On Mon, Jul 9, 2018 at 8:02 PM Joao Eduardo Luis <joao@xxxxxxx> wrote: > > On 07/09/2018 05:48 PM, Jesse Williamson wrote: > > > > Hi everyone, > > > > I'm adding a feature to allow groups of flags to be set or unset (e.g. > > "ceph osd set noup norecover noscrub"). Some of these flags (such as > > "recovery_deletes") require the user passing "--yes-i-really-mean-it" > > (YIRMI) on the command line in addition to requesting setting the flag > > before they will operate. > > > > What do we want the behavior to be in the case where a user tries to set > > multiple flags that require YIRMI? Here are the possibilities I've > > considered so far: > > > > 1) We can take a single YIRMI to imply approval for any such flags; > > 2) We can refuse to set groups of flags in the presence of a single > > YIRMI (i.e. a single YIRMI causes all sets to be rejected); > > 3) We can require an individual YIRMI for each flag that requires it > > (one for each flag); > > 4) We have have two commands (currently implemented): the existing "set" > > behavior handles a single flag and YIRMI as it currently does, and a new > > "setall" command only handles non-YIRMI flags. > > > > What of those (or some other behavior) sounds most correct to everyone? > > #1 seems the correct one. I think it's reasonable to expect that a > single `yes I really mean it` flag will be applicable to all flags being > set/unset. I agree with Joao. John > -Joao > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
