Ezra offers a lighter task queue for small web apps
Ezra is a lightweight for handling in small web and app services. Large systems like Kafka can be too heavy because they often need clusters, dedicated servers, and extra operations work. Many small teams and indie builders avoid that setup and settle for in-memory queues, which can lose work when a process stops.
Ezra aims to sit in the middle. It runs as one binary and stores jobs in one file. This makes jobs easier to inspect and recover after crashes.
It is built with Elixir, which helps isolate failures and recover from crashes. It also speaks the Redis Streams wire protocol, so existing Redis clients in different s can connect without a special new client library.
Key points
- Ezra is meant to be simpler to run than Kafka-style queue systems.
- It runs as one binary and writes jobs to one file.
- Stored jobs can be inspected by opening ezra.db in a SQLite browser.
- It uses the Redis Streams wire protocol, so normal Redis clients can work with it.
- It targets small teams and indie builders who need reliable without large .