spinnaker/README.adoc

175 строки
7.9 KiB
Plaintext
Исходник Обычный вид История

2015-11-17 00:00:02 +03:00
:doctype: book
2015-11-13 10:08:12 +03:00
2015-11-17 00:11:19 +03:00
image:http://join.spinnaker.io/badge.svg[Slack Status,link=http://join.spinnaker.io]
2015-11-17 00:00:02 +03:00
= Getting Started with Spinnaker
These instructions cover pulling Spinnaker from source and setting up to run locally against Amazon Web Services and/or Google Cloud Platform accounts. ( If you would prefer to try a container based solution, see https://github.com/spinnaker/spinnaker/tree/master/experimental/docker-compose[docker-compose installation] )
We will clone into `$SPINNAKER_HOME` and create that as our working directory, including this repo for configuration scripts, as well as the various
service repos.
2015-11-17 00:00:02 +03:00
*Note* If you are going to create a Virtual Machine in Amazon EC2 or
Google Compute Engine for your development, then a reasonable machine
type is m4.2xlarge (Amazon) or n1-standard-8 (Google). If using Google,
you will need to add "Read Write" Compute access scope when creating the
instance, and may also wish to add "Full" Storage scope to later write
releases to Google Cloud Storage buckets. The Amazon credentials are
2015-11-17 00:00:02 +03:00
discussed below in link:#configure-your-aws-account[Configure your AWS Account].
2015-11-11 00:43:29 +03:00
These scripts are tested on:
* Ubuntu 14.04 LTS
* Mac OS X 10.11
2015-11-17 00:00:02 +03:00
== Get the bootstrap / configuration repo
2015-11-17 00:00:02 +03:00
[source,bash]
----
#export SPINNAKER_HOME=/path/to/your/Spinnaker/workspace
mkdir -p $SPINNAKER_HOME
cd $SPINNAKER_HOME
git clone git@github.com:spinnaker/spinnaker.git
2015-11-17 00:00:02 +03:00
----
2015-11-17 00:00:02 +03:00
== Configure your environment
2015-11-10 01:24:51 +03:00
The Spinnaker platform has a few prerequisites, which are installed as a part of this configuration process. They are:
2015-11-17 00:00:02 +03:00
* https://www.oracle.com/java/index.html[JDK8]
* http://redis.io/[Redis]
* http://cassandra.apache.org/[Cassandra]
* https://www.packer.io/[Packer]
2015-11-17 00:00:02 +03:00
=== MacOSX
2015-11-17 00:00:02 +03:00
You need to have http://brew.sh/[homebrew] installed and ensure your version of git is above 2.0.
2015-11-17 00:00:02 +03:00
[source,bash]
----
brew install redis cassandra brew-cask packer
brew cask install java
cd $SPINNAKER_HOME
mkdir build
cd build
../spinnaker/dev/refresh_source.sh --pull_origin --use_ssh --github_user default
2015-11-17 00:00:02 +03:00
----
2015-11-17 00:00:02 +03:00
=== Debian-linux
2015-11-17 00:00:02 +03:00
[source,bash]
----
cd $SPINNAKER_HOME
spinnaker/dev/install_development.sh --package_manager
2015-11-04 04:17:18 +03:00
spinnaker/dev/bootstrap_dev.sh
2015-11-17 00:00:02 +03:00
----
The `--package_manager` argument requests the Debian Package Manager be used
to the greatest extent possible. This permits adding new source repositories
where package dependencies are distributed from non-standard repositories
(e.g. cassandra). The alternative is `--nopackage_manager` to use the
Package Manager for standard dependencies but explicitly install packages
that come from other sources. `--nopackage_manager` is intended for
installations where IT policies discourage or forbid adding additional
source locations.
The `bootstrap_dev.sh` script will ask to install additional components.
`packer` is only needed if you plan on building VM images. `gcloud`
is only needed to write releases to Google Cloud Storage, but is convenient
to have if you plan on using or accessing Google Cloud Platform resources
from your development environment. These could be installed at a system level
rather than user level, but the default install requires updating your path
so is performed here.
2015-11-17 00:00:02 +03:00
== Configure Spinnaker
We will create a directory for Spinnaker configuration overrides, copy the default configuration template there, and edit to select
2015-10-30 04:57:19 +03:00
the appropriate cloud provider(s).
2015-11-17 00:00:02 +03:00
[source,bash]
----
cd $SPINNAKER_HOME
mkdir -p $HOME/.spinnaker
cp spinnaker/config/default-spinnaker-local.yml $HOME/.spinnaker/spinnaker-local.yml
chmod 600 $HOME/.spinnaker/spinnaker-local.yml
2015-11-17 00:00:02 +03:00
----
Edit `$HOME/.spinnaker/spinnaker-local.yml` and set the enabled option for the cloud provider(s) of your choice.
