the best of the three

this one checks for more than just the status, it considers submirror count too, thus ensuring you don’t forget to complete the process of replacing a failed disk!

it’d be even better if it checked the metadb too.