cloudera.cloud.env failing when environment already exists

[nmarian85]

I have one question on the behavior below. I re-ran the same pipeline that created this environment and, when the environment is present, it fails. From what I saw, the environment is the only component where this happens. When trying to install a CDE/CML/etc, cluster that is already installed, the same warnings appear but the module does not fail.
Any insights on this ? I could go around it by tweaking the failed_when: never flag and having another task which checks if the installation went fine by querying the environment status. However, that is not oki since, in case of an environment creation failure, the failed_when will hide the error.

TASK [env : Install CDP environment ] ********************************
Friday 29 April 2022  13:56:09 +0000 (0:00:00.039)       0:00:00.039 ********** 
[WARNING]: Environment SSH tunneling specified. Currently, the SSH tunnel
setting cannot be reconciled. To update the tunneling setting, explicitly
delete and recreate the environment.
[WARNING]: Environment workload analytics specified. Currently, the
environment's workload analytics setting cannot be reconciled. To update the
workload analytics setting for theenvironment, explicitly delete and recreate
the environment.
[WARNING]: Environment tags specified. Currently, tags cannot be reconciled. To
update tags, explicitly delete and recreate the environment.
fatal: [localhost]: FAILED! => {
    "changed": false,
    "violations": [
        [
            "public_key_text",
            "ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQDZkgFHHkJltjGKOM93SwPQ+qM16zBh7rXLdgvwGhoLUDIqJgtqTBW03i8w8q4LxjZLTCV2LEiLTQIog+CcuzOPK9XWadzCW1LkovlZ5nG4Me+F0aN8I9jupYvMe1f8+xk4ZAKHLeI14gWmti+fJkTN4q1cK8cX0gzG1pEPXMdYtStkjVBk65z/VsqmqjinoQnuDSaa6Xd16dXMbbID9y2qRgAM91MWGKS0DGyH0YTSjkid3ZADQKgUblYdMRNSoaktSKhV68BLUtuWM0MfCbAk4LSxO5L9dB6RCjPVybrHcRmb5wtJvBeHdIcKdTQdnDPsi6w4SziRlgtcw+aec2+v cloudbreak"
        ]
    ]
}
MSG:
Environment exists and differs from expected:
Parameter 'public_key_text' found to be 'ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQDZkgFHHkJltjGKOM93SwPQ+qM16zBh7rXLdgvwGhoLUDIqJgtqTBW03i8w8q4LxjZLTCV2LEiLTQIog+CcuzOPK9XWadzCW1LkovlZ5nG4Me+F0aN8I9jupYvMe1f8+xk4ZAKHLeI14gWmti+fJkTN4q1cK8cX0gzG1pEPXMdYtStkjVBk65z/VsqmqjinoQnuDSaa6Xd16dXMbbID9y2qRgAM91MWGKS0DGyH0YTSjkid3ZADQKgUblYdMRNSoaktSKhV68BLUtuWM0MfCbAk4LSxO5L9dB6RCjPVybrHcRmb5wtJvBeHdIcKdTQdnDPsi6w4SziRlgtcw+aec2+v cloudbreak'

[Chaffelson]

the public_key_text value looks to be identical to me, so it shouldn't be triggering this behaviour, but I also cannot reproduce it.
Are you still seeing the issue, and can you give me more details about the environment configuration that is triggering it?

[belardgx-dotcom]

Hi Chaffelson,

I was able to replicate the error. This happened when I tried updating our CDP environment by including new groups with its corresponding resource role assignment like MLUser, DEUser, ODUser, etc. Our environment is running with version 7.15. It is hosted in AWS and it comes with CDE, CML, and CDW. Any specific configuration that you are looking for that can help identify the issue? Thank you

[WARNING]: Environment tags specified. Currently, tags cannot be reconciled. To update tags, explicitly delete and recreate the environment. fatal: [localhost]: FAILED! => { "changed": false, "violations": [ [ "public_key_text", "ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQDZkgFHHkJltjGKOM93SwPQ+qM16zBh7rXLdgvwGhoLUDIqJgtqTBW03i8w8q4LxjZLTCV2LEiLTQIog+CcuzOPK9XWadzCW1LkovlZ5nG4Me+F0aN8I9jupYvMe1f8+xk4ZAKHLeI14gWmti+fJkTN4q1cK8cX0gzG1pEPXMdYtStkjVBk65z/VsqmqjinoQnuDSaa6Xd16dXMbbID9y2qRgAM91MWGKS0DGyH0YTSjkid3ZADQKgUblYdMRNSoaktSKhV68BLUtuWM0MfCbAk4LSxO5L9dB6RCjPVybrHcRmb5wtJvBeHdIcKdTQdnDPsi6w4SziRlgtcw+aec2+v cloudbreak" ] ] } MSG: Environment exists and differs from expected: Parameter 'public_key_text' found to be 'ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQDZkgFHHkJltjGKOM93SwPQ+qM16zBh7rXLdgvwGhoLUDIqJgtqTBW03i8w8q4LxjZLTCV2LEiLTQIog+CcuzOPK9XWadzCW1LkovlZ5nG4Me+F0aN8I9jupYvMe1f8+xk4ZAKHLeI14gWmti+fJkTN4q1cK8cX0gzG1pEPXMdYtStkjVBk65z/VsqmqjinoQnuDSaa6Xd16dXMbbID9y2qRgAM91MWGKS0DGyH0YTSjkid3ZADQKgUblYdMRNSoaktSKhV68BLUtuWM0MfCbAk4LSxO5L9dB6RCjPVybrHcRmb5wtJvBeHdIcKdTQdnDPsi6w4SziRlgtcw+aec2+v cloudbreak'

[Chaffelson]

Ah ok, so it's probably a different violation that is not being reported, so the public key text is actually probably a red herring.
Can you please share the invocation of the module you are using with the before and after change? I'll replicate that in our dev environment then look to generate a patch.

[belardgx-dotcom]

We are using the following module to execute updates on cred

- name: "Create credential {{ env_credential['credential_name'] }}"
  cloudera.cloud.**env_cred**:
    state: present
    cloud: aws
    name: "{{ env_credential['credential_name'] }}"
    description: "{{ env_credential['description'] }}"
    role: "{{ env_credential['role_arn'] }}"
    debug: true
    cp_region: "{{ cp_region }}"