the orphan volumes, since it’s potentially bloating my usage.

[mayailurus]

@rnkn

or migrate my volume to a working region

Hi… You should be able to clone it into a different region via fly vol fork --region ewr.

Another route is to manually create a snapshot and then use that to create a new volume elsewhere.

@rnkn

force-deletion of the orphan volumes, since it’s potentially bloating my usage.

The screenshot says Pending_destroy, which doesn’t count against your bill…

Volumes in pending_destroy state do not accrue charges.

The CLI (flyctl) is generally more authoritative than the dashboard’s lists are, in my opinion.

[rnkn]

Thank you so much, the fork fixed it! I figured it was probably something simple.