kubernetes
/
kops
mirror of https://github.com/kubernetes/kops.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #823 from DualSpark/plane-docs 

Work from 30k before kubecon - getting some notes into the tree for discussion points
This commit is contained in:
Chris Love 2016-11-08 01:15:12 -08:00 committed by GitHub
parent 8532644166 941ce272db
commit 60054495b0
1 changed files with 30 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
upup/pkg/fi/README.md Normal file
View File

@ -0,0 +1,30 @@
# fi - Fast Install
The `fi` package is responsible for holding the infrastructure that supports installing, provisioning and deploying resources to get kubernetes running in a cloud.
## Cloudup
The `cloudup` package within `fi` contains infrastructure that handles deploying cloud resources.
Cloudup has a concept of `models`. These are a core component of kops, and critical to understanding how the tool works. They offer flexibility for the tool, and serve as the main representation of infrastructure in the cloud. They are the glue that will map resources to kops functionality. These `models` are stored in `upup/models/cloudup`.
#### AWS
Currently `aws` is the most verbose and feature rich of the 2 primarily supported cloud providers. Cloudup will handle reading the `aws` models (found in `_aws`) and apply these resources to the cloud.
A user will notice `awstasks` and `awsup` directores within cloudup. These are the files that will actually map an `aws` resource (API Request) to go code.
By convention, every used API request in kops for `aws`, has an associated `awstask` with it.
#### GCE
## Nodeup
This tool is secondary to cloudup.
After the resources to the cloud have been deployed, we will now need to provision the resources.
Nodeup runs as a binary **on VMs in the cloud** and will handle bringing up a kubernetes cluster.
The core bits of `nodeup` infrastructure can be found here.