Terraform provider to create cassandra tables

Question

I'm looking to create cassandra DB tables through terraform, on Azure. I already have the relative keyspaces in place.

My deployment is leveraging azurerm, however their provisioner is lacking a cassandra-tables resources. As of now, I can only deploy cassandra tables through Azure UI on the portal or with Azure CLI scripting, however this isn't the best solution for a variety of reasons.

Is there a provider that could help me with this? I'm giving a look around but it seems that there isn't much I could leverage.

Answer 1

Apparently a workaround could be deploying the resource to Azure as ARM using the ARM provider in an "incremental" mode.

 resource "azurerm_resource_group_template_deployment" "example" {
   depends_on = [module.cassandratest]

   name                = "example-cassandra-tables"
   resource_group_name = azurerm_resource_group.test.name
   deployment_mode     = "Incremental"
   template_content    = templatefile("resources/templatecosmos.json", {
     cosmos_db_account_name    = "test-cassandra-2",
     keyspace_name             = "keyspace1",
     table_name                = "test-table-2",
     autoscale_max_throughput  = 4000
   })
 }

CosmosDB Cassandra templates are documented here. An example of the contents of resources/templatecosmos.json:

{
 "$schema": "https://schema.management.azure.com/schemas/2015-01-01/deploymentTemplate.json#",
 "contentVersion": "1.0.0.0",
 "parameters": {
    "accountName": {
       "type": "string",
       "defaultValue": "${cosmos_db_account_name}",
       "metadata": {
          "description": "Cosmos DB account name, max length 44 characters"
       }
    },
    "keyspaceName": {
       "type": "string",
       "defaultValue": "${keyspace_name}",
       "metadata": {
          "description": "The name for the Cassandra Keyspace"
       }
    },
    "tableName": {
       "type": "string",
       "defaultValue": "${table_name}",
       "metadata": {
          "description": "The name for the Cassandra table"
       }
    },
    "autoscaleMaxThroughput": {
       "type": "int",
       "defaultValue": "[int(${autoscale_max_throughput})]",
       "minValue": 4000,
       "maxValue": 1000000,
       "metadata": {
          "description": "Maximum autoscale throughput for the Cassandra table"
       }
    }
 },
 "variables": {
    "accountName": "[toLower(parameters('accountName'))]",
    "databaseRef": "[resourceId('Microsoft.DocumentDB/databaseAccounts', parameters('accountName'))]",
    "keyspaceRef": "[resourceId('Microsoft.DocumentDB/databaseAccounts/cassandraKeyspaces', parameters('accountName'), parameters('keyspaceName'))]"
 },
  "resources": [
    {
       "type": "Microsoft.DocumentDb/databaseAccounts/cassandraKeyspaces/tables",
        "name": "[concat(variables('accountName'), '/', parameters('keyspaceName'), '/', parameters('tableName'))]",
        "apiVersion": "2020-04-01",
        "properties": {
           "resource": {
              "id": "[concat(parameters('tableName'))]",
              "schema": {
                 "columns": [
                    {
                       "name": "loadid",
                       "type": "uuid"
                    }
                 ],
                 "partitionKeys": [
                    { "name": "machine" },
                    { "name": "cpu" },
                    { "name": "mtime" }
                 ],
                 "clusterKeys": [
                    {
                       "name": "loadid",
                       "orderBy": "asc"
                    }
                 ]
              }
           },
           "options": {
              "autoscaleSettings": { 
                 "maxThroughput": "[parameters('autoscaleMaxThroughput')]" 
              }
           }
        }
     }
 ]
} 

Answer 2

For whatever reason it looks like hashicorp never implemented cassandra table in their provider. Their source code is missing the implementation for it.

I suggest filing a new bug on their repo. You can do that here