Chapter 74. On-Call Health: The Metric You Could Be Measuring
Caitie McCaffrey
A sad trombone noise emanated from my phone one Saturday afternoon. The services I was on call for requested attention, yet again. It was the end of my first week on call for a new team, and I had already been paged at least 50 times. I was bleary-eyed and anxious. As I acknowledged the alert I had a strong urge to throw my phone into an adjacent brick wall. Instead I took a deep breath and scrolled through the alert details while firing up my laptop yet again.
We define SLIs, SLOs, and SLAs for service health. We measure availability and reliability, run postmortems that focus on customer impact, and implement health checks for services to detect failures quickly. As an industry, we’re quick to know whether our services are healthy, but we overlook a crucial component of running a successful service: whether the people on call are as healthy.
Are they sleeping through the night? How often are they being paged outside of business hours? Does the work of being on call fit within a reasonable workweek? Fortunately, we can use several of the tools and best practices for monitoring a service’s health to monitor this other critical component.
- Metrics to measure
- To understand service health, we define SLIs; for measuring on-call health, we need similar metrics. Note the number of alerts per week, ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access