Getting Started¶
Requirements¶
- Python 3.9 – 3.14
- An OpenStack account with API credentials
Installation¶
After installation the orca command is available globally.
Configuration¶
Interactive setup¶
This prompts for all fields and stores them in ~/.orca/config.yaml (mode 0600). The token cache lives next to it as ~/.orca/token_cache.yaml.
Profiles¶
orca supports multiple named profiles for managing several clouds:
orca profile add # add a new profile interactively
orca profile list # list all profiles
orca profile switch prod # switch the active profile
orca --profile dev server list # use a specific profile for one command
Import from clouds.yaml¶
Environment variables¶
Standard OpenStack OS_* variables are supported:
export OS_AUTH_URL="https://keystone.example.com:5000/v3"
export OS_USERNAME="myuser"
export OS_PASSWORD="mypassword"
export OS_USER_DOMAIN_NAME="Default"
export OS_PROJECT_NAME="myproject"
Tip
Resolution priority: --profile flag → ORCA_PROFILE → OS_* env vars → OS_CLOUD → clouds.yaml → active orca profile.
Shell Auto-Completion¶
Use orca completion install — it generates a static script under $XDG_DATA_HOME/orca/completion.<shell> and adds a single source line to your rc file. This avoids re-spawning orca on every shell startup (cf. ADR 0010).
Open a new shell and orca <TAB> should suggest commands. Resource IDs/names (servers, volumes, flavors, images, networks, security groups, server groups, keypairs) are completed via per-profile cached lookups (5-minute TTL).
Verify Installation¶
orca --version
orca catalog # list available service endpoints
orca server list # list your compute instances
orca doctor # health-check config and connectivity
Next Steps¶
Browse the command reference for detailed usage of each service, or the CLI Reference for an exhaustive single-page view.