Aurora promotes an existing replica in another AZ to a new primary instance, so nothing needs to be done

You must manually create one or more new DB instances in another AZ

Aurora automatically creates a new primary instance in the same AZ

For a cluster using single-master replication, Aurora can create up to 15 read-only Aurora Replicas to serve requests from users

Create the AMI with No reboot option enabled

Create the AMI by disabling the No reboot option

Create an EBS-backed AMI for application consistency

Create the AMI with Delete on termination enabled

Configure Amazon CloudWatch Events to pick data from Amazon Inspector

Configure Amazon CloudTrail to generate logs for the service limits. CloudTrail and CloudWatch are integrated and hence alarm can be generated for customized service checks

Use CloudWatch ServiceLens to monitor data on service limits of various AWS services

Monitor Trusted Advisor service check results with Amazon CloudWatch Events

Configure the RDS to be a multi Availability Zone (AZ) deployment

Configure your JVM with a TTL value of no more than 60 seconds, to help you re-establish the connection to your database, in case of failure

Scale up your DB instance when you are approaching storage capacity limits

Configure automated backups for the RDS instance, to retrieve data and instance status, if needed after a failure

Use the AWS-UpdateCloudFormationStackWithApproval document to update impaired instances

Automatic recovery of impaired instances is not possible currently

Use the AWS-UpdateWindowsAmi document to recover impaired instances

Use the AWSSupport-ExecuteEC2Rescue document to recover impaired instances

Use nested stacks that will retain the configuration in the parent configuration even if the child configuration is lost or cannot be used

Use Stack policies to protect critical stack resources from unintentional updates

Use parameter constraints to specify the Identities that can update the Stack

Use revision controls to protect critical stack resources from unintentional updates

The EBS volume could be encrypted and the custom KMS key used to encrypt the snapshot is missing. The custom KMS key needs to be added to the volume configuration

Check that the device name you specified when you attempted to attach the EBS volume isn't already in use. Attempt to attach the volume to the instance, again, but use a different device name

The attaching status indicates that the underlying hardware related to your EBS volume has failed. This issue cannot be fixed. Raise a service request on AWS and request for a new volume. You are not charged for volumes that are in error state

Each EBS volume receives an initial I/O credit balance, an error in accumulating the credit balance can stop the volume from attaching properly to the instance. Restart the instance to fix the error

Create a Role in production account, that defines the development account as a trusted entity and specify a permissions policy that allows trusted users to update the bucket. Then, modify the IAM group policy in development account, so that testers are denied access to the newly created role. Developers can use the newly created role to access the live S3 buckets in production environment

Create a Role in development account, that defines the production account as a trusted entity and specify a permissions policy that allows trusted users to update the bucket. Then, modify the IAM group policy in development account, so that testers are denied access to the newly created role. Developers can use the newly created role to access the live S3 buckets in production environment

Use Inline policies to be sure that the permissions in a policy are not inadvertently assigned to an identity other than the one they're intended for

Create a Role in Production account, that defines the Development account as a trusted entity and specify a permissions policy that allows trusted users to update the bucket. Developers can use the newly created role to access the live S3 buckets in production environment

To safely write to tables on a read replica, create indexes on the table after setting the read_only parameter to 0

Writing to tables on a read replica can break the replication

Though read replicas can work on both transactional and nontransactional storage engines, nontransactional engines are error-prone because of the way memory is managed on these engines

Statements containing non-deterministic functions like SYSDATE() should be predefined in the configuration to successfully create the read replica

If the value for the max_allowed_packet parameter for a read replica is less than the max_allowed_packet parameter for the source DB instance, replica errors occur

Use Amazon S3 bucket owner condition to restrict access to unintended users

Use Amazon S3 replication to replicate critical objects to avoid losing them from unintended deletes

Enable S3 versioning on all the buckets used in the project

Enable S3 Object Lock to lock all the objects that are used in the project