Myujth

i have been working on creating EC2 instance from java SDK and I have been doing it successfully but now there is a thought that let us assume the case when AWS servers are down although I know it violates the concept of cloud still as if we want to handle this case I am not able to understand how can I create the scenario and handle this case?

public static void main(String args) {

 // Set up the amazon ec2 client
 AmazonEC2 ec2Client = AmazonEC2ClientBuilder.standard()
 .withCredentials(new AWSStaticCredentialsProvider(AWS_CREDENTIALS))
 .withRegion(Regions.US_EAST_1)
 .build();

 // Launch an Amazon EC2 Instance
 RunInstancesRequest runInstancesRequest = new RunInstancesRequest().withImageId("ami-777777")
 .withInstanceType("t2.micro") // https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instance-types.html
 .withMinCount(1)
 .withMaxCount(1)
 .withNetworkInterfaces(new InstanceNetworkInterfaceSpecification()
 .withAssociatePublicIpAddress(true)
 .withDeviceIndex(0)
 .withSubnetId("subnet-777777")
 .withGroups("sg-777777"));

 RunInstancesResult runInstancesResult = ec2Client.runInstances(runInstancesRequest);

 Instance instance = runInstancesResult.getReservation().getInstances().get(0);
 String instanceId = instance.getInstanceId();
 System.out.println("EC2 Instance Id: " + instanceId);

 // Setting up the tags for the instance
 CreateTagsRequest createTagsRequest = new CreateTagsRequest()
 .withResources(instance.getInstanceId())
 .withTags(new Tag("Name", "codeflex-ec2"));
 ec2Client.createTags(createTagsRequest);

 // Starting the Instance
 StartInstancesRequest startInstancesRequest = new StartInstancesRequest().withInstanceIds(instanceId);

There is no difference in the downtime in your datacenter and in AWS datacenter. Its design level scenario where you will need to have additional server on standby to handle your traffic in another AZ or DR (Disaster recovery) site in another region.

1. 
   

   You can Stop / terminate instance in AWS region to create the downtime scenario. Please use Stop Ec2 instance to stop the instance manually.

   
   
   

   public StopInstancesResult stopInstances(StopInstancesRequest request) 
   

   
   



2. To handle the availability of your application you need to setup application load balancer with auto-scaling group enabled. The load balancer checks for the server after every configured time (e.g. 5 second 2 min, you can choose any value) and route traffic to another instance if your current EC2 instance is unhealthy.



3. Since all AZ are in same availability zone in same region, if your region is down then you need to have your DR site ready with all your latest code deployed. This site can be used in case of region level downtime.