Amazon ElastiCache for Redis

This topic gives you reference information about the Amazon ElastiCache for Redis (csb-aws-redis) service. It details the plans, configuration parameters, and binding credentials.

Note
Tanzu Cloud Service Broker for AWS v1.5.0 removes brokerpak pre-configured plans for Amazon ElastiCache for Redis. All plans must be configured through the tile. If you have instances of Amazon ElastiCache for Redis that were created in previous versions and that you want to maintain, see Previously Provided Pre-configured Plans. For more information about upgrade steps, see Upgrading Tanzu Cloud Service Broker for AWS. Also, in Tanzu Cloud Service Broker for AWS v1.5.0 and later, changes made to this service by using the cf update-service command are applied immediately. This is instead of being queued to be applied during the maintenance window. This might result in brief downtime while the server reboots.

Plan Configuration Parameters

When configuring Cloud Service Broker for AWS you can add additional plans. For how to configure plans, see Configure Services with Cloud Service Broker for AWS.

The following table lists parameters that can only be configured for additional plans:

Parameter name	Values	Default	Required
`name`	The plan name.	n/a	Yes
`id`	A unique GUID.	n/a	Yes
`description`	Description of the new plan.	n/a	Yes
`free`	When false, service instances of this service plan have a cost.	true	No
`bindable`	Specifies whether service instances of the service plan can bind to applications.	true	No
`plan_updateable`	Whether the plan supports upgrading, downgrading, or sidegrading to another version.	true	No
`cache_size`	Deprecated. Use `node_type` instead. The cache size is in GB. 2–256, multiples of 2.	n/a	Yes
`metadata.displayName`	Name to use when displaying the plan in the Marketplace.	n/a	No
`metadata.bullets`	List of bullet points to display in Apps Manager.	n/a	No

You can also add any of the parameters listed in the Configuration Parameters section to your plan.

Note
If you set a parameter at plan level, developers cannot change the value when creating or updating service instances.

Configuration Parameters

You can provision a service by running:

cf create-service csb-aws-redis PLAN-NAME SERVICE-INSTANCE-NAME -c '{"PARAMETER-NAME": "PARAMETER-VALUE"}

You can update the configuration parameters for a service instance by running:

cf update-service SERVICE-INSTANCE -c '{"PARAMETER-NAME": "PARAMETER-VALUE"}'

The following table lists the parameters that you can configure, by using the -c flag, when provisioning or updating a csb-aws-redis service. The Operation column displays whether a parameter is supported for both provision and update, or for provision only:

