Made minor changes.

Author: kcantrel

CloudFormation/README.md

@@ -1,5 +1,5 @@
 # Deploy an Amazon FSx for NetApp ONTAP with CloudFormation
-This repository contains a CloudFormation template that deploys an Amazon FSx for NetApp ONTAP file system. 
+This repository contains a CloudFormation template that can be used to deploy an Amazon FSx for NetApp ONTAP file system. 
 
 ## Configuration
 The CloudFormation template allowed the following configuration parameters to be provided:
@@ -9,7 +9,7 @@ The CloudFormation template allowed the following configuration parameters to be
 |FileSystemType|String|Yes|This should be set to `ONTAP` to deploy an FSx for ONTAP file system.|
 |KmsKeyId|String|No|The ID of the AWS Key Management Service (AWS KMS) key used to encrypt the data at rest. If not provided, a key managed by AWS will  be used.|
 |SecurityGroupIds|Array of Strings|No|While technically not required; without a security group you won't be able to access the FSx for ONTAP file system.|
-|StorageCapacity|Integer|Yes|The number of gigabytes of SSD storage to provision. If you are using tiering, this is the amount to allocate in the "performance tier." The minimum is 1024 (i.e. 1TiB) per HA pair. The maximum is 1,048,576 (i.e. 1 PiB).|
+|StorageCapacity|Integer|Yes|The number of gigabytes of SSD storage to provision. If you are using tiering, this is the amount to allocate in the "performance tier." The minimum is 1024 (i.e. 1TiB) **per** HA pair. The maximum depends on the DeploymentType being used. For MULTI_AZ_1 and SINGLE_AZ_1 the maximum is 196,608 (192Tib). For MULTI_AZ_2, the maximum is 524,288 (512TiB). For SINGLE_AZ_2 the maximum is 1,048,576 (1 PiB).|
 |SubnetIds|string|Yes|These are the subnets that a Network Interface will be created that allows access to the FSx for ONTAP storage and management endpoints.<br><br>When deploying in a single availability zone, you can only specify one subnet. If deploying in a multi availability zone environment, then you need to specify two subnets. In the OntapConfiguration you will specify which subnet contains the primary node of the HA pair.|
 |OntapConfiguration|Structure|Yes|See the below table for the properties that make up this structure.|
 
