Streamline getting started.

pull/17/head
jackgr 9 years ago
parent 1b0bb08207
commit b6b5c42866

2
.gitignore vendored

@ -3,4 +3,4 @@ expandybird/expansion/*.pyc
resourcifier/pkg/* resourcifier/pkg/*
resourcifier/bin/* resourcifier/bin/*
manager/pkg/* manager/pkg/*
.project

@ -21,65 +21,68 @@ Kubernetes configuration SIG.
## Getting started ## Getting started
For the following steps, it is assumed that you have a Kubernetes cluster up There are two ways to get started...
and running, and that you can run kubectl commands against it. It is also
assumed that you're working with a clone of the repository installed in the src
folder of your GOPATH, and that your PATH contains $GOPATH/bin, per convention.
Since Deployment Manager uses Python and will be running locally on your * The quick way simply installs Deployment Manager in your cluster using
machine, you will first need to make sure the necessary Python packages are kubectl. This is the fastest way to get started and takes only a few seconds.
installed. This assumes that you have already installed the pip package
management system on your machine.
``` * The interesting way bootstraps Deployment Manager, by building and running a
pip install -r expandybird/requirements.txt local instance on your machine, and then using it to install another instance
``` in your cluster. You might want to go this way if you're interested in contributing
to Deployment Manager.
Next, you'll build and install the binaries, and bootstrap Deployment Manager Both assume that you have a Kubernetes cluster up and running, and that you can
into the cluster. Finally, you'll deploy an example application on the run `kubectl` commands against it. They both also assume that that you're working
cluster using Deployment Manager. with a clone of the repository installed in the src folder of your GOPATH, per
convention.
### Building and installing the binaries Instructions for the quick install follow here. Instructions for bootstrapping
Deployment Manager can be found in [examples/bootstrap/README.md](examples/bootstrap/README.md).
In this step, you're going to build and install the Deployment Manager binaries. ### Quick Install
You can do this by running make in the repository root.
For the quick install, you're going to use `kubectl` to create the replication
controllers and services that comprise a Deployment Manager instance from a predefined
configuration file, as follows:
``` ```
make kubectl create -f install.yaml
``` ```
### Bootstrapping Deployment Manager That's it. You should now be able to see Deployment Manager running in your cluster
using:
In this step, you're going to bootstrap Deployment Manager into the cluster.
Next, start the three Deployment Manager binaries on localhost using the supplied
bootstrap script.
``` ```
./examples/bootstrap/bootstrap.sh kubectl get pod,rc,service
``` ```
The script starts the following binaries: If you see replication controllers named expandybird-rc, manager-rc and resourcifier-rc
* manager (frontend service) running on port 8080 with pods that are READY, and services with corresponding names, then Deployment
* expandybird (expands templates) running on port 8081 Manager is up and running.
* resourcifier (reifies primitive Kubernetes resources) running on port 8082
Note that you can also tear down Deployment Manager using the same file, with:
It also starts kubectl proxy on port 8001. ```
kubectl delete -f install.yaml
```
Next, use the Deployment Manager running on localhost to deploy itself onto the The easiest way to interact with Deployment Manager, now that it's up and running,
cluster using the supplied command line tool and template. is to use a `kubectl` proxy:
``` ```
client --name test --service=http://localhost:8080 examples/bootstrap/bootstrap.yaml kubectl proxy --port=8001 &
``` ```
You should now have Deployment Manager running on your cluster, and it should be This command will start a proxy that lets you interact with the Kubernetes api
visible using kubectl (kubectl get pod,rc,service). server through port 8001 on you local host. However, there are other ways to access
Deployment Manager. We won't go into them here, but if you know how to access
services running on Kubernetes, you should be able to use any of the supported
methods to access Deployment Manager.
### Deploying your first application (Guestbook) ### Deploying your first application (Guestbook)
In this step, you're going to deploy the canonical guestbook example to your Next, you're going to deploy the canonical guestbook example to your Kubernetes
Kubernetes cluster. cluster.
``` ```
client --name guestbook --service=http://localhost:8001/api/v1/proxy/namespaces/default/services/manager-service:manager examples/guestbook/guestbook.yaml client --name guestbook --service=http://localhost:8001/api/v1/proxy/namespaces/default/services/manager-service:manager examples/guestbook/guestbook.yaml
@ -94,7 +97,7 @@ kubectl get service
You should see frontend-service running. If your cluster supports external You should see frontend-service running. If your cluster supports external
load balancing, it will have an external IP assigned to it, and you should be load balancing, it will have an external IP assigned to it, and you should be
able to navigate to it to see the guestbook in action. able to navigate to it in your browser to see the guestbook in action.
## Building the container images ## Building the container images
@ -113,12 +116,14 @@ make push
## Design of Deployment Manager ## Design of Deployment Manager
There is a more detailed [design document](https://github.com/kubernetes/deployment-manager/blob/master/docs/design/design.md) There is a more detailed [design document](docs/design/design.md)
available. available.
## Status of the project ## Status of the project
The project is still in a very active development mode so you might run into issues, The project is still under active development, so you might run into issues. If
please don't be shy about letting us know when you run into issues. you do, please don't be shy about letting us know, or better yet, contributing a
fix or feature. We use the same contribution conventions as the main Kubernetes
repository.

