Code review comment for lp:~mpontillo/maas/bug-1458894-ioerror-during-image-download

I don't really like this as it affects the region calling the cluster to update the boot images. I don't know if waiting in the thread for 30 seconds is the best thing to do as it will hold the import lock and it will look like from the region that the cluster is importing images, but really it is waiting 30 seconds to try again.

I think this retry logic should be handled in the ImageDownloadService, and can be used to pause inside the reactor. Causing it to not block a thread.