Pulumi errors with AWS's required format for Elasticache cluster version

[followben]

I tried to create an Redis cluster like so:

cluster = aws.elasticache.Cluster(
            name,
            engine="redis",
            engine_version="6.0.5",
            node_type="cache.m4.large",
            num_cache_nodes=1,
            port=6379,
            subnet_group_name=subnetGroup.name,
            security_group_ids=[self.securityGroup.id],
        )

Which threw the following error:

aws:elasticache:Cluster (apinfra):
    error: 1 error occurred:
    	* error creating ElastiCache Cache Cluster: InvalidParameterValue: Specific version selection is not allowed for '6.0.5', please use '6.x'
    	status code: 400, request id: 1068124d-ae27-474c-8a90-eda86ecf29a7

After updating the engine_version to "6.x", the cluster created successfully. However thereafter pulumi up fails with the following:

% pulumi up                                                                                    
Previewing update (saberastro/dev)

View Live: https://app.pulumi.com/org/project/dev/previews/821241f1-12ea-w0f0-9fdc-712b8d7aa9e8

     Type                        Name         Plan     Info
     pulumi:pulumi:Stack         project-dev           
     └─ aws:elasticache:Cluster  apinfra               1 error
 
Diagnostics:
  aws:elasticache:Cluster (project):
    error: diffing urn:pulumi:dev::apinfra::aws:elasticache/cluster:Cluster::project: Malformed version: 6.x

Expected behavior

Pulumi should handle wildcard versions as requested by AWS.

Current behavior

Pulumi throws a Malformed version error when trying to compare the stack.

Steps to reproduce

See above.

Context (Environment)

We are currently unable to maintain or update our project's infra with Redis.

Affected feature

aws.elasticache.Cluster

[followben]

Appears this could be related to the terraform provider: hashicorp/terraform-provider-aws#18920

[stack72]

Hi @followben

I believe this has been solved by pulumi-aws 4.2.0

have you been able to upgrade to use this?

Paul

[stack72]

Hi @followben

I am going to close this out as we have ingested the upstream bug fix - please let me know if this is still an issue and I can investigate further

Thanks

Paul