@ -0,0 +1,67 @@
# Bootstrapping Deployment Manager
Welcome to the bootstrap example. The instructions below will step you through
the process of building and running a local instance of Deployment Manager on
your local machine, and then using it to deploy another instance in your cluster.
This example provides insights into how Deployment Manager works, and is
recommended for anyone interested in contributing to the project.
The instructions below assume that you have a Kubernetes cluster up and running,
and that you can run `kubectl` commands against it. They also assume that that
you're working with a clone of the repository installed in the src folder of your
GOPATH and that your PATH contains `$GOPATH/bin`, per convention.
## Installing required python packages
Since Deployment Manager uses Python and will be running locally on your
machine, you will first need to make sure the necessary Python packages are
installed. This assumes that you have already installed the pip package
management system on your machine.
```
sudo pip install -r expandybird/requirements.txt
```
## Building and installing the binaries
Next, you're going to build and install the Deployment Manager binaries. You can
do this by running make in the repository root.
```
make
```
## Bootstrapping Deployment Manager
Now, you're ready to bootstrap Deployment Manager into the cluster.
### Start Deployment Manager on localhost
First, start the three Deployment Manager binaries on localhost using the supplied
bootstrap script.
```
./examples/bootstrap/bootstrap.sh
```
The script starts the following binaries:
* manager (frontend service) running on port 8080
* expandybird (expands templates) running on port 8081
* resourcifier (reifies primitive Kubernetes resources) running on port 8082
It also starts kubectl proxy on port 8001.
### Deploy Deployment Manager into your cluster
Next, use the Deployment Manager running on localhost to deploy itself onto the
cluster using the supplied command line tool and template.
```
client --name test --service=http://localhost:8080 examples/bootstrap/bootstrap.yaml
```
You should now have Deployment Manager running on your cluster, and it should be
visible using kubectl (kubectl get pod,rc,service).

@ -0,0 +1,145 @@
######################################################################
# Copyright 2015 The Kubernetes Authors All rights reserved.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# http://www.apache.org/licenses/LICENSE-2.0
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
######################################################################
---
apiVersion: v1
kind: Service
metadata:
labels:
app: dm
name: expandybird-service
name: expandybird-service
namespace: default
spec:
ports:
- name: expandybird
port: 8081
targetPort: 8080
selector:
app: dm
name: expandybird
---
apiVersion: v1
kind: ReplicationController
metadata:
labels:
app: dm
name: expandybird-rc
name: expandybird-rc
namespace: default
spec:
replicas: 2
selector:
app: dm
name: expandybird
template:
metadata:
labels:
app: dm
name: expandybird
spec:
containers:
- env: []
image: gcr.io/dm-k8s-testing/expandybird:latest
name: expandybird
ports:
- containerPort: 8080
name: expandybird
---
apiVersion: v1
kind: Service
metadata:
labels:
app: dm
name: resourcifier-service
name: resourcifier-service
namespace: default
spec:
ports:
- name: resourcifier
port: 8082
targetPort: 8080
selector:
app: dm
name: resourcifier
---
apiVersion: v1
kind: ReplicationController
metadata:
labels:
app: dm
name: resourcifier-rc
name: resourcifier-rc
namespace: default
spec:
replicas: 2
selector:
app: dm
name: resourcifier
template:
metadata:
labels:
app: dm
name: resourcifier
spec:
containers:
- env: []
image: gcr.io/dm-k8s-testing/resourcifier:latest
name: resourcifier
ports:
- containerPort: 8080
name: resourcifier
---
apiVersion: v1
kind: Service
metadata:
labels:
app: dm
name: manager-service
name: manager-service
namespace: default
spec:
ports:
- name: manager
port: 8080
targetPort: 8080
selector:
app: dm
name: manager
---
apiVersion: v1
kind: ReplicationController
metadata:
labels:
app: dm
name: manager-rc
name: manager-rc
namespace: default
spec:
replicas: 1
selector:
app: dm
name: manager
template:
metadata:
labels:
app: dm
name: manager
spec:
containers:
- env: []
image: gcr.io/dm-k8s-testing/manager:latest
name: manager
ports:
- containerPort: 8080
name: manager
Loading…
Cancel
Save