Deploying TIBCO ComputeDB from AWS Marketplace

You can quickly set up a simple TIBCO ComputeDB cluster on AWS using its Amazon Machine Image (AMI) that is available in the AWS Marketplace. The AMI comes with the sample CloudFormation templates.

Note

The TIBCO ComputeDB 1.2.0 AMI will be available shortly.

Note

The IAM user or the role with which you create the stack and other AWS resources as part of the stack must have adequate AWS permissions.

To deploy TIBCO ComputeDB from AWS Marketplace, do the following:

  1. Go to the TIBCO ComputeDB's page on AWS Marketplace.
  2. Select the region where you want to launch the instances from the dropdown list.

  3. Select an appropriate Fulfillment option from the following:

    • 64-bit (x86) Amazon Machine Image (AMI)
    • TIBCO ComputeDB - Developer Configuration
    • TIBCO ComputeDB - Secured Configuration

    The first option only launches the EC2 instance from the AMI, but it does not start the TIBCO ComputeDB cluster. You must configure and launch the cluster from the instance manually.

    The next two options allow you to launch the cluster using AWS CloudFormation via sample templates, which are pre-loaded on the AWS CloudFormation's Create stack page.

    TIBCO ComputeDB - Developer Configuration is the recommended option in most cases. This option lets you launch the TIBCO ComputeDB cluster with no cluster-level security enabled. This is preferred option if you want to quickly try out the TIBCO ComputeDB product.

    TIBCO ComputeDB - Secured Configuration option lets you launch the TIBCO ComputeDB cluster with security enabled. As TIBCO ComputeDB relies on an externally managed LDAP server for authentication, a test-only LDAP setup is provided in the AMI.
    Therefore, in this Fulfillment option, the test-only LDAP server is launched along with the TIBCO ComputeDB cluster. This LDAP setup has one admin user (snappyadmin) and one non-admin user (snappyuser).
    The password for snappyadmin is available in the /opt/snappydata/props.txt file, which is located in the EC2 instance. The password for snappyuser is the ID of the EC2 instance that will be launched.

    Note

    You must provision your own LDAP server to work with the TIBCO ComputeDB cluster in production and must not rely on the test-only LDAP setup provided in the AMI.

  4. Click Continue to Subscribe.

  5. Go through the Terms and Conditions and click Continue to Configure.
  6. Select CloudFormation Template as the Fulfillment option and select one of the two templates explained above. Then click Continue to Launch.
  7. Select Launch CloudFormation as the action and then click Launch. Alternatively, if you selected Amazon Machine Image as the Fulfillment option on the previous page, select Launch through EC2 as the action on this page. Click Launch and then perform these steps to complete the setup.
    The AWS CloudFormation’s Create stack page is displayed. The Amazon S3 URL of the template is already populated here.
  8. Click Next. The Specify stack details page is displayed.

  9. Enter the Stack name and the configuration details of the cluster using the following parameters and then click Next:

    Parameters Description
    ClientCIDR The allowed CIDR IP range for client (JDBC/ODBC) connections. To access the cluster from your personal laptop, first, find your public IP address. You can go to whatismyip.com to get your IPv4 address and then append /32 to it. For example, 123.201.112.0/32.
    EBSVolumeSize Specify the Amazon EBS volume size (in GB), which will be attached to the EC2 instance that gets launched. It can be any number between 16 and 1024, depending on your data volume. The TIBCO ComputeDB catalog metadata and its database are managed in this volume.
    InstanceTypeName Select one instance type from the dropdown list as per your workload requirement. For more details, refer https://aws.amazon.com/ec2/pricing/on-demand.
    KeyPairName Select one from the dropdown list whose key file (.pem) is available with you. This is needed to connect to the EC2 instance through SSH.
    VPCID Select the DEFAULT VPC ID of this region from the dropdown list. Your instance is launched within this VPC.
    WebUICIDR The allowed CIDR IP range that can access the TIBCO ComputeDB Monitoring Console. It can also be the same as ClientCIDR, for example, 123.201.112.0/32. To access the cluster from your personal laptop, first, find your public IP address. You can go to whatismyip.com to get your IPv4 address and append /32 to it.

    Note

    On the next page, you may see the following error message if you do not have adequate permissions for AWS SNS topics:
    Failed to retrieve sns topics
    You can safely ignore this message and proceed to create the stack.

