paulie57/aws-fsx-ontap

Deploy a demo Amazon FSx for NetApp ONTAP filesystem with terraform and configure it further w/ ansible

Deployment and Config Management for FSxN Filesystems

As a Cloud Solution Architect for AWS with NetApp I use this code to deploy a complete fsxn filesystem with different volumes for cifs and nfs for demo purposes. With that I can easily spin up a demo environment and destroy it afterwards so that I get a quick and cost-effective environment to showcase the solution or do some testings with fsxn. You can find examples for the deployment with Cloud Formation and Terraform and the configuration with Ansible. Feel free to adapt it according to your needs.

Recent Changes

15.09.2022

Changed folder structure to Ansible, Terraform and CloudFormation Added CloudFormation Template to deploy FSxN with an SVM joined to an AD

01.08.2022

Add deployment folder for single az deployment Add variable for deployment type New aws provider version required, update your environment with terraform init -upgrade in advance

Terraform

• As a prerequisite you have to configure the aws cli with your credentials to access aws
• Edit the terraform.tfvars file and put in your settings
• Create secret.tfvars file and provide sensitiv information. You can use the sample_secret.tfvars file as a reference and copy it to secret.tfvars
• When running terraform plan make sure to specify -var-file secret.tfvars otherwise you will get prompted for the sensitive information
• Then you can do the normal terraform procedure to deploy the fsxn filesystem
• There are different directories to create a filesystem and an svm, only an svm or only a volume
• There is also a directory for a flexcache demo environment. It deploys two FSxN instances. One with an active directory configuration and another one just with a vserver. In the subdirectory config-flexcache you can find Ansible playbooks to create the relationship with the cluster peering and the vserver peering between both instances as the prerequisite for flexcache. In a second playbook there is the play for creating the flexcache that caches the volume on the first instance. In addition playbooks to rollback the config are contained as well.

Ansible

• Edit the variables file and put in the settings for your environment
• Sensitive information can be put into the vault_vars.yml file. Then it should be encrypted with ansible-vault. In the ansible.cfg there's a file specified where you can store your decrypt password. It defaults to ~/.vault_password and can be changed on your behalf. There's also a sample_vault_vars.yml file as a reference that you can copy.

Playbooks

• Create nfs volume
• Create cifs volume, DNS, Cifs Server and cifs share
• Create qtree and a qtree quota on a volume (New!)
• Plus remove playbooks for these resources

CloudFormation

• Edit the template and replace object ids with the ids from your environment
• Create stack in cloudformation console and upload template from file or copy it to an S3 bucket
• The template fetches the passwords from a secret string in aws secretsmanager
• Create this secret in secretsmanager in advance and change the credential name in the template accordingly