Fuzz Testing and Verification of Apache Cassandra with 'Harry' - Alex Petrov

STRATEGIC
---------------
Google

PLATINUM
--------------
Apple
Huawei
Instaclustr
Tencent Cloud

GOLD
-----------
Aiven OY
AWS
Baidu
Cerner
Didi Chuxing
Dremio
Fiter
Gradle
Red Hat
Replicated Fuzz Testing and Verification of Apache Cassandra with "Harry" - Alex Petrov

Cassandra is a mature database, and most of the “straightforward” issues are already found, fixed, and covered with regression tests. To gain more confidence in Apache Cassandra releases, we need to search for issues that reveal themselves only during extensive testing in scenarios that are as close to the real world as possible. Current testing tooling in Apache Cassandra largely tests for common- and edge-cases, and most of the tests use predefined datasets. Property-based tests can help explore a broader range of states, but often require either a complex model and a large state to test against. To solve this problem we’ve introduced, Harry, a component for fuzz testing and verification of Apache Cassandra.

Using Harry, we’ve been able to identify and stably reproduce issues that would be difficult to find by hand. Found issues, when applicable, was reduced to a minimum number of steps needed to reproduce them using tools that comes with Harry.

In this talk, you will learn about the main concepts behind Harry, how these concepts help to keep verification state to the minimum, make verifications performant, learn about the structure of its primary components are structured, and hear about the process of finding issues using Harry.