opsreview

Compile a report of recent PagerDuty alerts for a given list of escalation policies.

Purpose

To keep the on-call duty tolerable, it's important to regularly review recent alerts and determine what actions need to be taken. That might be fixing the originating issue, or modifying your alerting behaviors.

Using the opsreview tool, you can quickly compile a report on recent alerts for a single escalation policy and determine what actions should be taken.

Example Output

> python pull_alerts.py

Time: Tuesday, January 19 - 9:19 AM
Service: api
Description: Failing API alarm: CRON failure
URL: https://subdomain.pagerduty.com/incidents/ABC123
Notes:
['bob@lyft.com: #a - Re-ran CRON']

Time: Tuesday, January 19 - 12:03 PM
Service: api
Description: Failing API alarm: CPU percent > 10
URL: https://subdomain.pagerduty.com/incidents/ABC124
Notes:
['bob@lyft.com: #na - Temporary spike in CPU, no action taken. Alert is too sensitive.']

Time: Tuesday, January 19 - 1:35 PM
Service: www
Description: Failing WWW alarm: 5XX percent > 5
URL: https://subdomain.pagerduty.com/incidents/ABC125
Notes:
['sally@lyft.com: #a - Bad deploy']

Setup

Update your settings. To get an API token, go to the "User Settings" tab on your PagerDuty profile and click "Create API User Token". Make sure to use a v2 token.

cp settings_example.py settings.py

# Update settings.py for your PagerDuty escalation policy.
#
# PAGERDUTY_API_TOKEN = 'yourapitoken'
# ESCALATION_POLICIES = ['PYODVQB']  # Get from https://yoursubdomain.pagerduty.com/escalation_policies#PYODVQB

Set up your environment.

pip3 install virtualenv
virtualenv venv3
source venv3/bin/activate
pip install -r requirements.txt

Run the script.

python pull_alerts.py