Andreagit97
commented
8 months ago Thank you very much for the amazing proposal! I'm fully onboard!
-[ ] Refactor Install, Running, and Deployment merging all steps for each type under the type (host, docker, Kubernetes) simplifies accessing the information you need and also navigating the documents. Deployment is a generic term that covers Kubernetes (so it should be renamed).
Just one side question/suggestion, why don't we use minikube+modern-ebpf as a unique example way to deploy Falco on Linux/Mac/Windows? I see here https://falco.org/docs/getting-started/falco-linux-quickstart/ that we have quite detailed steps for example with lima on Apple silicon (M1/M2), is there a particular reason for using lima? If not, we could simplify the whole page, with a simple:
minikube start --driver=virtualbox
helm install falco falcosecurity/falco \
--set driver.kind=modern-bpf
poiana
Dear Falco community, @mikegcoleman, @n1g3ld0ugla5 and I would like to propose improvements to the "Install and Operate" documentation.
The current structure is the following:
The new (proposed) structure is the following:
The steps to improve the documentation and the rationale are the following:
[ ] Split Install and Operate into separate sections. split and operate are two different parts. Install contains day-1 tasks, while operate contains day-2 tasks.
[ ] Add new "Introduction" sections the install section is one of the first sections that users go to. It is important to have an overview of Falco and discuss some central topics that apply to every installation. The "Operate" section will also need an introduction.
[ ] Refactor Install, Running, and Deployment merging all steps for each type under the type (host, docker, Kubernetes) simplifies accessing the information you need and also navigating the documents. Deployment is a generic term that covers Kubernetes (so it should be renamed).
[ ] Merge "Third Party Integrations" content into other sections the current content under this section is not about third-party integration, but about other environments (so it should be merged).
[ ] Move Upgrade and Falco Performance to a separate entry as we said before, Operate should have day-2 tasks. For now, those are the two pages with day-2 content. We expect later expansion of this section later (e.g. troubleshooting).
[ ] Rethink Learning and Production Environments this one is not set in stone as there are a few possibilities and we couldn't reach an agreement (please, let us know your thoughts). "Production Environments" is a misleading title for CoreOS, EKS, and GKE (every production environment is a production environment regardless of the infra). "Learning Environments" is not a great name either. Also, k3s, minikube, kind, EKS and GKE could be under Kubernetes. However, there might be more tips for specific environments in the future, such as open-shift. Shall we get rid of this page and merge the content into other pages or keep all tips in here?
/area documentation