matano init should create a unique resource identier

[mdfranz]

When running multiple inits (assuming a new directory) is used I would expect a new unique identifier would be created each time

 CDKToolkit |  0/12 | 9:06:46 AM | CREATE_FAILED        | AWS::S3::Bucket         | StagingBucket cdk-hnb659fds-assets-XXXXXX-us-east-2 already exists

I ran init twice and cdk-hnb659fds seems to be re-used, i would expect this to be unique each run, but maybe this is a constraint of CDK

When you have multiple repeated failures to deploy this makes cleanup difficult. I would also expect each deployment to have unique roles.

{
  "version": "20.0.0",
  "files": {
    "70f03c831095bf0345af1dac68037dcb2b95a9fe0c4b4d27738cfad55da1c8c7": {
      "source": {
        "path": "DPCommonStack.template.json",
        "packaging": "file"
      },
      "destinations": {
        "647303185053-us-east-2": {
          "bucketName": "cdk-hnb659fds-assets-XXXXX-us-east-2",
          "objectKey": "70f03c831095bf0345af1dac68037dcb2b95a9fe0c4b4d27738cfad55da1c8c7.json",
          "region": "us-east-2",
          "assumeRoleArn": "arn:${AWS::Partition}:iam::XXX:role/cdk-hnb659fds-file-publishing-role-XXX-us-east-2"
        }
      }
    }
  },
  "dockerImages": {}
}mfranz@pixel-slate-cros:~$ cat /tmp/matanocdkoutonT9xy/DPCommonStack.assets.json 
{
  "version": "20.0.0",
  "files": {
    "70f03c831095bf0345af1dac68037dcb2b95a9fe0c4b4d27738cfad55da1c8c7": {
      "source": {
        "path": "DPCommonStack.template.json",
        "packaging": "file"
      },
      "destinations": {
        "647303185053-us-east-2": {
          "bucketName": "cdk-hnb659fds-assets-XXXX-us-east-2",
          "objectKey": "70f03c831095bf0345af1dac68037dcb2b95a9fe0c4b4d27738cfad55da1c8c7.json",
          "region": "us-east-2",
          "assumeRoleArn": "arn:${AWS::Partition}:iam::XXX:role/cdk-hnb659fds-file-publishing-role-XXX-us-east-2"
        }
      }
    }
  },
  "dockerImages": {}

[shaeqahmed]

Multiple Matano deployments within the same region + account is currently not supported.

We can use this issue to track that if it is a feature request, and feel free to describe why this would be useful.

If it is mostly for helping with cleanups, we can solve that with a simpler fix (is_production: false option for the matano.config.yml to disable aggressive retain-on-delete policies specified on stateful resources, a matano destory command, etc.)

[mdfranz]

Cool, will try the config changes to make life easier during testing. More "principle of least suprise" that if I have a expected a different config directory that word correspond to a different backend infrastructure. I guess it is also reasonable that multiple matano deployments would be in different AWS accounts or regions so there would not be collision.

[gdrapp]

A matano destroy command would be great. Currently it’s a pain to clean up SQS queues and S3 buckets when experimenting.

[timoguin]

The matano destroy command was implemented in PR #34 a few days ago. Could y'all test it out and let us know if that resolves the issue?