[Enhancement]: resource/aws_neptune_cluster_instance: Missing options for snapshots

[gdavison]

Description

The resource aws_neptune_cluster_instance does not currently support options for controlling snapshots, including:

• FinalDBSnapshotIdentifier
• SkipFinalSnapshot

Affected Resource(s) and/or Data Source(s)

aws_neptune_cluster_instance

Potential Terraform Configuration

No response

References

No response

Would you like to implement a fix?

None

[github-actions[bot]]

Community Note

Voting for Prioritization

• Please vote on this issue by adding a 👍 reaction to the original post to help the community and maintainers prioritize this request.
• Please see our prioritization guide for information on how we prioritize.
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for issue followers and do not help prioritize the request.

Volunteering to Work on This Issue

• If you are interested in working on this issue, please leave a comment.
• If this would be your first contribution, please review the contribution guide.

[triggan]

Snapshots on Neptune are controlled at the Neptune Cluster level. No data is persisted on Neptune instances, as all instances in the cluster use a shared cluster volume for storage. Is there a reason why you need the listed settings for snapshots for an instance?

[gdavison]

The AWS API allows taking a snapshot when deleting an instance. We support this for most other AWS managed data stores, so for consistency, we should support it for the aws_neptune_cluster_instance resource.

[triggan]

I can see where you are coming from, but there's a bit more to the story. Neptune's control plane is based on that of the RDS control plane and the underlying CLI/SDK options are based on the same options that you would see for RDS. In some cases, these options do not apply, due to differences in the data plane/engine architecture for Neptune vs RDS or Aurora.

If you try to specify a final snapshot when deleting a read replica in a Neptune cluster (for example):

aws neptune delete-db-instance --db-instance-identifier neptune-instance-replica-xlarge-2 \
    --region us-west-2 \
    --debug --no-skip-final-snapshot \
    --final-db-snapshot-identifier testinstsnap

You'll get an error, such as the following:

An error occurred (InvalidParameterCombination) when calling the DeleteDBInstance operation:
 FinalDBSnapshotIdentifier can not be specified when deleting a cluster instance

The man page for the aws neptune delete-db-instance also states:

       --skip-final-snapshot | --no-skip-final-snapshot (boolean)
          Determines whether a final DB snapshot is created before the DB
          instance is deleted. If true is specified, no DBSnapshot is created.
          If false is specified, a DB snapshot is created before the DB
          instance is deleted.

          Note that when a DB instance is in a failure state and has a status
          of 'failed', 'incompatible-restore', or 'incompatible-network', it
          can only be deleted when the SkipFinalSnapshot parameter is set to
          "true".

          **Specify true when deleting a Read Replica.**

... which is different than the man page description for the same option for RDS.

Even if you specify --no-skip-final-snapshot, the option gets ignored.

These options only apply when deleting the cluster.