ansible-collection-netscaleradc icon indicating copy to clipboard operation
ansible-collection-netscaleradc copied to clipboard
verified publisher icon netscaler

[BUG] netscaler.adc.snmptrap is failing

Open boleslawlucjanek opened this issue 2 years ago • 1 comments

Describe the bug Task using netscaler.adc.snmptrap is failing if there is already snmp trap bind command configured for the same below parameters:

  • trapdestination

  • trapclass

Example: If there is already in config: add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv

Task using netscaler.adc.snmptrap is failing in such scenario with bellow error: "msg": "ERROR: Found more than one resource with the same primary key generic and get arguments {'trapdestination': '5.5.5.5', 'version': 'V3'}"}

If there is no snmp trap bind command configured for the same trapdestination and trapclass, then task using netscaler.adc.snmptrap works as expected:

  • is adding snmp trap if it is not configured (task key "changed " is true)
  • is doing nothing if snmp trap is already configured for the same trapdestination and trapclass

To Reproduce Steps to reproduce the behaviour:

  1. My ansible-playbook is... Note: You need to configure snmp group and snmp user before using this playbook: image

tasks:

- name: V2 | Sample Task | login
  delegate_to: localhost
  register: login_result
  netscaler.adc.login:
    nsip: "{{ ansible_host }}"
    validate_certs: no
    username: "{{ nitro_user }}"
    password: "{{ nitro_pass }}"

- name: Sample Task | snmptrap
  delegate_to: localhost
  netscaler.adc.snmptrap:
    nitro_auth_token: "{{ login_result.sessionid }}"
    validate_certs: no
    nsip: "{{ ansible_host }}"
    state: present
    trapclass: generic
    trapdestination: 5.5.5.5
    version: V3
    allpartitions: ENABLED   

- name: Sample Task | snmptrap_snmpuser_binding
  delegate_to: localhost
  netscaler.adc.snmptrap_snmpuser_binding:
    nitro_auth_token: "{{ login_result.sessionid }}"
    validate_certs: no
    nsip: "{{ ansible_host }}"
    state: present
    username: xxx_user
    securitylevel : authPriv
    trapclass: generic
    trapdestination: 5.5.5.5
    version: V3  

- name: V2 | Sample Task | logout
  delegate_to: localhost
  netscaler.adc.logout:
    nitro_auth_token: "{{ login_result.sessionid }}"
    nsip: "{{ ansible_host }}"
    validate_certs: no

In first iteration playbook finishes with success as snmp trap is being added and binded to the snmp user. Tasks using modules netscaler.adc.snmptrap and netscaler.adc.snmptrap_snmpuser_binding have response key "changed " as true:

PLAY [baseline] ************************************************************************************************************************

TASK [V2 | Sample Task | login] ******************************************************************************************************** changed: [test_LB -> localhost]

TASK [Sample Task | snmptrap] ********************************************************************************************************** changed: [test_LB -> localhost]

TASK [Sample Task | snmptrap_snmpuser_binding] ***************************************************************************************** changed: [test_LB -> localhost]

TASK [V2 | Sample Task | logout] ******************************************************************************************************* changed: [test_LB -> localhost]

PLAY RECAP ***************************************************************************************************************************** test_LB : ok=4 changed=4 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0

In cli configured commands are seen:

show ns runningConfig | grep trap add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv

But during next playbook execution playbook if failing on task "Sample Task | snmptrap": PLAY [baseline] ************************************************************************************************************************

TASK [V2 | Sample Task | login] ******************************************************************************************************** changed: [test_LB -> localhost]