2015-11-17 00:00:02 +03:00
== Configure your AWS Account
If you enabled AWS for Spinnaker, there are some requirements for the AWS account:
2015-10-30 01:36:24 +03:00
Decide which region you want Spinnaker to index. In `$HOME/.spinnaker/spinnaker-local.yml` fill in that value in providers.aws.defaultRegion. (The default is us-east-1).
2015-11-17 00:00:02 +03:00
Sign into the https://console.aws.amazon.com/[AWS console], and select the region Spinnaker will manage.
. Click on Networking > VPC
. Name your vpc (edit the name tag, and give it a value with no spaces or dots in the name) (e.g. defaultvpc)
. Name your subnets (edit the name tag and name following the pattern vpcName.internal.<availabilityZone>)
** e.g. defaultvpc.internal.us-east-1a, defaultvpc.internal.us-east-1b, defaultvpc-internal.us-east-1c
+
image:https://cloud.githubusercontent.com/assets/74310/11158314/5dcee17e-8a0c-11e5-9f2b-4d3965b4aba7.png[width="355"]
+
* Create an EC2 role called BaseIAMRole
** Console > Identity & Access Management > Roles > Create New Role. Select Amazon EC2.
** You don't have to apply any policies to this role. EC2 instances launched with Spinnaker will have this role associated.
* Create an EC2 keyPair for connecting to your instances.
** Console > EC2 > Key Pairs > Create Key Pair. Name the key pair `my-aws-account-keypair` (`my-aws-account` should match the aws.primaryCredentials.name in `$HOME/.spinnaker/spinnaker-local.yml`, so if the name is awsprod, the name for the key pair should be awsprod-keyspace).
* Create AWS credentials for Spinnaker
** Console > Identity & Access Management > Users > Create New Users. Enter a username.
** Create an Access Key for the user. Save the access key and secret key into `~/.aws/credentials` as shown http://docs.aws.amazon.com/cli/latest/userguide/cli-chap-getting-started.html#cli-config-files[here]. Alternatively, add the keys to `$HOME/.spinnaker/spinnaker-local.yml`
** Edit the users Permissions.
** Attach a Policy to the user granting PowerUserAccess.
+
image:https://cloud.githubusercontent.com/assets/74310/11158315/5dd056f8-8a0c-11e5-8a85-bf78f6d8db94.png[width="591"]
+
** Create an inline policy for IAM granting PassRole on the resource '*' <img width="500" src="https://cloud.githubusercontent.com/assets/74310/11158316/5dd1e86a-8a0c-11e5-8dc9-c06a6ec616cf.png">
== Configure your Google Cloud Platform Account
If you enabled Google for Spinnaker, perform the following steps for your project in the Google Developer's Console:
2015-11-17 00:00:02 +03:00
. Enable APIs (click 'Enable API' button for each):
* https://console.developers.google.com/project/_/apiui/apiview/compute_component/overview[Compute Engine]
* https://console.developers.google.com/project/_/apiui/apiview/autoscaler/overview[Compute Engine Autoscaler]
. Add and Obtain Credentials:
* Navigate to https://console.developers.google.com/project/_/apiui/credential[Credentials]
* New credentials &gt; Service account key
* Select… &gt; New service account
* Provide a Name and click Create
* `chmod 400` the file that downloads
. Add GCP credentials in `$HOME/.spinnaker/spinnaker-local.yml`:
* Set project ID for `provider.google.primaryCredentials.project`
* Set full absolute path of downloaded file for `providers.google.primaryCredentials.jsonPath`
== Start Spinnaker Services
( If you're running on Mac OS, please make sure that your redis server and cassandra server is up and running. Make sure that cqlsh is accessible in your path and you can connect to cassandra by calling `cqlsh` in your terminal ).
2015-11-17 00:00:02 +03:00
[source,bash]
----
cd $SPINNAKER_HOME/build
../spinnaker/dev/run_dev.sh [service]
2015-11-17 00:00:02 +03:00
----
If a service is provided, then just that one service will be started.
If no service is provided, then all the services will be started
(including redis and cassandra unless they are specified with a remote host).
If a service is already running (even if not yet available) then it will
not be restarted.
2015-11-17 00:00:02 +03:00
*Note* `run_dev.sh` might get stuck waiting on a service to start. Hitting CTRL-C just stops the waiting on service it doesn't terminate the services. If it seems stuck
stop and restart run_dev.sh.
2015-11-17 00:00:02 +03:00
== Stop Spinnaker Services
[source,bash]
----
cd $SPINNAKER_HOME/build
../spinnaker/dev/stop_dev.sh [service]
2015-11-17 00:00:02 +03:00
----
If a service is provided, then just that one service will be stopped.
If no service is provided then all the spinnaker services will be stopped.
2015-11-17 00:11:19 +03:00
Cassandra and redis are not affected by stop_dev.sh