On Thu, Oct 27, 2016 at 5:08 PM, Rodrigo Campos wrote:
> On Thursday, October 27, 2016, 'Robert Bailey' via Kubernetes user
> discussion and Q wrote:
>
>> Hello all,
>>
>> *TL;DR*
>>
>> I've just enabled notifications from Stack Exchange
There are two things, the deployment itself and the live/ready checks.
A new deploy should be easy because you can gracefully stop the leader,
that results in a new leader elected between the others and everything is
fine. You can even change the default setting on the rolling update to
first
Does a new replica actively have to join the cluster and decide whether
it's going to be a leader or follower? If that's the case, then you could
manage the set of (unknown, follower, leader) states in your
liveness/readiness probe and only return a positive result on successful
determination. (My