Network requirements

To prepare your network for running Elastic Disaster Recovery, you need to set the following connectivity settings:

Note

All communication is encrypted with TLS.

Communication over TCP Port 443:

Communication over TCP Port 1500:

• Between the Source Machines and the staging area Subnet

Communication over TCP port 443

Add the following IP addresses and URLs to your firewall:

The Elastic Disaster Recovery AWS Region-specific Console address:

• (drs.<region>.amazonaws.com example: drs.eu-west-1.amazonaws.com)

Amazon S3 service URLs (required for downloading AWS Elastic Disaster Recovery software)

• The AWS Replication Agent installer should have access to the S3 bucket URL of the AWS Region you are using with Elastic Disaster Recovery.

• The staging area subnet should have access to S3.

• The following S3 buckets should be allowed:

  
                          https://aws-drs-clients-<REGION>.s3.<REGION>.amazonaws.com/
                          https://aws-drs-clients-hashes-<REGION>.s3.<REGION>.amazonaws.com/
                          https://aws-drs-internal-<REGION>.s3.<REGION>.amazonaws.com/
                          https://aws-drs-internal-hashes-<REGION>.s3.<REGION>.amazonaws.com/
                          https://aws-elastic-disaster-recovery-<REGION>.s3.<REGION>.amazonaws.com/
                          https://aws-elastic-disaster-recovery-hashes-<REGION>.s3.<REGION>.amazonaws.com/
                      

Note

• Agent installation and replication server components require Amazon S3 bucket for service functionality.

• Ensure the relevant VPC endpoint policy includes access to all the required Amazon S3 buckets.

When using an S3 VPC Endpoint, you must provide sufficient permissions for service functionality. See example policy for replicating to us-east-1:

{
	"Version": "2008-10-17",
	"Statement": [
		{
			"Effect": "Allow",
			"Principal": {
				"AWS": "*"
			},
			"Action": "s3:GetObject",
			"Resource": [
				"arn:aws:s3:::aws-drs-clients-us-east-1/*",
				"arn:aws:s3:::aws-drs-clients-hashes-us-east-1/*",
				"arn:aws:s3:::aws-drs-internal-us-east-1/*",
				"arn:aws:s3:::aws-drs-internal-hashes-us-east-1/*",
				"arn:aws:s3:::aws-elastic-disaster-recovery-us-east-1/*",
				"arn:aws:s3:::aws-elastic-disaster-recovery-hashes-us-east-1/*"
			]
		}
	]
} 

AWS specific

The staging area subnet requires outbound access to the Amazon EC2 endpoint of its AWS Region.

TCP port 443 is used for two communication routes:

1. Between the source servers and Elastic Disaster Recovery.

2. Between the staging area subnet and AWS Elastic Disaster Recovery.

Communication between the source servers and Elastic Disaster Recovery over TCP port 443

Each source server that is added to AWS Elastic Disaster Recovery (AWS DRS) must continuously communicate with AWS DRS (DRS.<region>.amazonaws.com) over TCP port 443.

The following are the main operations performed through TCP port 443:

• Downloading the AWS Replication Agent on the source servers.

• Upgrading installed Agents.

• Connecting the source servers to the AWS DRS Console and displaying their replication status.

• Monitoring the source servers for internal troubleshooting and the use of resource consumption metrics (such as CPU, RAM).

• Reporting source server-related events (for example, a removal of resizing of a disk).

• Transmit source server-related information to the AWS DRS Console (including hardware information, running services, installed applications and packages, and more).

• Preparing the source servers for drill or recovery.

Important

Make sure that your corporate firewall allows connections over TCP port 443.

Solving communication problems over TCP port 443 between the source servers and AWS Elastic Disaster Recovery

If there is no connection between your source servers and AWS Elastic Disaster Recovery, make sure that your corporate firewall facilitates connectivity from the source servers to AWS Elastic Disaster Recovery over TCP Port 443. If the connectivity is blocked, activate it.

Enabling Windows Firewall for TCP port 443 connectivity

Important

The information provided in this section is for general security and firewall guidance only. The information is provided on "AS IS" basis, with no guarantee of completeness, accuracy or timeliness, and without warranty or representations of any kind, expressed or implied. In no event will AWS Elastic Disaster Recovery and/or its subsidiaries and/or their employees or service providers be liable to you or anyone else for any decision made or action taken in reliance on the information provided here or for any direct, indirect, consequential, special or similar damages (including any kind of loss), even if advised of the possibility of such damages. AWS Elastic Disaster Recovery is not responsible for the update, validation or support of security and firewall information.

Note

Enabling Windows Firewall for TCP port 443 connectivity will allow your servers to achieve outbound connectivity. You may still need to adjust other external components, such as firewall blocking or incorrect routes, in order to achieve full connectivity.

Note

These instructions are intended for the default OS firewall. You will need to consult the documentation of any third-party local firewall you use to learn how to enable TCP port 443 connectivity.

1. On the source server, open the Windows Firewall console.

2. On the console, select the Outbound Rules option from the tree.

   

3. On the Outbound Rulestable, select the rule that relates to the connectivity to Remote Port - 443. Check if the Enabled status is Yes.

   

4. If the Enabled status of the rule is No, right-click it and select Enable Rule from the pop-up menu.

   

Enabling Linux Firewall for TCP port 443 connectivity

1. Enter the following command to add the required Firewall rule:

   sudo iptables -A OUTPUT -p tcp --dport 443 -j ACCEPT

2. To verify the creation of the Firewall rule, enter the following commands:

   sudo iptables -L

   Chain INPUT (policy ACCEPT)

   target prot opt source destination

   Chain FORWARD (policy ACCEPT)

   target prot opt source destination

   Chain OUTPUT (policy ACCEPT)

   target prot opt source destination

   ACCEPT tcp -- anywhere anywhere tcp dpt:443

Communication between the staging area subnet and AWS Elastic Disaster Recovery over TCP port 443

The replication servers in the staging area subnet must continuously communicate with Elastic Disaster Recovery over TCP port 443. The main operations that are performed through this route are:

• Downloading the replication software by the replication servers.

• Connecting the replication servers to AWS Elastic Disaster Recovery, and displaying their replication status.

• Monitoring the replication servers for internal troubleshooting use and resource consumption metrics (such as CPU, RAM).

• Reporting replication-related events.

Note

The staging area subnet requires S3 access.

Configuring communication over TCP port 443 between the staging area subnet and AWS Elastic Disaster Recovery

You can establish communication between the staging area subnet and AWS Elastic Disaster Recovery over TCP port 443 directly.

There are two ways to establish direct connectivity to the Internet for the VPC of the staging area, as described in the VPC FAQ.

1. Public IP address + Internet gateway

2. Private IP address + NAT instance

Communication between the source servers and the Staging Area Subnet over TCP port 1500

Each source server with an installed AWS Replication Agent continuously communicates with the AWS Elastic Disaster Recovery replication servers in the staging area subnet over TCP port 1500. TCP port 1500 is needed for the transfer of replicated data from the source servers to the staging area subnet.

The replicated data is encrypted and compressed when transferred over TCP port 1500. Prior to being moved into the Staging Area Subnet, the data is encrypted on the source infrastructure. The data is decrypted after it arrives at the staging area subnet and before it is written to the volumes.

TCP port 1500 is primarily used for the replication server data replication stream.

Elastic Disaster Recovery uses TLS 1.2 end to end from the agent installed on the source server to the replication server. Each replication server gets assigned a specific TLS server certificate, which is distributed to the corresponding agent and validated against on the agent side.

Establishing communication over TCP port 1500

Important

To allow traffic over TCP port 1500, make sure that your corporate firewall enables this connectivity.

Required bandwidth between the source servers and the staging area subnet

Replicated data is transferred from the source servers to the staging area over the network. For replication to succeed, your average network bandwidth must be higher than the write rate on the source servers. If you attempt to conduct a replication of a write intensive source server under low bandwidth conditions, it will likely lag.