Automated Diagnostics

Sending Automation Output to PagerDuty Incidents

Introduction

By default, your automated diagnostics jobs will display their results in logging visible in the Automation interface. There are a few extra steps necessary to transport those logs or some subset over to your PagerDuty incidents. This section will walk you through steps to simplify output in a job and send it to a PagerDuty incident. These steps assume that you have already completed Tour 4: Send Diagnostic Data to PagerDuty, which ensures that you have connected your Automation interface to your PagerDuty instance at the infrastructure level.

Example

To carry out these steps, you’ll need a job that we can modify for our purposes. Since most of the jobs in this Solution already have these steps completed, we’ll start with a fresh job which you can import and work with. You can copy the job definition at the bottom of the page.

1. In the Jobs section of your project, click New Job
   
   
2. Click Upload Definition
   
   
3. Click Choose File to select the downloaded file
   
   
4. Select YAML format
5. Click Upload
6. Edit your job using the Actions menu at the far right of the jobs list.

Add an option for Incident ID

1. Click Workflow at the top of the Job Edit page
2. Toward the top of the Workflow section, click Add an option.
   
3. Enter an Option Name and Option Label as specified in the screenshot.
4. Click Save for this option

Add a log filter to your job step

1. Using the gear icon at the right of the job step, click Add Log Filter
2. Choose Progress Badge from the available filters
   

Setting your Filter

1. Set the following fields:
   Regex: .*Active: active.*
   Text: Healthy
   Status Symbol: Select the Green Checkmark
   Context Variable: ssh-status
   
   
   
2. Save your filter. You should see the log filter visually now as part of your job step.
   
   
3. Click the +Add button to add another Progress Badge filter
4. Set the following fields:
   Regex:!.*Active: active.*
   Text: Unhealthy
   Status Symbol: Select the Red X
   Context Variable: ssh-status
   

Add an Incident Note job step

1. On the Edit Job page, click Add a step.
   
   
   
2. Under Workflow Steps, search for and select PagerDuty Incident Note
   

Configure the Incident Note job step

1. Enter or select a PagerDuty API Key
   If in keystore, use Select button. Otherwise, you can paste the API Key in this field
2. Enter an Email to be used for incident updates. If you’ve established a project variable, you can use that as specified in screenshot
3. Update the Incident ID field as specified in screenshot to use the pd_incident_id variable
4. Update the Note field to include variable reference (${data.variablename}) as shown in screenshot.

   SSH Status: ${data.ssh-status*}\n\nClick here for detailed diagnostics: ${job.url}#output

   This note will show the resulting progress badge as well as a link to the full log
5. Set a Step Label to identify what this job step does
6. Save the job step and job to save your changes
   

Job Definition

Copy and paste this content into a file and import to your project.

yaml

- defaultTab: nodes
  description: ''
  executionEnabled: true
  group: Administration/Linux
  id: e4b9f6db-efda-43a4-8f10-52eba1b6c56d
  loglevel: INFO
  name: SSH Status
  nodeFilterEditable: true
  nodefilters:
    dispatch:
      excludePrecedence: true
      keepgoing: false
      rankOrder: ascending
      successOnEmptyNodeFilter: false
      threadcount: '1'
    filter: 'name: employees_mysql_db_1 '
  nodesSelectedByDefault: true
  plugins:
    ExecutionLifecycle: {}
  scheduleEnabled: true
  schedules: []
  sequence:
    commands:
    - description: Check status of SSH service
      exec: sudo systemctl status ssh
    keepgoing: false
    strategy: node-first
  uuid: e4b9f6db-efda-43a4-8f10-52eba1b6c56d