Using Tooz Locking Abstraction

Picking the right distributed lock mechanism once and for all for your application is not necessarily an obvious choice. Consider these factors:

First, some solutions are heavier than others to deploy and maintain. For example, installing a memcached server is pretty straightforward, but maintaining a ZooKeeper cluster is much more complicated. Clearly, the two solutions are not strictly equivalent in terms of safety and guaranteed operation. However, as a developer, it might be handy to test using a small backend and operate at scale with a scalable backend.

Secondly, it is not always obvious which solution to pick. A few years ago, ZooKeeper was the hot thing and the only widely available implementation of the Paxos algorithm. Nowadays, solutions such as etcd and its Raft implementation are getting more traction: the algorithm is simpler to understand and the project is less complicated to deploy and operate.

All those backends offer different levels of abstraction on top of distributed features. Some projects provide a full locking implementation whereas some others are only key/value stores.

tooz library

The Tooz library was created a few years ago to solve those problems. It provides an abstraction on top of a varied set of backends, making it easy to switch from one service to another. This can be quite powerful, as it allows you to use memcached to test your distributed code on your laptop, for example, as it is lightweight to run and install, while you can also support something like a ZooKeeper cluster as a more robust solution. Moreover, whatever backend you pick, the distributed features that you need, such as locking, are provided using the same API for your application, whatever primitives provided by the underlying service.

To achieve that, Tooz provides a Coordinator object that represents the coordination service your application is connected to. The method tooz.coordinator.get_coordinator allows the program to get a new coordinator. All it needs is the URL to connect to and a unique identifier for the node, as you can see in the following example.

Note: You will need to start etcd to make the examples work. Simply run etcd from the command line in a separate terminal (open new terminal using + button beside Terminal tabs) before running each example as instructed. If you do not have etcd, you can use memcached instead. Replace the etcd://localhost URL with memcached://localhost in the code files and instead of etcd, run memcached -u memcache in a separate terminal.

After running etcd in a separate terminal you will see output similar to the following: