Yeah... but I think that stopping the last master shouldn't kill the cluster
either.

On Thu, Mar 3, 2011 at 4:17 PM, Ryan Rawson <[email protected]> wrote:

> with multiple masters, shutting down a master should NOT cause a
> cluster death!
>
> I ran in to this once, sucked.
>
> I have previously commented, I thought we had removed the 'master exit
> = cluster death' but I'm not sure.
>
> -ryan
>
> On Thu, Mar 3, 2011 at 4:14 PM, Ted Dunning <[email protected]> wrote:
> > I think that the proposal on the table is to actually simplify things a
> bit
> > by making the shutdown of the master
> > not cause the shutdown of the regions.  Less coupling is simpler.
> >
> > On Thu, Mar 3, 2011 at 2:47 PM, M. C. Srivas <[email protected]> wrote:
> >
> >> To tell you the truth, I really like the simplicity that exists today in
> >> launching a hbase cluster
> >>   - start the master somewhere
> >>   - start the RS's wherever you like
> >>
> >> And it just works! Very nice.
> >>
> >> Perhaps adding an acl on the master in is order to prevent the
> unexpected
> >> shutdown (or extra master, or other funky stuff from accidentally
> >> happening).
> >>
> >> Making it more complex to start/shutdown doesn't really help. Many of us
> >> will then write a wrapper script to make it simple again, and we are
> back
> >> to
> >> square one.
> >>
> >>
> >>
> >> On Wed, Mar 2, 2011 at 8:57 PM, Stack <[email protected]> wrote:
> >>
> >> > On Wed, Mar 2, 2011 at 5:23 PM, Bill Graham <[email protected]>
> >> wrote:
> >> > > So the question is, is this a bug or a feature? If it's a feature it
> >> > > seems like an incredibly dangerous one. Once our live cluster is
> >> > > running, those configs will also be needed on the client so really
> bad
> >> > > things could happen by mistake.
> >> > >
> >> >
> >> > One time it might have been thought a 'feature' but now its a bug I'd
> >> > say given the above.  I opened
> >> > https://issues.apache.org/jira/browse/HBASE-3590 using this thread as
> >> > context.
> >> >
> >> > St.Ack
> >> >
> >>
> >
>

Reply via email to