New module 'efs_tag'

[zeten30]

SUMMARY

New module - efs_tag

ISSUE TYPE

• New Module Pull Request

COMPONENT NAME

efs_tag

ADDITIONAL INFORMATION

There's a similar 'ec2_tag' module amazon.aws.ec2_tag that allows tagging on various EC2 resources. I'm adding a module that allows all EFS resource tagging (both filesystem and access point)

- name: Ensure tags are present on a resource
  community.aws.efs_tag:
    resource: fsap-123456ab
    state: present
    tags:
      Name: MyEFSAccessPoint
      Env: Production

[ansibullbot]

cc @akazakov @jillr @ryansydnor @s-hertel @tremble @wimnat
click here for bot help

[zeten30]

@goneri Hi, any idea why checks are failing? It looks like a check problem, not the new module code.

The task includes an option with an undefined variable. The error was: '_hashed_targets' is undefined

The error appears to be in '/var/lib/zuul/builds/e8ae5b2a68be41ca958f877439a2b3e3/untrusted/project_0/github.com/ansible/ansible-zuul-jobs/roles/ansible-test/tasks/split_targets.yaml': line 20, column 3, but may
be elsewhere in the file depending on the exact syntax problem.

The offending line appears to be:

  loop: '{{ ansible_test_targets.stdout_lines }}'
- set_fact:
  ^ here

[goneri]

@goneri Hi, any idea why checks are failing? It looks like a check problem, not the new module code.

The task includes an option with an undefined variable. The error was: '_hashed_targets' is undefined

The error appears to be in '/var/lib/zuul/builds/e8ae5b2a68be41ca958f877439a2b3e3/untrusted/project_0/github.com/ansible/ansible-zuul-jobs/roles/ansible-test/tasks/split_targets.yaml': line 20, column 3, but may
be elsewhere in the file depending on the exact syntax problem.

The offending line appears to be:

  loop: '{{ ansible_test_targets.stdout_lines }}'
- set_fact:
  ^ here

Hi @zeten30, Sorry for the situation. This should not happen anymore now.

[goneri]

recheck

[tremble]

Thanks for the PR @zeten30

[ansibullbot]

cc @markuman
click here for bot help

[tremble]

Tags look good. You also need to explicitly enable the AWS Retries when you make the client calls. You're actually passed a decorated copy of the boto3 client where most calls have an extra "aws_retry" parameter.

[markuman]

Thanks for your work. Looks great so far.
You need also a changelog/fragments file.

[zeten30]

Making as WIP again. I'm going to create a complete set of integration tests, remove module from 'unsupported' and apply all suggested fixes.

[zeten30]

BTW: Do you know why ansible/gate Expected — Waiting for status to be reported is hanging in this state more that a day?

[markuman]

BTW: Do you know why ansible/gate Expected — Waiting for status to be reported is hanging in this state more that a day?

A PR needs two reviewer that confirms the PR.
...and the ansible/gate is waiting for it before the PR is merging it automagically.

[zeten30]

Getting ready for the integration tests - mattclay/aws-terminator#157

I assume the next check will fail:) We're missing some EFS: actions ^.

[zeten30]

All the EFS related problems should be fixed. Integration test are failing on unrelated module:

ERROR: The 1 integration test(s) listed below (out of 20) failed. See error output above for details:
ec2_vpc_endpoint_service_info
Cleaning up temporary python directory: /tmp/python-01sgnf9j-ansible

Can we merge this? Thanks in advance.

[zeten30]

recheck

[zeten30]

And another error: ERROR: Command "ansible-playbook ec2_vpc_endpoint_service_info-ox_pgtf5.yml -i ../../../../../../../venv/lib/python3.6/site-packages/ansible_test/_data/inventory --skip-tags False --diff -vvvvvv" returned exit status 4.

@goneri Any idea what is wrong here? Thank you very much in advance.

[markuman]

It looks like all last pipeline runs are affected.
But I've no idea atm. Locally the efs integrationtest runs successful for me.

ansible-test integration -v --python 3.7 efs --docker
...
localhost                  : ok=58   changed=16   unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   

[markuman]

recheck

[zeten30]

It looks like all last pipeline runs are affected.
But I've no idea atm. Locally the efs integrationtest runs successful for me.

ansible-test integration -v --python 3.7 efs --docker
...
localhost                  : ok=58   changed=16   unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   

yes, EFS works. This is now the problem:

ERROR: The 1 integration test(s) listed below (out of 9) failed. See error output above for details:
ec2_vpc_endpoint_service_info

And test timeout sometimes. Timeout needs to be increased or split the test even more.

[markuman]

recheck

[markuman]

recheck

[markuman]

recheck

[markuman]

recheck

[zeten30]

Really weird. Failed with:

botocore.exceptions.ClientError: An error occurred (ValidationException) when calling the DescribeFileSystems operation: 1 validation error detected: Value 'ansible-test-44911258-fedora-34-1vcpu-vexxhost-sjc1-0001719877-efs' at 'creationToken' failed to satisfy constraint: Member must have length less than or equal to 64

Playbook: tests/integration/targets/efs/playbooks/version_fail.yml

- hosts: localhost
  connection: local
  environment: "{{ ansible_test.environment }}"
  vars:
     resource_prefix: 'test-'

  tasks:
    - name: 'efs graceful failure tests'
      collections:
        - amazon.aws
      module_defaults:
        group/aws:
          aws_access_key: '{{ aws_access_key }}'
          aws_secret_key: '{{ aws_secret_key }}'
          security_token: '{{ security_token | default(omit) }}'
          region: '{{ aws_region }}'
      block:

        - name: create efs with provisioned_throughput options (fails gracefully)
          efs:
            state: present
            name: "{{ resource_prefix }}-efs"
            throughput_mode: 'provisioned'
            provisioned_throughput_in_mibps: 8.0
          register: efs_provisioned_throughput_creation
          ignore_errors: yes

        - name: check that graceful error message is returned when creation with throughput_mode and old botocore
          assert:
            that:
              - efs_provisioned_throughput_creation.failed
              - 'efs_provisioned_throughput_creation.msg == "throughput_mode parameter requires botocore >= 1.10.57"'

[markuman]

recheck

[markuman]

New day new random errors?

"assertion": "efs_provisioned_throughput_creation.msg == \"throughput_mode parameter requires botocore >= 1.10.57\"",

That's also not affected by this PR

[markuman]

recheck

[zeten30]

@tremble 👍 thx for help with integration tests cleanup.
@markuman I hope that the last commit will finally solve our CI issues.

[tremble]

test failure is unrelated.

#675 is running through gating and will likely generate a merge conflict I'll rebase this for you once the gating completes

[zeten30]

recheck

[zeten30]

\o/ - a successful ansible/check finally

[markuman]

lgtm

[ansible-zuul]

LGTM!

[tremble]

recheck