Jepsen

Breaking distributed systems so you don't have to.

Jepsen is a Clojure library. A test is a Clojure program which uses the Jepsen library to set up a distributed system, run a bunch of operations against that system, and verify that the history of those operations makes sense. Jepsen has been used to verify everything from eventually-consistent commutative databases to linearizable coordination systems to distributed task schedulers. It can also generate graphs of performance and availability, helping you characterize how a system responds to different faults. See aphyr.com for examples of the sorts of analyses you can carry out with Jepsen.

Design overview

A Jepsen test runs as a Clojure program on a control node. That program uses SSH to log into a bunch of db nodes, where it sets up the distributed system you're going to test using the test's pluggable os and db.

Once the system is running, the control node spins up a set of logically single-threaded processes, each with its own client for the distributed system. A generator generates new operations for each process to perform. Processes then apply those operations to the system using their clients. The start and end of each operation is recorded in a history. While performing operations, a special nemesis process introduces faults into the system--also scheduled by the generator.

Finally, the DB and OS are torn down. Jepsen uses a checker to analyze the test's history for correctness, and to generate reports, graphs, etc. The test, history, analysis, and any supplementary results are written to the filesystem under store/<test-name>/<date>/ for later review. Symlinks to the latest results are maintained at each level for convenience.

Setting up a Jepsen environment

Your local machine needs a JVM and leiningen 2 installed. Probably want JNA for SSH auth too.

sudo apt-get install openjdk-8-jre openjdk-8-jre-headless libjna-java

For your db nodes, you'll need some (I use five) debian boxes. I run debian jessie, but some DBs don't need the latest packages so you might get away with an older distribution, or possibly ubuntu. Each one should be accessible from the control node via SSH. By default they're named n1, n2, n3, n4, and n5, but that (along with SSH username, password, identity files, etc) is all definable in your test. The account you use on those boxes needs sudo access to set up DBs, control firewalls, etc.

Be advised that tests may mess with clocks, add apt repos, run killall -9 on processes, and generally break things, so you shouldn't, you know, point jepsen at your prod machines unless you like to live dangerously, or you wrote the test and know exactly what it's doing.

See lxc.md for some of my notes on setting up LXC instances.

Running a test

Once you've got everything set up, you should be able to run cd aerospike; lein test, and it'll spit out something like

INFO  jepsen.core - Analysis invalid! (ﾉಥ益ಥ）ﾉ ┻━┻

{:valid? false,
 :counter
 {:valid? false,
  :reads
  [[190 193 194]
   [199 200 201]
   [253 255 256]
   ...}}

Writing a test

If you don't know Clojure, you'll want to learn some of the basics. Try Clojure From the Ground Up and Clojure for the Brave and True. Or you can reimplement Jepsen's ideas in a language you do know. Either way's fine!

For a complete guide to writing a test, start with scaffolding.

FAQ

JSCH auth errors

You might be hitting a jsch bug which doesn't know how to read hashed known_hosts files. Remove all keys for the DB hosts from your known_hosts file, then:

ssh-keyscan -t rsa n1 >> ~/.ssh/known_hosts
ssh-keyscan -t rsa n2 >> ~/.ssh/known_hosts
ssh-keyscan -t rsa n3 >> ~/.ssh/known_hosts
ssh-keyscan -t rsa n4 >> ~/.ssh/known_hosts
ssh-keyscan -t rsa n5 >> ~/.ssh/known_hosts

to add unhashed versions of each node's hostkey to your ~/.ssh/known_hosts.