如何实现Ansible Playbook错误逐层传递至Shell与Jenkins Groovy任务？

Absolutely! You can propagate Ansible errors all the way up to your Jenkins Groovy pipeline without any hassle—here's a step-by-step breakdown to make this work smoothly:

1. Ensure Ansible Returns a Non-Zero Exit Code on Failure #

Ansible already returns a non-zero exit code by default when a playbook fails, but double-check your playbook to make sure you aren't accidentally suppressing errors with settings like ignore_errors: yes or failed_when: false. These will force Ansible to return 0 even if tasks fail, breaking the error chain.

2. Capture Ansible Errors in Your Shell Script #

Modify your shell script to capture both the exit code and output of the Ansible command, then propagate the error if something goes wrong. Here's an example:

#!/bin/bash

# Execute the Ansible playbook, merging stderr into stdout to capture all output
ANSIBLE_FULL_OUTPUT=$(ansible-playbook -vvv -c local "route53Mapping.yml" --extra-vars '{"hostedZoneId":'${hostedZoneId}', "route53Url":'${route53Url}'}' 2>&1)
ANSIBLE_EXIT_CODE=$?

# If Ansible failed, print the full error output and exit with the same code
if [ $ANSIBLE_EXIT_CODE -ne 0 ]; then
    echo "ERROR: Ansible Playbook execution failed:"
    echo "$ANSIBLE_FULL_OUTPUT"
    exit $ANSIBLE_EXIT_CODE
fi

# If successful, proceed with any post-tasks (optional)
echo "Ansible Playbook executed successfully"

The 2>&1 ensures both standard output and error are captured, so you don't miss any critical debug info from the -vvv flag.

3. Catch the Shell Script Error in Jenkins Groovy #

In your Jenkins pipeline, use the sh step with returnStatus and returnStdout enabled to capture both the exit code and output of your shell script. Then, explicitly fail the pipeline if the script returns a non-zero code:

pipeline {
    agent any

    stages {
        stage('Run Route53 & SNS Deployment') {
            steps {
                script {
                    // Run the shell script and capture results
                    def deploymentResult = sh(
                        script: "./your-deployment-script.sh", // Replace with your script path
                        returnStdout: true,
                        returnStatus: true
                    )

                    // Check if the script failed
                    if (deploymentResult.status != 0) {
                        // Throw an error to mark the pipeline as failed and display the full error
                        error "Deployment failed with the following details:\n${deploymentResult.stdout}"
                    }
                }
            }
        }
    }
}

This will make Jenkins mark the stage (and pipeline) as failed, and the full error output from Ansible will be visible in the Jenkins console logs.

4. Optional: Improve Error Clarity in Ansible #

To make debugging easier, add explicit failure messages in your Ansible playbook using the fail module. For example:

- name: Verify Route53 record creation
  aws_route53:
    hosted_zone_id: "{{ hostedZoneId }}"
    record: "{{ route53Url }}"
    type: A
    state: present
  register: route53_result

- name: Fail if Route53 record creation failed
  fail:
    msg: "Failed to create Route53 record for {{ route53Url }} in zone {{ hostedZoneId }}. AWS error: {{ route53_result.msg }}"
  when: route53_result.failed

This ensures the error message includes context-specific details that will bubble up to Jenkins, making it faster to diagnose issues.

内容的提问来源于stack exchange，提问作者Monish Das