TASK [Sample Task | snmptrap] ********************************************************************************************************** fatal: [test_LB -> localhost]: FAILED! => {"changed": false, "loglines": ["DEBUG: Initializing ModuleExecutor for resource snmptrap", "TRACE: ENTRY: get_valid_desired_states() called with ('snmptrap',), {}", "TRACE: EXIT: get_valid_desired_states() returned {'absent', 'present'}", "TRACE: ENTRY: get_netscaler_version() called with (<ansible_collections.netscaler.adc.plugins.module_utils.client.NitroAPIClient object at 0x7f83eedfefd0>,), {}", "TRACE: ENTRY: get_resource() called with (<ansible_collections.netscaler.adc.plugins.module_utils.client.NitroAPIClient object at 0x7f83eedfefd0>, 'nsversion'), {}", "TRACE: ENTRY: send() called with (<ansible_collections.netscaler.adc.plugins.module_utils.client.NitroAPIClient object at 0x7f83eedfefd0>,......................................................... ................................................................................................................ "msg": "ERROR: Found more than one resource with the same primary key generic and get arguments {'trapdestination': '5.5.5.5', 'version': 'V3'}"}

PLAY RECAP ***************************************************************************************************************************** test_LB : ok=1 changed=1 unreachable=0 failed=1 skipped=0 rescued=0 ignored=0

But if below command is removed from configuration: bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv

show ns runningConfig | grep trap add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv unbind snmp trap generic 5.5.5.5 -userName xxx_user Done show ns runningConfig | grep trap add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED

then playbook will finish with success and task "Sample Task | snmptrap" that is using module netscaler.adc.snmptrap has response key "changed " as false because there is already snmp trap configured with the same trapclass and trapdestination:

PLAY [baseline] ************************************************************************************************************************

TASK [V2 | Sample Task | login] ******************************************************************************************************** changed: [test_LB -> localhost]

TASK [Sample Task | snmptrap] ********************************************************************************************************** ok: [test_LB -> localhost]

TASK [Sample Task | snmptrap_snmpuser_binding] ***************************************************************************************** changed: [test_LB -> localhost]

TASK [V2 | Sample Task | logout] ******************************************************************************************************* changed: [test_LB -> localhost]

PLAY RECAP ***************************************************************************************************************************** test_LB : ok=4 changed=3 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0

In cli again will be seen config with snmp bind command:

show ns runningConfig | grep trap add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv

And if playbook would be executed again, it would fail again with below error: "msg": "ERROR: Found more than one resource with the same primary key generic and get arguments {'trapdestination': '5.5.5.5', 'version': 'V3'}"}

The ansible-playbook command I executed is... ansible-playbook --limit test_LB baseline.yml

The ansible-playbook logs are... PLAY [baseline] ************************************************************************************************************************

TASK [V2 | Sample Task | login] ******************************************************************************************************** changed: [test_LB -> localhost]

TASK [Sample Task | snmptrap] ********************************************************************************************************** fatal: [test_LB -> localhost]: FAILED! => {"changed": false, "loglines": ["DEBUG: Initializing ModuleExecutor for resource snmptrap", "TRACE: ENTRY: get_valid_desired_states() called with ('snmptrap',), {}", "TRACE: EXIT: get_valid_desired_states() returned {'absent', 'present'}", "TRACE: ENTRY: get_netscaler_version() called with (<ansible_collections.netscaler.adc.plugins.module_utils.client.NitroAPIClient object at 0x7f83eedfefd0>,), {}", "TRACE: ENTRY: get_resource() called with (<ansible_collections.netscaler.adc.plugins.module_utils.client.NitroAPIClient object at 0x7f83eedfefd0>, 'nsversion'), {}", "TRACE: ENTRY: send() called with (<ansible_collections.netscaler.adc.plugins.module_utils.client.NitroAPIClient object at 0x7f83eedfefd0>,......................................................... ................................................................................................................ "msg": "ERROR: Found more than one resource with the same primary key generic and get arguments {'trapdestination': '5.5.5.5', 'version': 'V3'}"}

PLAY RECAP ***************************************************************************************************************************** test_LB : ok=1 changed=1 unreachable=0 failed=1 skipped=0 rescued=0 ignored=0

I am seeing the below error... "msg": "ERROR: Found more than one resource with the same primary key generic and get arguments {'trapdestination': '5.5.5.5', 'version': 'V3'}"}

