◀️ Hazelcast on AWS EC2 🔗 SnappyData CLASSPATH ▶️ 🔗 Home ▶️

This article describes how to reactivate VM workspaces that have been detached from AWS EC2 VMs.

You have created a VM workspace with EC2 instances. After you were done running Hazelcast cluster(s) on the EC2 instances from the VM workspace, you terminated the EC2 instances without removing the workspace.

A few days later, you return to your work and are in need of reinstating the same workspace environment with new EC2 instances.

VM workspace activation steps involve updating two (2) workspace and cluster configuration files with the public IP addresses of new EC2 instances.

• Workspace: vmenv.sh
• Cluster: cluster.properties

Let's walk through the workspace reactivation steps using the same example workspace described in the following article:

Hazelcast on AWS EC2

First, launch the desired number of EC2 instances from the EC2 Dashboard. For our example, we launched five (5) instances as shown below.

Gather the public IP addresses of the EC2 instances and list them in the workspace vmenv.sh file.

switch_workspace ws-aws-hazelcast
vi vmenv.sh

Set the VM_HOSTS property with the public IP addresses and make sure VM_USER is set to the correct user name.

VM_HOSTS="3.18.113.154,3.21.44.203,18.218.42.247,18.222.225.117,18.188.216.237"
VM_USER="ec2-user"

Update the etc/cluster.properties file with the new EC2 IP addresses. For our example, the VM cluster name is myhz.

switch_cluster myhz
vi etc/cluster.properties

Set the following VM properties.

# Set Management Center host
mc.host=3.18.113.154

# Set member hosts
vm.hosts=18.218.42.247,18.222.225.117,18.188.216.237,3.18.113.154

❗ If you have reinstated the workspace with an EC2 instance type that is different from the previous instance type then you may also need to change resource properties such as heap.min and heap.max.

Run vm_sync to synchronize the workspace.

vm_sync

The above command reports the following:

Deploying padogrid_0.9.4-SNAPSHOT to 3.18.113.154...
Deploying padogrid_0.9.4-SNAPSHOT to 3.21.44.203...
Deploying padogrid_0.9.4-SNAPSHOT to 18.218.42.247...
Deploying padogrid_0.9.4-SNAPSHOT to 18.222.225.117...
Deploying padogrid_0.9.4-SNAPSHOT to 18.188.216.237...

Workspace sync: ws-aws-hazelcast
   Synchronizing 3.18.113.154...
   Synchronizing 3.21.44.203...
   Synchronizing 18.218.42.247...
   Synchronizing 18.222.225.117...
   Synchronizing 18.188.216.237...

Updating remote (VM) '.bashrc' if needed...
...
Workspace sync complete.

vm_sync will display warning messages similar to the output shown above since the new EC2 instances do not have the required software installed. Download the required software and install them by running the vm_install command as shown below.

vm_install -product ~/Downloads/jdk-8u333-linux-x64.tar.gz
vm_install -product ~/Downloads/hazelcast-enterprise-5.1.3.tar.gz
vm_install -product ~/Downloads/hazelcast-management-center-5.1.4.tar.gz

Start the cluster.

start_cluster

For our example, we have already created the perf_test app. Because we have configured the EC2 discovery service, the client app requires no changes. Let's run the app.

cd_app perf_test; cd bin_sh
./test_ingestion -run

Reactivating a VM workspace with new EC2 instances is just a matter of updating the workspace and cluster configuration files with the new instance IP addresses. Note that you can also upgrade your software components as part of reactivation.

◀️ Hazelcast on AWS EC2 🔗 SnappyData CLASSPATH ▶️ 🔗 Home ▶️