Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Keepalived based HA master #128

Open
orymate opened this issue Oct 29, 2020 · 2 comments
Open

Keepalived based HA master #128

orymate opened this issue Oct 29, 2020 · 2 comments

Comments

@orymate
Copy link
Contributor

orymate commented Oct 29, 2020

Support automatic keepalived installation/configuration

@immanuelfodor
Copy link

Extremely useful feature and great idea to support it out of the box, huge +1 :)

May I ask why did you choose Cisco's VRRP over the open-source CARP protocol? Do you just want keepalived and that's it? Or is there a specific use case why one is better than another in light of PKE?

For example, when I was looking for a solution of API server HA, I've found UCarp much easier to install and configure than keepalived (example repo: https://github.com/immanuelfodor/kube-karp). Since then we are using it within our in-house clusters without a problem. I'm honestly interested in why keepalived, maybe we need to switch in the future and good to think ahead.

@gecube
Copy link
Contributor

gecube commented Jan 25, 2021

@immanuelfodor Hi! Thanks for your comment! We are in the same issue. I am building the kubernetes-based PaaS platform with my colleagues and we checked a lot of approached for building HA clusters. Unfortunately, ANY of the possible approaches has its own disadvantages and nuances. For example, keepalived or CARP are working only in L2 environments, AFAIK. So it is not the option for cloud providers and deploying the kubernetes on top of VMs. But on the other hand it's well known and working solution for bare metal and vmware environments.
The approach used by kubespray - deploying of haproxy with the list of all master nodes on any of worker nodes - has another issues. Etc. etc.

p.s. I'd like to emphasize that I am not connected in anyway with BanzaiCloud and all written above is my personal opinion.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants