Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/superfly/corrosion

Gossip-based service discovery (and more) for large distributed systems.
https://github.com/superfly/corrosion

database distributed distributed-systems gossip rust sqlite swim

Last synced: 4 days ago
JSON representation

Gossip-based service discovery (and more) for large distributed systems.

Awesome Lists containing this project

README

        

# Corrosion
Gossip-based service discovery (and more) for large distributed systems.

## Why we built Corrosion

We built Corrosion specifically for service discovery across a large global network, replacing Consul’s central state database with eventually consistent state distributed across our hosts.

Our new tool needed to deliver the following:

### Fast reads and writes

Getting state (data) from a central remote source can be incredibly expensive (at least 300ms for a round-trip to something on the other side of the world), but usually takes less than 1ms from a local source.

### Fast, eventual, consistency

Many use cases can cope with eventual consistency, especially if a consistent state is attained sooner than later. Raft fell short for some use cases at Fly.io where round-trips to a centralized location are too expensive.

### Flexibility

Global state for a distributed system isn't one-size-fits-all. Flexible schemas and queries are essential.

## How Corrosion works

In a nutshell, Corrosion:

- Maintains a SQLite database on each node
- Gossips local changes throughout the cluster
- Uses [CR-SQLite](https://github.com/vlcn-io/cr-sqlite) for conflict resolution with CRDTs
- Uses [Foca](https://github.com/caio/foca) to manage cluster membership using a SWIM protocol
- Periodically synchronizes with a subset of other cluster nodes, to ensure consistency

## Features

- A flexible API to read from and write to Corrosion's store using SQL statements
- File-based schemas with on-the-fly updates
- HTTP streaming subscriptions based on SQL queries
- Live population of configuration files from Corrosion state with user-defined [Rhai](https://rhai.rs/) templates
- Storage and propagation of state from locally registered Consul services, replacing the central database with Corrosion's distributed state
- Secure peer-to-peer communication with the [QUIC](https://datatracker.ietf.org/doc/html/rfc9000) transport protocol (using [Quinn](https://github.com/quinn-rs/quinn))

## Usage overview

Run the Corrosion agent on every node/host in the cluster. Other programs running on the node use [Corrosion's HTTP API](https://superfly.github.io/corrosion/api/index.html) to query the local Corrosion SQLite database, add and update data, and subscribe to change notifications.

The [Corrosion CLI](https://superfly.github.io/corrosion/cli/index.html) provides commands for administration and access to database and features.

### Quick start

- [Prepare the Corrosion configuration file](https://superfly.github.io/corrosion/config/)
- [Specify the initial database schema](https://superfly.github.io/corrosion/schema.html)
- [Start the Corrosion agent](https://superfly.github.io/corrosion/cli/agent.html)

See the WIP [Corrosion documentation](https://superfly.github.io/corrosion/) for more details.

## Building Corrosion

Clone [https://github.com/superfly/corrosion.git](https://github.com/superfly/corrosion.git).

From within the repo directory:

```
cargo build --release && mv target/release/corrosion ./
```