Admin cannot acquire machine that failed to deploy without first releasing it

[Nathaniel W. Turner]

MAAS 3.0-rc1. Personal severity level: "not a big deal"

A newly-commissioned machine failed to deploy because the storage config was wrong.
The only way to fix this was to first Release the machine back to the pool and then quickly Acquire it before anyone else tried to deploy it.

Could perhaps "Abort..." be an available action for nodes in the "Failed to deploy" state, allowing them to transition directly to Allocated?

[Björn Tillenius (bjornt)]

I think going from Failed Deployment to Acquired makes sense. We have a similar case, that you should be able to release a Deployed node and have it got to Acquired rather than Ready.

[Jerzy Husakowski (jhusakowski)]

MAAS should offer more flexibility in changing deployment parameters. We want to look at node lifecycle management and state transitions, and it's likely that this particular issue will be addressed together with other lifecycle management issues in a future release.