Raft-based, highly-available distributed cache implementation
Aravinth Manivannan
7d5ca30ae8
Some checks are pending
ci/woodpecker/push/woodpecker Pipeline is pending
Reviewed-on: #14 |
||
---|---|---|
bench | ||
dcache_py | ||
proto/dcache | ||
src | ||
tests | ||
.dockerignore | ||
.gitignore | ||
.woodpecker.yml | ||
build.rs | ||
Cargo.lock | ||
Cargo.toml | ||
check.sh | ||
Dockerfile | ||
launch.sh | ||
LICENSE.md | ||
Makefile | ||
README.md | ||
renovate.json | ||
requirements.txt | ||
test-cluster.sh | ||
test.py |
dcache: Distributed, Highly Available cache implementation for mCaptcha
Overview
Tips
We recommend running at least three instances of dcache in your deployment.
NOTE: Catastrophic failure will occur when n/2 + 1 instances are down.
Usage
Firewall configuration
dcache uses a single, configurable port for both server-to-server and client-to-server communications. Please open that port on your server.
Launch
dcache --id 1 \
--http-addr 127.0.0.1:9001 \
--introducer-addr 127.0.0.1:9001 \
--introducer-id 1 \
--cluster-size 3
Options
Name | Purpose |
---|---|
--id | Unique integer to identify node in network |
--http-addr | Socket address to bind and listen for connections |
--introducer-addr | Socket address of introducer node; required to join network |
--intdocuer-id | ID of the introducer node; required to join network |
--cluster-size | Total size of the cluster |