@@ -19,12 +19,12 @@ This table list the properties that make up the OntapConfiguration structure:
 |---|---|---|---|
 |AutomaticBackupRetentionDays|Integer|No|Sets the number of days to retain an automatic backup. Defaults to 30. Set to 0 to disable automatic backups.|
 |DailyAutomaticBackupStartTime|String|No|Sets when the daily automatic backup should occur. The format should be HH:MM. It should be in 24-hour notation with each number zero filled. The time zone is always UTC. For example, "22:05" means 5 minutes after 10PM UTC. The default is randomly chosen by AWS.
-|DeploymentType|String|Yes|Sets the generation of file servers to use, and whether you want both nodes of the HA pair to be deploy within a single availability zone or spread across multiple availability zones. The supported vales are:<br><ul><li>MULTI_AZ_1 - Deploy generation 1 file servers with each node of the HA pair in different availability zones.</li><li>MULTI_AZ_2 - Deploy generation 2 file servers with each node of the HA pair in different availability zones.</li><li>SINGLE_AZ_1 - Deploy generation 1 file servers with both nodes of the HA pair in the same availability zone.</li><li>SINGLE_AZ_2 - Deploy generation 2 file servers with all nodes in the same availability zone. This specific deployment type allows you to deploy from one to twelve HA pairs in a single file system. Use the HAPairs property to specify how many HA pairs you want deployed.</li></ul>|
-|DiskIopsConfiguration|Structure|No|Sets the maximum IOPS (I/Os per second) the file system can have when accessing the SSDs. It does not set the maximum IOPS for the file system itself, since the file system can response with data that is already in its cache.  The definition of the structure is below.|
-|EndpointIpAddress|String|No|Set the CIDR (i.e. IP Address range) that the file system endpoints will be allocated from. This can only be specified for multiple availability zone deployments. For more information how this parameter is used, please visit this webpage. The default is 198.19.0.0/16.|
+|DeploymentType|String|Yes|Sets the generation of file servers to use, and whether you want both nodes of the HA pair to be deployed within a single availability zone or spread across multiple availability zones. The supported vales are:<br><ul><li>MULTI_AZ_1 - Deploy generation 1 file servers with each node of the HA pair in different availability zones.</li><li>MULTI_AZ_2 - Deploy generation 2 file servers with each node of the HA pair in different availability zones.</li><li>SINGLE_AZ_1 - Deploy generation 1 file servers with both nodes of the HA pair in the same availability zone.</li><li>SINGLE_AZ_2 - Deploy generation 2 file servers with all nodes in the same availability zone. This specific deployment type allows you to deploy from one to twelve HA pairs in a single file system. Use the HAPairs property to specify how many HA pairs you want deployed.</li></ul>|
+|DiskIopsConfiguration|Structure|No|Sets the maximum IOPS (I/Os per second) the file system can have when accessing the backend SSDs. It does not set the maximum IOPS for the file system itself, since the file system can response with data that is already in its cache.  The definition of the structure is below.|
+|EndpointIpAddress|String|No|Set the CIDR (i.e. IP Address range) that the file system endpoints will be allocated from. This can only be specified for multiple availability zone deployments. For more information how this parameter is used, please visit this [webpage](https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/getting-started-step1.html). The default is 198.19.0.0/16.|
 |FsxAdminPassword|String|No|Sets the default user's (fsxadmin) password. It is not recommended to store the password in plain text anywhere, so the best practice is to not set it when you deploy the file system, but set it later via the AWS console, CLI or API. Or, you can use CloudFormation to read a secret to set the password.|
-|HPApairs|Integer|No|Allows you to set the number of HA pairs to deploy in a single file system. Only used when the DeploymentType is set to "SINGLE_AZ_2". Note, you cannot change the number of nodes that are in a file system once it has been deployed. If you need to change it, will have to deploy a new filesystem and migrate your data to it. The default is number of HA pairs is 1.|
-|PreferredSubnetId|String|No|Required in multiple availability zone deployments. This allows you to specify which subnet you want the primary node to be connected to. It should be one of the subnets specified in the SubnetIds property.|
+|HPApairs|Integer|No|Allows you to set the number of HA pairs to deploy in a single file system. Only used when the DeploymentType is set to "SINGLE_AZ_2". The default is number of HA pairs is 1.|
+|PreferredSubnetId|String|No|Required in multiple availability zone deployments. This allows you to specify which subnet you want the primary node to be connected to. It must be one of the subnets specified in the SubnetIds property.|
 |RouteTableIds|Array of Strings|No|This allows you to specify any route tables you want CloudFormation to update so that it will have a route to the file system. This is only needed for a multiple availability zone deployment. The default is the VPC's default route table.|
 |ThroughputCapacity|Integer|No|This sets the throughput capacity of Gen 1 based file system (i.e. when the DeploymentType is set to "SINGLE_AZ_1" or "MULTI_AZ_1"). The default is based on the amount of storage provisioned. Valid values are:<br><ul><li>128</li><li>256</li><li>512</li><li>1024</li><li>2048</li><li>4096</li></ul>**Note**: You can only specify ThroughputCapacity or ThroughputCapacityPerHAPair but not both. And, since you can use ThroughputCapacityPerHAPair for all deployment types, I would avoid using this one.|
 |ThroughputCapacityPerHaPair|Integer|No|This sets the throughput capacity of each HA pair. This property can be used for both generation 1 and generation 2 type deployments. Use the supported values above if using it with generation 1 type deployment.<br><br>For generation 2 type deployments (i.e. when the deployment type is "MULTI_AZ_2" or "SINGLE_AZ_2") and with only 1 HA pair, the following are the allowed values:<ul><li>384</li><li>768</li><li>1536</li><li>3072</li><li>6144</li></ul>For generation 2 type deployments with more than 1 HA pair, the following are the allowed values:<ul><li>1536</li><li>3072</li><li>6144</li></ul>The default is based on the amount of storage provisioned.<br>**Note:** You can only specify ThroughputCapacity or ThroughputCapacityPerHAPair but not both.|
@@ -42,6 +42,7 @@ There are two main ways of deploying a CloudFormation "Stack" using a template f
 ### Using the AWS Console to deploy a CloudFormation Template
 To use the console, first log into the AWS console (https://console.aws.com) and then go the CloudFormation page. From there select "Stacks," then "Create stack" and finally "With new resources (standard)":
 ![Create Stack](./images/create_stack-01.png)
+
 On the next page, select "Choose an existing template." As you can see there are multiple ways to pass the template file to CloudFormation. If you stored your template file in an S3 bucket, click on "Amazon S3 URL" and filling the URL to the file. Otherwise, if you just are keeping the template as a file on your PC click on "Upload a template file" and then click on "Choose File." 
 ![Choose Template](./images/create_stack-02.png)
 That will bring up a file selection box. Find the file that holds your template and select it. After AWS has read in your file, click "Next."