Parameter name	Type	Description	Default	Operation
`instance_name`	String	The name of the AWS instance to create	`csbINSTANCE-ID`	provision
`region`	String	The AWS region to deploy the service in. For more information about available regions, see the AWS documentation.	The value the operator entered for Region in Ops Manager.	provision
`node_type`	String	The AWS instance class to be used. The `node_type` you select impacts costs, performance, and fault tolerance. The `node_type` cannot be updated from a family of type `r6gd` to all other families and the reverse. For more information about node types and pricing, see the AWS documentation and Choosing your node size.	`""`	provision and update
`node_count`	Integer	Number of nodes (primary and replicas) in the cluster	`2`	provision and update
`redis_version`	String	The version for the Redis instance to be used in the replication group. If the version is 6 or later, you can set the major and minor version. For example, `6.2`. Alternatively, the minor version can be unspecified, which uses the latest version at creation time. For example, `6.x`. Otherwise, specify the full version that you want. For example, `5.0.6`. In other words, `redis_version` must match `major.minor` when using version `6` or higher, or `major.minor.patch`. For more information about upgrading engine versions, see the AWS documentation. The downgrade of the version is not allowed because it involves the recreation of the instance. If you try to perform the operation, the Cloud Service Broker incurs an error because of the mechanism established to prevent the destruction of the instance.	None	provision and update
`auto_minor_version_upgrade`	Boolean	(Optional) Indicates that minor engine upgrades are applied automatically to the redis instance during the maintenance window. Only supported for redis version 6 and later. When setting this parameter to `true`, `redis_version` must not specify a minor version and must be in the form ` .x`. For example, `6.x`.	`false`	provision and update
`data_tiering_enabled`	Boolean	Enables data tiering. Data tiering is only supported for replication groups that use the `r6gd` node type. This parameter must be set to `true` when using `r6gd` nodes. Not all versions of Redis support this feature. For more information about node types and supported Redis versions, see the AWS documentation.	`false`	provision
`aws_vpc_id`	String	The Virtual Private Cloud (VPC) to connect the instance to	The default VPC	provision
`elasticache_subnet_group`	String	Pre-existing AWS Elasticache subnet group to use	The value the operator entered for Subnet group in Ops Manager. A new one is created if none is specified in Ops Manager or provision parameters.	provision
`elasticache_vpc_security_group_ids`	String	Comma-delimited list of security group IDs for the instance	The value the operator entered for Security group IDs in Ops Manager. A new one is created if none is specified in Ops Manager or provision parameters.	provision
`at_rest_encryption_enabled`	Boolean	Whether to enable encryption at rest	`true`	provision
`multi_az_enabled`	Boolean	Whether to enable Multi-AZ support for the replication group. Only applies when `node_count` is greater than 1. If `true`, `automatic_failover_enabled` must also be `true`. Enabling Multi-AZ and then trying to deactivate the automatic failover causes an error.	`true`	provision and update
`automatic_failover_enabled`	Boolean	Automatically promote a replica to primary if the existing primary fails. Only applies when `node_count` is greater than 1. You can't set this property to `false` unless you also deactivate Multi-AZ. Enabling automatic failover and then trying to reduce the number of nodes to 1 causes an error.	`true`	provision and update
`kms_key_id`	String	The ARN of the key to use if encrypting at rest. Defaults to AWS-managed key.	`""`	provision
`aws_access_key_id`	String	The AWS Access Key to use for an instance	The value the operator entered for AWS Access Key in Ops Manager	provision and update
`aws_secret_access_key`	String	The corresponding secret for the AWS Access Key to use for an instance	The value the operator entered for AWS Secret Access Key in Ops Manager	provision and update
`maintenance_day`	String	The preferred maintenance day: `Sun`, `Mon`, `Tue`, `Wed`, `Thu`, `Fri`, or `Sat`. It can only be updated to a non-null value.	Uses AWS default value for the region if all `maintenance_*` properties are null when creating the instance	provision and update
`maintenance_start_hour`	String	The preferred maintenance start hour. Valid values are `00` through `23`. It can only be updated to a non-null value.	Uses AWS default value for the region if all `maintenance_*` properties are null when creating the instance	provision and update
`maintenance_start_min`	String	The preferred maintenance start minute. Valid values are `00`, `15`, `30`, `45`. It can only be updated to a non-null value.	Uses AWS default value for the region if all `maintenance_*` properties are null when creating the instance	provision and update
`maintenance_end_hour`	String	The preferred maintenance end hour. Valid values are `00` through `23`. It can only be updated to a non-null value. The minimum maintenance window is a 60 minute period.	Uses AWS default value for the region if all `maintenance_*` properties are null when creating the instance	provision and update
`maintenance_end_min`	String	The preferred maintenance end minute. Valid values are `00`, `15`, `30`, `45`. It can only be updated to a non-null value. The minimum maintenance window is a 60 minute period.	Uses AWS default value for the region if all `maintenance_*` properties are null when creating the instance	provision and update
`backup_retention_limit`	Integer	Number of days for which ElastiCache retains automatic cache cluster snapshots before deleting them. If set to `0`, backups are deactivated.	`1`	provision and update
`final_backup_identifier`	String	The name of the final node group (shard) snapshot. ElastiCache creates the snapshot from the primary node in the cluster. If omitted, no final snapshot is made.	`null`	provision and update
`backup_name`	String	The name of an existing snapshot to be restored in this new instance.	`""`	provision
`backup_start_hour`	String	The preferred backup start hour. Valid values are `00` to `23`. It can only be updated to a non-null value.	Uses AWS default value for the region if all `backup_*` properties are null when creating the instance.	provision and update
`backup_start_min`	String	The preferred backup start minute. Valid values are `00`, `15`, `30`, `45`. It can only be updated to a non-null value. The minimum backup window is a 60 minute period.	Uses AWS default value for the region if all `backup_*` properties are null when creating the instance.	provision and update
`backup_end_hour`	String	The preferred backup end hour. Valid values are `00` to `23`. It can only be updated to a non-null value. The minimum backup window is a 60 minute period.	Uses AWS default value for the region if all `backup_*` properties are null when creating the instance.	provision and update
`backup_end_min`	String	The preferred backup end minute. Valid values are `00`, `15`, `30`, `45`. It can only be updated to a non-null value.	Uses AWS default value for the region if all `backup_*` properties are null when creating the instance.	provision and update
`parameter_group_name`	String	Name of the custom parameter group to associate with this instance. If left unset, the default parameter group for the specified `redis_version` is used. After the parameter is set, make sure that it is updated to an appropriate value when updating the `redis_version`. To set it to the default value, specify the name of the default parameter group for the Redis version. For example, `default.redis6.x`. For more information about parameter groups, see the AWS documentation.	`""`	provision and update
`preferred_azs`	Array	List of EC2 availability zones in which the nodes are created. The first item in the list is the primary node. The number of entries must equal `node_count`. After this property is set, `node_count` cannot be changed.	`[]`	provision
`logs_slow_log_enabled`	Boolean	Enable the streaming of Redis Slow Log to CloudWatch in JSON format. Slow Log is supported for Redis replication groups using v6.0 and later. A new CloudWatch log group is created with the name `/aws/elasticache/cluster/SERVICE-INSTANCE-NAME/slow-log`. Deleting the instance also deletes the log group and permanently deletes all the archived log events associated with it.	`false`	provision and update
`logs_slow_log_loggroup_retention_in_days`	Number	Specifies the number of days to retain log events in the specified log group. If `0` is specified, the events in the log group are always retained and never expire. When specifying `logs_slow_log_loggroup_retention_in_days`, set `logs_slow_log_enabled` as `true`. For more information and the list of valid values, see the AWS documentation.	`0`	provision and update
`logs_slow_log_loggroup_kms_key_id`	String	The ARN for the KMS key to encrypt Slow Log CloudWatch logs. When specifying `logs_slow_log_loggroup_kms_key_id`, set `logs_slow_log_enabled` as `true`. If omitted, CloudWatch default encryption applies. For information about CloudWatch log data encryption and how to configure a KMS key, see the AWS documentation.	`""`	provision and update
`logs_engine_log_enabled`	Boolean	Enable the streaming of Redis Engine Log to CloudWatch in JSON format. Engine Log is supported for Redis replication groups using v6.2 and later. A new CloudWatch log group is created with the name `/aws/elasticache/cluster/SERVICE-INSTANCE-NAME/engine-log`. Deleting the instance also deletes the log group and permanently deletes all the archived log events associated with it.	`false`	provision and update
`logs_engine_log_loggroup_retention_in_days`	Number	Specifies the number of days to retain log events in the specified log group. If `0` is specified, the events in the log group are always retained and never expire. When specifying `logs_engine_log_loggroup_retention_in_days`, set `logs_engine_log_enabled` to `true`. For more information and the list of valid values, see the AWS documentation.	`0`	provision and update
`logs_engine_log_loggroup_kms_key_id`	String	The ARN for the KMS key to encrypt Engine logs CloudWatch logs. When specifying `logs_engine_log_loggroup_kms_key_id`, set `logs_engine_log_enabled` to `true`. If omitted, CloudWatch default encryption applies. For information about CloudWatch log data encryption and how to configure a KMS key, see the AWS documentation.	`""`	provision and update

Binding

You can bind a service by running:

cf bind-service APP-NAME SERVICE-INSTANCE --binding-name BINDING-NAME

Cloud Service Broker for AWS returns the Redis AUTH token created at provision time alongside endpoints information to connect to the service instance.

Note
Because the Cloud Service Broker for AWS configures Redis AUTH as the authentication mechanism, all bindings share the same Redis AUTH token. Deleting the binding does not revoke the credentials.

Binding Credentials

The format for binding credentials for Redis is as follows:

{
    "name" : "REDIS-CLUSTER-NAME",
    "host" : "REDIS-PRIMARY-ENDPOINT-ADDRESS",
    "password" : "AUTHENTICATION-PASSWORD",
    "tls_port" : "REDIS-TLS-PORT",
    "reader_endpoint" : "REDIS-READER-ENDPOINT-ADDRESS"
}

host is the primary node endpoint address. This endpoint can be used both for writing and reading.

reader_endpoint is the IaaS-recommended endpoint to use for reads whenever eventual consistency is acceptable. The reader endpoint evenly splits incoming connections to the endpoint between all read replicas. For more information, see the AWS documentation.

Previously Provided Pre-configured Plans

The following table lists the previously provided plans for Amazon ElastiCache for Redis service:

Plan	Cache size	AWS cache node type	HA
small	minimum 1 GB	cache.t2.small	no
medium	minimum 4 GB	cache.m5.large	no
large	minimum 16 GB	cache.r4.xlarge	no
small-ha	minimum 1 GB	cache.t2.small	yes
medium-ha	minimum 4 GB	cache.m5.large	yes
large-ha	minimum 16 GB	cache.r4.xlarge	yes

To keep these plans in this version of the broker, add them through the tile as custom plans. For how to configure plans through the tile, see Configure services with Cloud Service Broker for AWS.

Important
Property cache_size is now deprecated. Consider replacing it with its equivalent node_type. For more information, see Equivalence of cache_size and node_type. Specifying node_type in a plan prevents any existing or future service from specifying node_type as a service-specific configuration parameter.

This has strong implications for existing plans because existing services might become unmanageable if they specified a custom node_type when they were created.

When in doubt, keep using cache_size for existing plans and start using node_type for any new plans.

For more information about node types and pricing, see the AWS documentation and Choosing your node size.

Tanzu Cloud Service Broker for AWS v1.5.0 sets the default values to true for at_rest_encryption_enabled and multi_az_enabled. Because previously created instances cannot be modified to use Multi-AZ or encryption at rest, set the properties multi_az_enabled and at_rest_encryption_enabled to false when replicating the old plans to match service instances that have already been created.

Add the following block to keep the small plan:

{
  "name": "small",
  "id": "ad963fcd-19f7-4b79-8e6d-645756e84f7a",
  "description": "Redis 6.0 with 1GB cache and 1 node.",
  "cache_size": 2,
  "redis_version": "6.0",
  "node_count": 1,
  "at_rest_encryption_enabled": false,
  "multi_az_enabled": false,
  "metadata": {
    "displayName": "Small",
    "bullets": ["Redis 6.0", "1GB Cache", "1 node"]
  }
}

Add the following block to keep the medium plan:

{
  "name": "medium",
  "id": "df41095a-43e8-4be4-b4d6-ae2d8a35068d",
  "description": "Redis 6.0 with at least 4GB cache and 1 node.",
  "cache_size": 4,
  "redis_version": "6.0",
  "node_count": 1,
  "at_rest_encryption_enabled": false,
  "multi_az_enabled": false,
  "metadata": {
    "displayName": "Medium",
    "bullets": ["Redis 6.0", "4GB Cache", "1 node"]
  }
}

Add the following block to keep the large plan:

{
  "name": "large",
  "id": "da4dc49c-a64f-4d2a-8490-5e456cbb0577",
  "description": "Redis 6.0 with at least 16 GB cache and 1 node.",
  "cache_size": 16,
  "redis_version": "6.0",
  "node_count": 1,
  "at_rest_encryption_enabled": false,
  "multi_az_enabled": false,
  "metadata": {
    "displayName": "Large",
    "bullets": ["Redis 6.0", "16GB Cache", "1 node"]
  }
}

Add the following block to keep the small-ha plan:

{
  "name": "small-ha",
  "id": "70544df7-0ac4-4580-ba51-c1fbdd6fdfd0",
  "description": "Redis 6.0 with 1GB cache and 2 nodes.",
  "cache_size": 2,
  "redis_version": "6.0",
  "at_rest_encryption_enabled": false,
  "multi_az_enabled": false,
  "metadata": {
    "displayName": "Small HA",
    "bullets": ["Redis 6.0", "1GB Cache", "2 nodes"]
  }
}

Add the following block to keep the medium-ha plan:

{
  "name": "medium-ha",
  "id": "a4235008-80f4-4053-924b-defcce17cb63",
  "description": "Redis 6.0 with at least 4GB cache and 2 nodes.",
  "cache_size": 4,
  "redis_version": "6.0",
  "at_rest_encryption_enabled": false,
  "multi_az_enabled": false,
  "metadata": {
    "displayName": "Medium HA",
    "bullets": ["Redis 6.0", "4GB Cache", "2 nodes"]
  }
}

Add the following block to keep the large-ha plan:

{
  "name": "large-ha",
  "id": "f26cda6f-d4b4-473a-966c-32d238f723ef",
  "description": "Redis 6.0 with at least 16 GB cache and 2 nodes.",
  "cache_size": 16,
  "redis_version": "6.0",
  "at_rest_encryption_enabled": false,
  "multi_az_enabled": false,
  "metadata": {
    "displayName": "Large HA",
    "bullets": ["Redis 6.0", "16GB Cache", "2 nodes"]
  }
}

Equivalence of cache_size and node_type

Table of equivalents for cache_size and node_type for the previously pre-configured plans:

Plan	Cache size	Description	AWS cache node type	HA
small	1	minimum 1 GB	cache.t2.small	no
medium	4	minimum 4 GB	cache.m5.large	no
large	16	minimum 16 GB	cache.r4.xlarge	no
small-ha	1	minimum 1 GB	cache.t2.small	yes
medium-ha	4	minimum 4 GB	cache.m5.large	yes
large-ha	16	minimum 16 GB	cache.r4.xlarge	yes

Table of equivalents for cache_size and node_type not used by any of the previously pre-configured plans:

Cache size	AWS cache node type
2	cache.t3.medium
8	cache.m5.xlarge
32	cache.r4.2xlarge
64	cache.r4.4xlarge
128	cache.r4.8xlarge
256	cache.r5.12xlarge