-
Notifications
You must be signed in to change notification settings - Fork 101
Running the Images
Once you have the images, get ready to run them. As documented in the launchcommands.txt, here are the steps and explanations:
The below instructions name the running containers so clients can access them via Docker linking. This means that when they exit you won't be able to re-run the container with the same name. If you stop a container, please docker rm the container.
The below assumes you have direct access to the container network. Depending on your environment, you may want to use docker port mapping in order to indirectly access that network. Additionally if you are running each container on a unique machine, you may want to use the host network directly.
Instead of using existing images for Java and Tomcat by other Docker community members, we have chosen to replicate these within our own Netflix trusted builds, so we have complete control over the versions of Java and Tomcat. Also, by creating our own we are certain about their contents making the builds even more trustable. You should not need to run these images by themselves as they are just base images for the following NetflixOSS images.
docker run -d --name asgard -v `pwd`/Config.groovy.mine:/root/.asgard/Config.groovy netflixoss/asgard:1.5.1
Before running this command, you must create a Config.groovy.mine by customizing the provided Config.groovy. You will see that you need to provide your Amazon EC2 AccountIDNumber, AccountAccessKeyId, AccountAccessKeySecretKey updating four fields. You can add these to Config.groovy using static keys, but we highly suggest using expiring and IAM role controller keys. By running this command you will be adding your Config.groovy into where Asgard looks for configuration.
Once started, you can get the IP address by running ./dinspect.sh that is in this project. Otherwise you can run docker inspect CONTAINERNAME to find individual ip addresses
Once you have the correct URL, load the following url:
http://ASGARDIPADDRESS:8080/
docker run -d --name exhibitor netflixoss/exhibitor:1.5.2
Once started, obtain the IP address using dinspect.sh and then load the following url:
http://EXHIBITORIPADDRESS:8080/exhibitor/v1/ui/index.html
docker run -d --name eureka -d netflixoss/eureka:1.3.1
This will start the Eureka server with a name of "eureka", which means anyone who links to the container will be able to address the server via the hostname eureka.
Once started, obtain the IP address using dinspect.sh and then load the following url:
http://EUREKAIPADDRESS:8080/eureka/
docker run -d --name hello-netflix-oss --link eureka:eureka netflixoss/hello-netflix-oss:1.0.28
This will start a sample Karyon server. The container will be linked to the Eureka container instance, so a minute or so after startup, you should see it show up in the Eureka console.
Once started, obtain the IP address using dinspect.sh and then load the following urls (the first is the application, the second is the embedded karyon admin services console):
http://HELLONETFLIXOSSIPADDRESS:8080/
http://HELLONETFLIXOSSIPADDRESS:8077/
docker run -e "origin.zuul.client.DeploymentContextBasedVipAddresses=HELLO-NETFLIX-OSS" -d --name zuul --link eureka:eureka netflixoss/zuul:1.0.28
This will start a Zuul server that will front end the sample application. The container will be linked to the Eureka container instance, so a minute or so after startup, you should see it show up in the Eureka console. Also, given we overrode the origin by providing a Eureka based vip address, all requests to the Zuul server will be proxied to any instances registered in Eureka for the Hello Netflix OSS sample application.
Once started, obtain the IP address using dinspect.sh and then load the following url:
http://ZUULIPADDRESS:8080/
The configuration of security monkey requires additional Amazon configuration. Also, it is composed of multiple images. Follow the guide to running Security Monkey on the wiki.