We use cinder devices heavily in our clusters, and failed kitchen runs will delete the instances, but not the associated volumes.
This seems to occur if there is a problem with instance creation (like a quota)....in these cases, kitchen fails to clean up the volume it created.
We use cinder devices heavily in our clusters, and failed kitchen runs will delete the instances, but not the associated volumes.
This seems to occur if there is a problem with instance creation (like a quota)....in these cases, kitchen fails to clean up the volume it created.