How to Survive Your First On-Call Rotation as a Software Engineer

Brian Jenney
5 min readAug 5, 2020
Me on my first on-call rotation — “Everything is fine, everything is fine”

I had been a software engineer at my last company for about a month when my manager let me know I’d be going on-call during the next rotation. “Cool” I replied. Not cool.

I installed PagerDuty, a popular app that is as necessary as it is hated, to alert me when any critical ticket was created. I had casually been keeping tabs on the kinds of bugs that were deemed critical to see what I could expect during my shift. Basically, anything that prevented our customers from normal usage of any one of our many applications would trigger an alert. Button to export a CSV is broken? Critical ticket. API returning 500’s on a Saturday at 2 a.m.? Extra critical ticket.

Now, we had a comprehensive test suite, incredibly high test coverage and a kick ass QA team but I had yet to see a week go by without at least two critical escalations. At my previous companies, critical tickets were few and far between, and as such, there wasn’t much formality around how to handle these show-stopping bugs. At one company, the rotation would change every day and on the weekends, if anything blew up, we would play the game of who is going to acknowledge this first. Not it.

Luckily, my new company had some decent protocols in place for handling on-call rotation, though this didn’t lessen my stress. If you are a new…

--

--

Brian Jenney
Brian Jenney

Written by Brian Jenney

full-stackish developer, late bloomer coder and power google user and owner of Parsity.io

No responses yet