Comment 12 for bug 1466982

the task will succeed on a second run as is and on a new deployment it seems to work. its only in the case of an already cluster environment where this fails. swallowing failures is bad however it would seem that the upstream module needs work and IMO i think we start there to make this better for everyone however in the mean time to fix this I say we fire and forget but maybe add a comment to the role indicating why and that we intend to revise when the module is made better upstream.