Cloud Formation Scripts (command line)

Recommended when installation needs to be automated or have large number of VexData servers to be deployed.

CFN(Cloud Formation) Scripts Download

Prerequisites

1. Ensure jq and "aws cli" are installed

2. Permissions for aws account to create ec2 instance, load balancer.

3. Ensure the private subnet where Vexdata server is installed has out going traffic enabled.

Download CFN Scripts

//clone the CFN scripts.
git clone https://bitbucket.org/fireman10/dq_devops.git

// cd to ec2_cfn 
cd dq_devops/ec2_cfn

1. Create a new file from the template file in dq_devops/ec2_cfn/cfn/cfn_templates_parameters/infrastructure/<ENV>.<ACCOUNT_>.<REGION>.ec2-infra.cfn.parameters.json

   Replace ENV with the environment. ex : dev, prod\

   Replace ACCOUNT with the environment. ex : Vexdatadv

   Replace REGION with the environment. ex : us-east-1

   Replace ENV with the environment. ex : dev, prod

If creating the server in dev environment, the new file name can be : dev.account.region.ec2-infra.cfn.parameters.json

1. Below five properties are mandatory.

   • VPCId

   • PrivateSubnets

   • PublicSubnets

   • KeyPairName

   • ALBSSLPolicy

   • ACMCertificateArn Note: PublicSubnets are coma separated values used by load balancer.

2. Update the CFN Stack Tags File as needed in ./cfn/cfn_templates_parameters/common/env.account.region.cfn-stack.tags.json

3. Update both the Parameter and Stack Tags files name ( env.account.region.emr-infra.cfn.parameters.json and env.account.region.cfn-stack.tags.json ) with corresponding env/account/region values you need. File should be like this dq_devops/ec2_cfn/cfn/cfn_templates_parameters/infrastructure/<ENV>.<ACCOUNT_>.<REGION>.region.cfn-stack.tags.json

4. Update sed command under replace_template_parameters() function in the script(ec2_cfn/shell_scripts/ec2_infra_deploy.sh) as below,

   • If running the scripts in Mac, UPDATE all 'sed' commands in below Format: >

     sed -i '' "s/ParameterNameValue/${PARAMETER_NAME}/g" $PARAMETER_FILE

   • If running the scripts in LINUX, UPDATE all 'sed' commands in below Format: >

     sed -i "s/ParameterNameValue/${PARAMETER_NAME}/g" $PARAMETER_FILE

   \

5. Execute Infrastructure Shell Script:

   Below example will create infra for dev environment and account Vexdata with internet-facing (public) Application Load Balancer and http protocol.

// Execute below command to launch.
./ec2_infra_deploy.sh --stack-tag <STACK_NAME> --subsystem app --env <ENV_TEMPLATE> --account Vexdata --region us-east-1 --ec2-ami-id 'ami-0947d2ba12ee1ff75' --alb-scheme 'internet-facing' --alb-protocol 'http' --create-r53 'false' --r53-hosted-zone-id '' --r53-record-set-name ''

// example command 
./ec2_infra_deploy.sh --stack-tag Vexdata --subsystem app --env dev --account Vexdata --region us-east-1 --ec2-ami-id 'ami-0947d2ba12ee1ff75' --alb-scheme 'internet-facing' --alb-protocol 'http' --create-r53 'false' --r53-hosted-zone-id '' --r53-record-set-name ''

Proceed with the following steps after successful installation

1. Login to the aws web console under cloud formation to confirm the stack is created successful.

2. Click on the stack name-> Outputs -> corresponding to the key "LoadBalancerDnsName", there will be a load balancer url. Click on the URL to login to Vexdata application.

3. if the jobs need to be executed on EMR cluster, ssh into the ec2 instance where Vexdata is running, and URL needs to be set in app.properties file.