community.general.pacemaker_resource: resource creation fails in cluster maintenance mode

[eshirke]

Summary

Hi @munchtoast,

I’ve encountered a similar issue as previously reported in #10180, now with community.general version 11.0.0, when attempting to create a HAProxy health check resource while the cluster is in maintenance mode.

Issue Type

Bug Report

Component Name

pacemaker_resource

Ansible Version

ansible [core 2.18.4]

Community.general Version

community.general 11.0.0

Configuration

$ ansible-config dump --only-changed

OS / Environment

RHEL 9.x

Steps to Reproduce

Ansible Code:

• name: pcs_resource_healthcheck | Create health check resource
  community.general.pacemaker_resource:
  name: "{{ hana_sid | upper }}_healthcheck"
  resource_type:
  resource_name: service:haproxy
  resource_operation:
• operation_action: monitor
  operation_option:
• "interval=10s"
• "timeout=20s"
  state: present
  notify: cluster_setup_pause

Expected Results

resource should get created without any error

Actual Results

Playbook Error:

TASK [clustering_setup_hana : pcs_resource_healthcheck | Create health check resource] **********************************************************************************
skipping: [ABC2]
An exception occurred during task execution. To see the full traceback, use -vvv. The error was: ansible_collections.community.general.plugins.module_utils.mh.exceptions.ModuleHelperException
fatal: [ABC1]: FAILED! => {"changed": false, "msg": "pcs failed with error (rc=1): Waiting for the cluster to apply configuration changes (timeout: 300 seconds)...\nError: resource 'HGC_healthcheck' is not running on any node\nError: Errors have occurred, therefore pcs is unable to continue\n", "output": {"previous_value": null, "value": null}, "previous_value": null, "value": null, "vars": {"previous_value": null, "value": null}}

Full Trace:

TASK [clustering_setup_hana : pcs_resource_healthcheck | Create health check resource] **********************************************************************************
The full traceback is:
Traceback (most recent call last):
File "/tmp/ansible_community.general.pacemaker_resource_payload_tei4a_ek/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/deco.py", line 56, in wrapper
func(self, *args, **kwargs)
File "/tmp/ansible_community.general.pacemaker_resource_payload_tei4a_ek/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/base.py", line 93, in run
self.run()
File "/tmp/ansible_community.general.pacemaker_resource_payload_tei4a_ek/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/mixins/state.py", line 37, in run
return func()
^^^^^^
File "/tmp/ansible_community.general.pacemaker_resource_payload_tei4a_ek/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/modules/pacemaker_resource.py", line 201, in state_present
File "/tmp/ansible_community.general.pacemaker_resource_payload_tei4a_ek/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/cmd_runner.py", line 223, in run
self.results_processed = self.output_process(*results)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/ansible_community.general.pacemaker_resource_payload_tei4a_ek/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/modules/pacemaker_resource.py", line 177, in process
File "/tmp/ansible_community.general.pacemaker_resource_payload_tei4a_ek/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/base.py", line 53, in do_raise
raise _MHE(*args, **kwargs)
ansible_collections.community.general.plugins.module_utils.mh.exceptions.ModuleHelperException
fatal: [ABC1]: FAILED! => {
"changed": false,
"invocation": {
"module_args": {
"name": "HGC_healthcheck",
"resource_argument": null,
"resource_meta": null,
"resource_operation": [
{
"operation_action": "monitor",
"operation_option": [
"interval=10s",
"timeout=20s"
]
}
],
"resource_option": [],
"resource_type": {
"resource_name": "service:haproxy",
"resource_provider": null,
"resource_standard": null
},
"state": "present",
"wait": 300
}
},
"msg": "pcs failed with error (rc=1): Waiting for the cluster to apply configuration changes (timeout: 300 seconds)...\nError: resource 'HGC_healthcheck' is not running on any node\nError: Errors have occurred, therefore pcs is unable to continue\n",
"output": {
"previous_value": null,
"value": null
},
"previous_value": null,
"value": null,
"vars": {
"previous_value": null,
"value": null
}
}

Full Trace After Rerun:

TASK [clustering_setup_hana : pcs_resource_healthcheck | Create health check resource] **********************************************************************************
ok: [ABC1] => {
"changed": false,
"cmd": [
"/sbin/pcs",
"resource",
"create",
"HGC_healthcheck",
"service:haproxy",
"op",
"monitor",
"interval=10s",
"timeout=20s",
"--wait=300"
],
"invocation": {
"module_args": {
"name": "HGC_healthcheck",
"resource_argument": null,
"resource_meta": null,
"resource_operation": [
{
"operation_action": "monitor",
"operation_option": [
"interval=10s",
"timeout=20s"
]
}
],
"resource_option": [],
"resource_type": {
"resource_name": "service:haproxy",
"resource_provider": null,
"resource_standard": null
},
"state": "present",
"wait": 300
}
},
"previous_value": " * HGC_healthcheck\t(service:haproxy):\t Stopped (unmanaged)",
"stderr": "Error: 'HGC_healthcheck' already exists\n",
"stderr_lines": [
"Error: 'HGC_healthcheck' already exists"
],
"stdout": "",
"stdout_lines": [],
"value": " * HGC_healthcheck\t(service:haproxy):\t Stopped (unmanaged)"
}

Code of Conduct

• I agree to follow the Ansible Code of Conduct

[ansibullbot]

Files identified in the description:

• plugins/modules/pacemaker_resource.py

If these files are incorrect, please update the component name section of the description or use the !component bot command.

click here for bot help

[eshirke]

@munchtoast

As you suggested, I have tried with the option as below. I got an error as below.

Ansible Code:

    - name: pcs_resource_healthcheck | Create health check resource
      community.general.pacemaker_resource:
        name: "{{ hana_sid | upper }}_healthcheck"
        resource_type:
          resource_name: haproxy
          resource_standard: service
        resource_operation:
          - operation_action: monitor
            operation_option:
              - "interval=10s"
              - "timeout=20s"
        state: present
      notify: cluster_setup_pause

Error Details:

The full traceback is:
Traceback (most recent call last):
  File "/tmp/ansible_community.general.pacemaker_resource_payload_klpqeczx/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/deco.py", line 56, in wrapper
    func(self, *args, **kwargs)
  File "/tmp/ansible_community.general.pacemaker_resource_payload_klpqeczx/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/base.py", line 93, in run
    self.__run__()
  File "/tmp/ansible_community.general.pacemaker_resource_payload_klpqeczx/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/mixins/state.py", line 37, in __run__
    return func()
           ^^^^^^
  File "/tmp/ansible_community.general.pacemaker_resource_payload_klpqeczx/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/modules/pacemaker_resource.py", line 201, in state_present
  File "/tmp/ansible_community.general.pacemaker_resource_payload_klpqeczx/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/cmd_runner.py", line 223, in run
    self.results_processed = self.output_process(*results)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/tmp/ansible_community.general.pacemaker_resource_payload_klpqeczx/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/modules/pacemaker_resource.py", line 177, in process
  File "/tmp/ansible_community.general.pacemaker_resource_payload_klpqeczx/ansible_community.general.pacemaker_resource_payload.zip/ansible_collections/community/general/plugins/module_utils/mh/base.py", line 53, in do_raise
    raise _MHE(*args, **kwargs)
ansible_collections.community.general.plugins.module_utils.mh.exceptions.ModuleHelperException
fatal: [ABC1]: FAILED! => {
    "changed": false,
    "invocation": {
        "module_args": {
            "name": "HGC_healthcheck",
            "resource_argument": null,
            "resource_meta": null,
            "resource_operation": [
                {
                    "operation_action": "monitor",
                    "operation_option": [
                        "interval=10s",
                        "timeout=20s"
                    ]
                }
            ],
            "resource_option": [],
            "resource_type": {
                "resource_name": "haproxy",
                "resource_provider": null,
                "resource_standard": "service"
            },
            "state": "present",
            "wait": 300
        }
    },
    "msg": "pcs failed with error (rc=1): Error: missing value of 'haproxy' option\n",
    "output": {
        "previous_value": null,
        "value": null
    },
    "previous_value": null,
    "value": null,
    "vars": {
        "previous_value": null,
        "value": null
    }
}

[eshirke]

@munchtoast

Hope you're doing well. Just wanted to follow up and check if you had a chance to look into replicating the issue.
Please let me know if you need any additional information from my side.

Thanks,
Eknath

[munchtoast]

@munchtoast

Hope you're doing well. Just wanted to follow up and check if you had a chance to look into replicating the issue. Please let me know if you need any additional information from my side.

Thanks, Eknath

Hi @eshirke,

Thank you for your patience. I apologize for the delay in getting back to you, as I’ve been stretched thin lately... I’ll submit a pull request as soon as I’m able to carve out time to work on this issue. There are two items that need to be addressed for this issue: maintenance resource creation and bugged provider/standard/name resource name.

[munchtoast]

Hi @eshirke , once again, I'm sorry for my delay in response.

I am not able to reproduce this error as I am not sure on how to install service:haproxy. Instead, I tested on my own cluster (in maintenance mode, using 11.2.0 release) with another resource type - and that test succeeded on my end even though the cluster was in maintenance mode. I noticed in the output provided on the issue:

"stderr": "Error: 'HGC_healthcheck' already exists\n",
"stderr_lines": [
"Error: 'HGC_healthcheck' already exists"
],

Which seems to be displaying some un-idempotent behavior. The module should return [ok] instead. Could you please verify if this is the correct release?