Please specify the equivalent NSCLI command, if you were to manually configure via CLI add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv

Expected behaviour I expect that task "Sample Task | snmptrap" that is using module netscaler.adc.snmptrap will finish with success with response key "changed " as false if there is already snmp trap configured with the same trapclass and trapdestination. I expect below ansible logs: PLAY [baseline] ************************************************************************************************************************

TASK [V2 | Sample Task | login] ******************************************************************************************************** changed: [test_LB -> localhost]

TASK [Sample Task | snmptrap] ********************************************************************************************************** ok: [test_LB -> localhost]

TASK [Sample Task | snmptrap_snmpuser_binding] ***************************************************************************************** ok: [test_LB -> localhost]

TASK [V2 | Sample Task | logout] ******************************************************************************************************* changed: [test_LB -> localhost]

PLAY RECAP ***************************************************************************************************************************** test_LB : ok=4 changed=3 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0

Environment (please complete the following information):

  • Python version python --version output Python 3.6.8
  • Ansible version ansible --version output ansible [core 2.11.12]
  • NetScaler version show ns version output NetScaler NS13.1: Build 42.47.nc, Date: Feb 22 2023, 08:08:13 (64-bit)

Additional context There is as well problem with module "netscaler.adc.snmptrap_snmpuser_binding" as I have tested execution of only one task with that module and it is always giving response with key "changed" as true, even if command is already in configuration and there was no modification. Below is example:

CLI config:

show ns runningConfig | grep trap add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv

Playbook yaml:

tasks:

- name: V2 | Sample Task | login
  delegate_to: localhost
  register: login_result
  netscaler.adc.login:
    nsip: "{{ ansible_host }}"
    validate_certs: no
    username: "{{ nitro_user }}"
    password: "{{ nitro_pass }}"

# - name: Sample Task | snmptrap
#   delegate_to: localhost
#   netscaler.adc.snmptrap:
#     nitro_auth_token: "{{ login_result.sessionid }}"
#     validate_certs: no
#     nsip: "{{ ansible_host }}"
#     state: present
#     trapclass: generic
#     trapdestination: 5.5.5.5
#     version: V3
#     allpartitions: ENABLED   

- name: Sample Task | snmptrap_snmpuser_binding
  delegate_to: localhost
  netscaler.adc.snmptrap_snmpuser_binding:
    nitro_auth_token: "{{ login_result.sessionid }}"
    validate_certs: no
    nsip: "{{ ansible_host }}"
    state: present
    username: xxx_user
    securitylevel : authPriv
    trapclass: generic
    trapdestination: 5.5.5.5
    version: V3  

- name: V2 | Sample Task | logout
  delegate_to: localhost
  netscaler.adc.logout:
    nitro_auth_token: "{{ login_result.sessionid }}"
    nsip: "{{ ansible_host }}"
    validate_certs: no

Playbook output: PLAY [baseline] ************************************************************************************************************************

TASK [V2 | Sample Task | login] ******************************************************************************************************** changed: [test_LB -> localhost]

TASK [Sample Task | snmptrap_snmpuser_binding] ***************************************************************************************** changed: [test_LB -> localhost]

TASK [V2 | Sample Task | logout] ******************************************************************************************************* changed: [test_LB -> localhost]

PLAY RECAP ***************************************************************************************************************************** test_LB : ok=3 changed=3 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0

So there was no config modification, but task "Sample Task | snmptrap_snmpuser_binding" is seen as "changed":

show ns runningConfig | grep trap add snmp trap generic 5.5.5.5 -version V3 -allPartitions ENABLED bind snmp trap generic 5.5.5.5 -userName xxx_user -securityLevel authPriv

boleslawlucjanek avatar Sep 28 '23 06:09 boleslawlucjanek

The NITRO API is responding with two objects in the GET response for the same resource. I am checking with the internal NITRO team on this.

sumanth-lingappa avatar Dec 17 '23 14:12 sumanth-lingappa