How to deploy and setup a Fleet cluster?

[Danny_Dumenigo (Danny Duménigo)]

Hello community!
I have two environments of ELK v7.17 deployed on-prem (test and prod), so I decided to try Fleet. Im still testing things, so, In my test env I was able to setup a Fleet server (with certificates generated using certutil tool), enroll some agents and everything goes fine. Now, before move to production and to fullfil requirements of high availabilty, fault tolerance, etc, I want to deploy a cluster of Fleet servers, but was unable to figure it out how to do it, and I haven't found any clue over the internet.
I have try to enroll a new server using the same fleet-server-policy (logically throw some errors, because an agent can only have one policy apply), install another fleet server (for a total of 2) but can't make them communicate each other (I don't know if I have to make an extra configuration... maybe in fleet.yml?)

Any help will be appreciate.
Best Regards

[nchaulet (Nicolas Chaulet)]

Hi @Danny_Dumenigo

In a high availability scenario Fleet servers will not have to talk to each other (they get the info they need via Elasticsearch)
You to follow the recommandation here and have multiple Fleet server behind a load balancer Fleet Server deployment models | Fleet and Elastic Agent Guide [8.0] | Elastic

[Danny_Dumenigo (Danny Duménigo)]

Hi @nchaulet Thanks for the answer!!

I understand the main idea behind the image. I will do some tests with this concept and hope everything will go well.

Thanks again!

Best Regards
Danny