Configuration

Sed ut perspiciatis unde omnis iste natus error sit voluptatem accusantium doloremque laudantium, totam rem aperiam. Flux reads configuration from three sources, in this order of precedence:

1. Command-line flags (--workspace, --log-level, …)
2. Environment variables prefixed with FLUX_
3. A TOML file at ~/.flux/config.toml (or $FLUX_CONFIG)

Example config file

# ~/.flux/config.toml

workspace = "acme"
log_level = "info"

[backend]
driver = "postgres"
dsn    = "postgres://flux:flux@localhost:5432/flux"

[telemetry]
otlp_endpoint = "http://localhost:4317"
sample_rate   = 1.0

[limits]
max_concurrent_runs  = 32
max_step_duration_ms = 600_000

Environment variables

Variable	Default	Description
FLUX_WORKSPACE	–	Logical tenant (required)
FLUX_TOKEN	–	Workspace API token (required)
FLUX_LOG_LEVEL	info	trace, debug, info, warn, error
FLUX_BACKEND_DSN	sqlite:./flux.db	Persistence DSN
FLUX_OTLP_ENDPOINT	–	OpenTelemetry collector URL

Secrets

Reference secrets in pipelines via the secret:// URI scheme. The CLI resolves them lazily at runtime, never at parse time:

step('publish')
  .to('kafka://orders.v1')
  .with({
    sasl: {
      username: 'flux',
      password: 'secret://kafka/password',
    },
  });

Supported secret backends: AWS Secrets Manager, HashiCorp Vault, Azure Key Vault, and a local .flux/secrets file for development.