Ansible : MODULE FAILURE\nSee stdout/stderr for the exact error

Question

I am trying to install AWS CloudWatch agent in EC2 Centos VM using ansible-playbook. It worked well in the sandbox but when I ran it in Production it fails (integrated with Jenkins pipeline)

Here is my task snippet:

      - name: setup temp directory for install
        file:
          path: /tmp/aws-cw-agent
          state: directory

      - name: download installer
        get_url:
          url: "{{ aws_cw_agent_url }}"
          dest: /tmp/aws-cw-agent/amazon-cloudwatch-agent.rpm

      - name: install agent
        become: true
        shell: rpm -U /tmp/aws-cw-agent/amazon-cloudwatch-agent.rpm

Until "download installer" it works fine and I could find the rpm binary by manually navigating to the directory. But the next task install agent fails. Instead of shell, if I use "yum" module also, it fails.

The error says:

17:16:07 task path: /home/jenkins/workspace/groupesiph-dsir/03227/03227_Cloudwatch_Agent_deploy_hprod/playbook/deployment/roles/aws_cw_agent/tasks/main.yml:22

>17:16:07 Tuesday 10 March 2020  17:16:07 +0100 (0:00:00.098)       0:00:05.352 ********* 
17:16:08 Using module file /usr/lib/python3.6/site-packages/ansible/modules/commands/command.py
17:16:08 Pipelining is enabled.
17:16:08 <10.45.1.136> ESTABLISH SSH CONNECTION FOR USER: ansible

>17:16:08 <10.45.1.136> SSH: EXEC sshpass -d10 ssh -C -o ControlMaster=auto -o ControlPersist=60s -o StrictHostKeyChecking=no -o 'User="ansible"' -o ConnectTimeout=10 -o ServerAliveInterval=60 -o ServerAliveCountMax=10 -o ControlPath=/home/jenkins/.ansible/cp/84b84369b7 10.45.1.136 '/bin/sh -c '"'"'sudo -H -S -n  -u root /bin/sh -c '"'"'"'"'"'"'"'"'echo BECOME-SUCCESS-syqwibhfpdecwpfqddhe ; /usr/bin/python'"'"'"'' && sleep 0'"'"''

>17:16:08 Escalation succeeded
17:16:08 <10.45.1.136> (1, b'', b'')
17:16:08 <10.45.1.136> Failed to connect to the host via ssh: 
17:16:08 fatal: [prod05]: FAILED! => {
17:16:08     "changed": false,
17:16:08     "module_stderr": "",
17:16:08     "module_stdout": "",
17:16:08     "msg": "MODULE FAILURE\nSee stdout/stderr for the exact error",
17:16:08     "rc": 1
17:16:08 

But if it is a real problem with ssh, then how it is succeeding in the previous task to download the installer.

What could be the problem?

Thanks in advance.

Solution

The problem was user did not have sudoers access to install binary which using yum to install "/tmp/aws-cw-agent/amazon-cloudwatch-agent.rpm". With debug message I found the error.

Comments

[mdaniel]

Those two tasks (the get_url and shell) differ in that one has become and the other not -- so I would guess that your sshpass is misconfigured; you can run ansible in debug mode as env ANSIBLE_DEBUG=1 ansible -vvvv and it will emit all kinds of helpful debug messaging, including why the module really failed

[Vorsprung]

post an answer to your own question, acknowledge @mdaniel and explain how you fixed it

[ARINDAM BANERJEE]

I tried something as well before posting it, but did not work.