From af1347785e15c66f7893b909b1250c9f3d9018f9 Mon Sep 17 00:00:00 2001
From: ohmystack <jiangjun1990@gmail.com>
Date: Fri, 1 Feb 2019 18:37:19 +0800
Subject: [PATCH] Cleanup pipeline bootstrapper

As pipeline bootstrapper has been removed in #739 , the gitignore file
and deployment guide should be updated as well.
---
 .gitignore         |  4 +---
 developer_guide.md | 40 ++++------------------------------------
 2 files changed, 5 insertions(+), 39 deletions(-)

diff --git a/.gitignore b/.gitignore
index 139dc24a40bb..c0a9308b9e60 100644
--- a/.gitignore
+++ b/.gitignore
@@ -12,8 +12,6 @@ bower_components/
 
 # Build output
 dist
-ml-pipeline/ml-pipeline-app
-ml-pipeline/kf-app
 
 # Web server
 frontend/server/*.js
@@ -51,4 +49,4 @@ vendor
 bazel-*
 
 # VSCode
-.vscode
\ No newline at end of file
+.vscode
diff --git a/developer_guide.md b/developer_guide.md
index a2becc2a286f..3d632512b88d 100644
--- a/developer_guide.md
+++ b/developer_guide.md
@@ -4,34 +4,11 @@ This document describes the development guideline to contribute to ML pipeline p
 
 ## ML pipeline deployment
 
-The ML pipeline system uses [Ksonnet](https://ksonnet.io/) as part of the deployment process.
-Ksonnet provides the flexibility to generate Kubernetes manifests from parameterized templates and
-makes it easy to customize Kubernetes manifests for different use cases.
-The Ksonnet is wrapped in a customized bootstrap container so a user don't need to explicitly deal
-with Ksonnet to install ML pipeline.
-
-The docker container accepts various parameters to customize your deployment.
-- **--namespace** the namespace to deploy to
-- **--api_image** the API server image to use
-- **--ui_image** the webserver image to use
-- **--report_usage** whether to report usage for the deployment
-- **--uninstall** to uninstall everything.
-
-See [bootstrapper.yaml](https://github.com/kubeflow/pipelines/blob/master/bootstrapper.yaml) for examples on how to pass in parameter.
-
-Alternatively, you can use [deploy.sh](https://github.com/kubeflow/pipelines/blob/master/ml-pipeline/deploy.sh) if you want to interact with Ksonnet directly.
-To deploy, run the script locally.
-```bash
-$ ml-pipeline/deploy.sh
-```
-And you will se a Ksonnet [APP](https://ksonnet.io/docs/concepts#application) folder generated in your current path. If you want to update or delete the K8s resource created by the deployment, run
-```bash
-# Update
-$ cd ml-pipeline && ks apply default
-# Delete
-$ cd ml-pipeline && ks delete default
-```
+### GKE
+Please refer to this [guide](https://www.kubeflow.org/docs/pipelines/pipelines-quickstart/)
 
+### Minikube
+Please use [kubeflow ksonnet registry](https://github.com/kubeflow/kubeflow/tree/master/kubeflow)
 
 ## Build Image
 
@@ -82,15 +59,6 @@ For example, to build API server image
 $ docker build -t ml-pipeline-api-server -f backend/Dockerfile .
 ```
 
-### Update deployment image
-If your change updates deployment image (e.g. add new service account, change image version etc.),
-remember to update the deployment image as well, and use that image to create deployment job.
-```bash
-$ docker build -t gcr.io/<your-gcp-project>/bootstrapper ml-pipeline/
-$ gcloud auth configure-docker
-$ docker push gcr.io/<your-gcp-project>/bootstrapper
-```
-
 ## Unit test
 
 ### API server