Deployment of my app broke today. Based on the logs, I think it’s an issue with the fly platform, and nothing to do with my application.
Running <app-name> release_command: /app/bin/migrate
> Created release_command machine 0801e43b690208
> Waiting for 0801e43b690208 to have state: stopped
> Machine 0801e43b690208 has state: stopped
Starting machine
> Waiting for 0801e43b690208 to have state: started
> Waiting for 0801e43b690208 to have state: started
> Machine 0801e43b690208 has state: started
> Waiting for 0801e43b690208 to have state: destroyed
> Machine 0801e43b690208 has state: destroyed
> Waiting for 0801e43b690208 to get exit event
✖ Failed: error finding the release_command machine 0801e43b690208 exit event: error getting machine 0801e43b690208 from api: failed to get VM 0801e43b690208: machine not found
Error: release command failed - aborting deployment. error finding the release_command machine 0801e43b690208 exit event: error getting machine 0801e43b690208 from api: failed to get VM 0801e43b690208: machine not found (Request ID: 01JMBCP3T0W0704QYMZ861ZPJ8-iad) (Trace ID: e98c1d999df370f8e67dacbde3b66997)
It looks to me like something in the deployment platform is polling to wait for the release_command to finish being destroyed, but then crashing once it’s successfully destroyed.