  10. In the Configure stack options page, you can configure advance options for the stack. Click Next. The Review page is displayed.

  11. In the Review page, you can review and edit the stack details before proceeding to create the stack. This page also displays the stack description. The stack description mentions if the cluster will be launched with security enabled. You can also view the details about the status of stack creation from the Events tab. The status changes to CREATE_COMPLETE when it is successfully completed.

    Note

    A test LDAP server is launched within the launched EC2 instance to demonstrate a TIBCO ComputeDB cluster with security enabled. You must provision your own LDAP server to work with the TIBCO ComputeDB cluster in production. For more details about configuring a secure TIBCO ComputeDB cluster with LDAP, refer Launching the Cluster in Secure Mode.

  12. After the CREATE_COMPLETE status is displayed, click the Outputs tab. The following two URLs are displayed:

    • ComputeDBConsole:
      Click ComputeDBConsole URL to access the TIBCO ComputeDB monitoring console. The monitoring console, which provides an overview of the cluster, is displayed. You can only access it from the machines whose IP falls in the CIDR IP range you provided as WebUICIDR while creating the stack.

      Note

      If your cluster has security enabled, you will be prompted for the username and password to access this console. Use snappyuser as the username and the EC2 instance ID as the password. You can find this instance ID on this page.

    • ZeppelinNotebooks:
      Click the ZeppelinNotebooks URL to access the Apache Zeppelin server that is launched along with the TIBCO ComputeDB cluster. You can only access it from the machines whose IP falls in the CIDR IP range you provided as WebUICIDR while creating the stack. It comes with a set of sample Notebooks for you to interact with the TIBCO ComputeDB cluster. Clone the Notebook you want to run. You can also create your own Notebooks. For more details, refer to Using Apache Zepplin with TIBCO ComputeDB.

Setting Credentials for SnappySession in Notebooks

If your cluster has security enabled, the credentials required for your Notebook to communicate with the TIBCO ComputeDB cluster are already pre-configured in the interpreter settings. However, some of the Notebooks that use SnappySession APIs must be modified to set the credentials in the session's conf:

To modify SnappySession APIs in the Notebooks, do the following:

  1. From the Notebook dropdown, select ComputeDB API Quickstart Notebook.

    Note

    You must clone this Notebook before you run its paragraphs.

  2. Go to Step 1, where the SnappySession instance is created and add the following lines after the initialization:

        snappyNoteSession.conf.set(“user”, “snappyuser”)
    snappyNoteSession.conf.set(“password”, “<EC2-Instance-ID>”)

    Refer to the following example:

        exec scala

    import snappy.implicits._
    val snappyNoteSession = new org.apache.spark.sql.SnappySession(sc)
    snappyNoteSession.conf.set(“user”, “snappyuser”)
    snappyNoteSession.conf.set(“password”, “<EC2-Instance-ID>”)

    Replace <EC2-Instance-ID> above with the actual instance ID.

Connecting to the TIBCO ComputeDB Cluster from JDBC URL

You can also connect to the cluster via your JDBC program using the following JDBC connection string: 

jdbc:snappydata://ec2-instance-IP:1527/

Note

Your program (or JDBC client tool) should be running on a machine whose IP address is in the range you provided as ClientCIDR while creating the stack.
If your cluster has security enabled, then you must also specify the username and password as connection properties.

Connecting to the EC2 Instance via SSH

You can log into the EC2 instance via SSH from your laptop using the following command:

chmod 700 /path/to/your/key/file.pem
ssh -i /path/to/your/key/file.pem ec2-user@public-IP-ec2-instance

Your laptop's IP address must be in the range you provided as ClientCIDR while creating the stack. Your key file (.pem) should not have read/write permissions for other users.

TIBCO ComputeDB is installed at /opt/tibco-computedb and Apache Zeppelin is installed at /opt/zeppelin. You can stop, configure, and start TIBCO ComputeDB or Zeppelin as required from these locations.

Deleting the Stack

After you have finished using the clusters, you can delete the CloudFormation stack.

On the AWS CloudFormation page, select the stack that you want to delete and click the Delete button located on the upper right side of the page. All the resources that were created as part of the stack creation will be deleted.