Configuration Overview
Styx can be configured through environment variables, command-line arguments, and runtime parameters.
Environment Variables
Server Configuration
Unique identifier for the Styx node in the cluster. Each node must have a distinct ID.
STYX_SERVER
string
default:"http://localhost:8080"
Server URL for the Styx CLI client to connect to the oracle.
Command-Line Arguments
Styx Server
The styx-server binary accepts an optional port argument:
Port number for the HTTP server to listen on.
# Run on default port 8080
./styx-server
# Run on custom port 9090
./styx-server 9090
Styx CLI
The styx CLI tool supports multiple commands:
Docker Configuration
Docker Run
Configure Styx using environment variables with Docker:
docker run -d \
--name styx \
-p 8080:8080 \
-e STYX_NODE_ID=1 \
styx:latest
Docker Compose
Set environment variables in your docker-compose.yml:
version: '3.8'
services:
styx:
build: .
ports:
- "8080:8080"
environment:
- STYX_NODE_ID=1
restart: unless-stopped
Kubernetes Configuration
ConfigMap
Create a ConfigMap for centralized configuration:
apiVersion: v1
kind: ConfigMap
metadata:
name: styx-config
data:
node_id: "1"
port: "8080"
Environment Variables in Deployment
Inject configuration into pods:
apiVersion: apps/v1
kind: Deployment
metadata:
name: styx
spec:
template:
spec:
containers:
- name: styx
image: styx:latest
env:
- name: STYX_NODE_ID
value: "1"
Server Endpoints
The Styx server exposes the following HTTP endpoints:
Health Check
Returns server health status. Used by Docker and Kubernetes health probes.
Query Node Status
GET /query?target=<node_id>
{
"alive_confidence": 0.85,
"dead_confidence": 0.10,
"unknown": 0.05,
"dead": false,
"refused": false,
"refusal_reason": ""
}
Submit Witness Report
Submit a witness report about a target node.
Request Body:
{
"witness": 1,
"target": 2,
"alive": 0.9,
"dead": 0.05,
"unknown": 0.05
}
Register Witness
Register a new witness node in the system.
CLI Configuration
Setting Server URL
Configure the CLI to connect to a remote Styx server:
export STYX_SERVER=http://styx.example.com:8080
styx health
Production Configuration
Recommended Settings
version: '3.8'
services:
styx:
image: styx:latest
ports:
- "8080:8080"
environment:
- STYX_NODE_ID=${NODE_ID:-1}
restart: always
healthcheck:
test: ["CMD", "wget", "--no-verbose", "--tries=1", "--spider", "http://localhost:8080/health"]
interval: 30s
timeout: 3s
retries: 3
start_period: 5s
deploy:
resources:
limits:
cpus: '0.5'
memory: 256M
reservations:
cpus: '0.1'
memory: 128M
Logging
Styx logs to standard output. Configure log aggregation:
Security Considerations
- Network Policies: Restrict ingress/egress traffic in Kubernetes
- TLS: Use a reverse proxy (nginx, Envoy) for HTTPS termination
- Authentication: Implement API authentication for production deployments
- Resource Limits: Always set CPU and memory limits
- Health Checks: Configure liveness and readiness probes
Troubleshooting
Check Configuration
docker inspect styx | jq '.[0].Config.Env'
Common Issues
Port Already in Use:
# Find process using port 8080
lsof -i :8080
# Use a different port
./styx-server 9090
# Check if server is running
ps aux | grep styx-server
# Set correct server URL
export STYX_SERVER=http://correct-host:8080