Hi Devin, What are the exact configuration changes that force you to restart the cluster? What is your process of onboarding a tenant?
The reason I'm asking is that adding new caches should not require a cluster restart, even in the current version, but it's possible that you're hitting one of the edge cases. Please provide more details so that I could properly address the issue. -Val On Tue, Sep 15, 2020 at 11:14 AM Devin Bost <devin.b...@gmail.com> wrote: > Hi Everyone, > > I'd like to start a discussion on how to eliminate the need to restart the > entire cluster whenever configuration changes need to be made to caches. > The need to deactivate and restart the entire cluster every time config > changes must be made has caused problems with our ability to scale. Every > time we onboard a new tenant to Ignite, we increase the frequency of > needing to restart the cluster because each tenant has their own needs, and > with each restart, we increase the downtime for all of the tenants. One > workaround we've considered is setting up a separate cluster for each > tenant, but this large number of clusters wastes hardware and is expensive > to maintain. How can we resolve this issue? > > I asked this question in the chat during the webinar today, but we ran out > of time before the question could be addressed. > > Thanks, > > Devin G. Bost >
