NEC Express Cluster 3.3

Product: NEC Express Cluster 3.3 (Product landing page)

Failover Services: HANA Scale Up data bases on Red Hat Linux

Licensing: NEC licenses depending on the services

Status: released, supported

The NEC Cluster relies on the SAP HANA system replication. It works across AWS availability zones within a region.

The NEC cluster uses AWS Overlay IP addresses which support a fast failover. The NEC Cluster will not shut down a node which isn't providing anymore the service. It will fail over to the standby node.

More Resources

AWS Specific Configuration Details

Be aware that the NEC cluster will change the network topology. The privileges required for these operations allow to change the AWS network topology in an account. Verify and test all entries very carefully. Limit access to user working on the NEC Express cluster nodes to the required minimum.

Required Routing Entries

The NEC Cluster will typically operate in a single VPC. The cluster nodes are typically located in different availability zones for increased availability. Therefore thew will have their primary IP addresses in different subnets.

The AWS overlay IP addresses are based on a concept which allows to create routing entries which point traffic to an IP address (NEC cluster node). The NEC Express Cluster will change these routing entries when needed. It will however not create the routing entries. The initial creation of the routing entries needs to happen manually. The same routing entry will have to be created in all routing tables of the given VPC. 

The AWS VPC console can be used to add this entry. The AWS Command Line Interface offers the following command as well:

ec2addrt ROUTE_TABLE -r CIDR -i INSTANCE

The user will have to pick an arbitrary AWS instance id from a cluster node as option -i. The NEC Express cluster will then update this entry as needed.

The NEC cluster will only operate in a  correct way if the routing entry in all routing tables of the VPC have been created!

AWS Instance Configuration for Cluster Nodes

The AWS cluster nodes will have to be able to communicate through a second IP address. The document IP Failover with Overlay IP Addresses on this site describe how to disable the source/destination check for AWS instances and how to host a second IP address on the same Linux system.

IAM Policies: NEC-HA-Policy

The cluster nodes will require the following privileges to operate:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "Stmt1424870324000",
            "Effect": "Allow",
            "Action": [
                "ec2:DescribeInstances",
                "ec2:DescribeInstanceAttribute",
                "ec2:DescribeTags",
                "ec2:DescribeVpcs",
                "ec2:DescribeNetworkInterfaces",
                "ec2:DescribeAvailabilityZones"
            ],
            "Resource": "*"
        },
        {
            "Sid": "Stmt1424860166260",
            "Action": [
                "ec2:CreateRoute",
                "ec2:DeleteRoute",
                "ec2:DescribeRouteTables",
                "ec2:ReplaceRoute"
            ],
            "Effect": "Allow",
            "Resource": "*"
        }
    ]
}