CloudFormation/template.yaml

@@ -83,52 +83,26 @@ Parameters:
 Conditions:
   MultiAZ:  !Or [!Equals [!Ref DeploymentType, MULTI_AZ_1 ], !Equals [!Ref DeploymentType, MULTI_AZ_2 ]]
 
-  SingleAZ: !Or [!Equals [!Ref DeploymentType, SINGLE_AZ_1], !Equals [!Ref DeploymentType, SINGLE_AZ_2]]
-
-  ScaleOut:       !Equals [!Ref DeploymentType, SINGLE_AZ_2]
-
-  ScaleUp:  !Not [!Equals [!Ref DeploymentType, SINGLE_AZ_2]]
+  ScaleOut: !Equals [!Ref DeploymentType, SINGLE_AZ_2]
 
   IpRange:  !Not [!Equals [!Ref EndpointIpAddressRange, ""]]
 
 Resources:
-#
-# Since there are properties (e.g. RouteTableIds) that can't be specified with a SAZ
-# deployment type we have to have two separate resources but only allow one to be deployed
-# based on the conditions set above.
   FileSystemMAZ:
     Type: "AWS::FSx::FileSystem"
-    Condition: MultiAZ
 
     Properties:
       FileSystemType: "ONTAP"
       StorageCapacity: !Ref StorageCapacity
-      SubnetIds: [!Ref PrimarySubnet, !Ref SecondarySubnet]
+      SubnetIds: [!Ref PrimarySubnet, !If [ MultiAZ, !Ref SecondarySubnet, !Ref "AWS::NoValue" ]]
       SecurityGroupIds: !Ref SecurityGroupIds
       OntapConfiguration:
         DeploymentType: !Ref DeploymentType
         ThroughputCapacityPerHAPair: !Ref ThroughputCapacity
         HAPairs: !If [ ScaleOut, !Ref HAPairs, 1 ]
         PreferredSubnetId: !Ref PrimarySubnet
-        RouteTableIds: !Ref RouteTableIds
+        RouteTableIds: !If [ MultiAZ, !Ref RouteTableIds, !Ref "AWS::NoValue" ]
         EndpointIpAddressRange: !If [ IpRange, !Ref EndpointIpAddressRange, !Ref "AWS::NoValue" ]
       Tags:
         - Key: Name
           Value: !Ref Name
-
-  FileSystemSAZ:
-    Type: "AWS::FSx::FileSystem"
-    Condition: SingleAZ
-
-    Properties:
-      FileSystemType: "ONTAP"
-      StorageCapacity: !Ref StorageCapacity
-      SubnetIds: [!Ref PrimarySubnet]
-      SecurityGroupIds: !Ref SecurityGroupIds
-      OntapConfiguration:
-        DeploymentType: !Ref DeploymentType
-        ThroughputCapacityPerHAPair: !Ref ThroughputCapacity
-        HAPairs: !If [ ScaleOut, !Ref HAPairs, 1 ]
-      Tags:
-        - Key: Name
-          Value: !Ref Name