DevOps Finland April meetup

Hosted by Zalando
Sponsored by incidentist.io (and last year by SOK.fi)

Agenda
17:00 Doors open
17:30 Welcome / DevOps Finland & Zalando
17:45 A perfect storm: how a tiny change in a CMS service authentication once broke up the Zalando Fashion Store? / Uri Savelchev, Senior DevOps Engineer @ Zalando Finland
18:30 ***break ***
18:45 Start SLOw! - How to ru(i)n your services using Service Level Objectives / Vitek Urbanec, Chief Resiliency & Governance Consultant @ incidentist.io
19:30 Free discussion
20:30 Doors close

A perfect storm: how a tiny change in a CMS service authentication once broke up the Zalando Fashion Store?

Is it possible that a little mistake in update of a CMS, that is pretty far from the "hot path" and the customer traffic, could break up the operations and lead to a high-profile incident? I'll tell you a thrilling story about one of the most puzzling incidents I have ever dealt with.

For the last fifteen years I'm specializing in DevOps/SRE and Platform Engineering.

Start SLOw! - How to ru(i)n your services using Service Level Objectives

In the vibe-coded world, it's essential to know when your change broke your company's stream of income (again), and how much more of that can your customers take before they vibe-code their own SaaS (with blackjack & webhooks). Calculating good_events / total_events * 100% is just the beginning. You've done the calculation, you've set a made-up arbitrary objective, but what do you do next? In this talk, we dig beyond the trivial topics, our starting line will be "we already have the SLOs, what now?"

Vitek Urbanec is a Chief Resiliency & Governance Consultant at incidentist.io, who has spent the better part of 13 years making sure other people's production systems don't catch fire — and occasionally helping them understand why they did. He's worked across industries in fintech, gaming, and SaaS, most recently embedded with a platform team, helping squads turn uptime feelings into actionab