Bad rollout took down Google Maps!

Watch the video explanation ➔

On the 18th of March, 2022, Google Maps faced a major outage affecting millions of people for a couple of hours. The outage happened due to a bad deployment.

Although a bad deployment does not sound too bad, the situation worsens when there are cascading failures if 3 services have a synchronous dependency, forming a chain-like A -> B -> C.

If A goes down, it will have some impact on B, and if the impact is big enough, we might see B going down as well; and extending it, we might see C going down.

This is exactly what happened in this Google Outage. Some services had a bad deployment, and they started crashing. Tile Rendering service depended on it, and the Tile rendering service went down because of retries.

The Direction SDK, Navigation SDK directly invoked the Tile rendering service for rendering the maps, which didn’t work, causing a big outage.

How to remediate a bad deployment?

Rollback as soon as possible.

Preventing cascading outages

  • Reject requests when the server is exhausted
  • Tune the performance of the webserver and networking stack of the server
  • Monitor the server resource consumption and set alerts
  • Add circuit breakers wherever possible.

Here's the video ⤵

Courses I teach

Alongside my daily work, I also teach some highly practical courses, with a no-fluff no-nonsense approach, that are designed to spark engineering curiosity and help you ace your career.


System Design Masterclass

A no-fluff masterclass that helps experienced engineers form the right intuition to design and implement highly scalable, fault-tolerant, extensible, and available systems.


Details →

System Design for Beginners

An in-depth and self-paced course for absolute beginners to become great at designing and implementing scalable, available, and extensible systems.


Details →

Redis Internals

A self-paced and hands-on course covering Redis internals - data structures, algorithms, and some core features by re-implementing them in Go.


Details →


Writings and Learnings

Knowledge Base

Bookshelf

Papershelf


Arpit's Newsletter read by 90000+ engineers

Weekly essays on real-world system design, distributed systems, or a deep dive into some super-clever algorithm.