fpt

fpt

fpt is a command line tool to aid in the development and testing of Flexera Policies. The tool is able to syntax check, upload, and run Policies.

See ChangeLog.md for changes.

• Installation
  • Configuration
• Usage
• Change log
• Contributors
• License

Installation

Since fpt is written in Go it is compiled to a single static binary. Extract and run the executable below:

• Linux: v1/fpt-linux-amd64.tgz
• macOS: v1/fpt-darwin-amd64.tgz
• Windows: v1/fpt-windows-amd64.zip

Configuration

fpt interfaces with the Policy API. Credentials for the API can be provided in two ways:

1. YAML-based configuration file - Run fpt config account <name>, where name is a nickname for the account, to interactively write the configuration file into $HOME/.fpt.yml for the first time. You will be prompted for the following fields:
   • Account ID - Numeric account number, such as 60073
   • API endpoint host - Hostname, typically governance-3.rightscale.com
   • Refresh Token - Either:
     • Your personal OAuth refresh token available from Settings > Account Settings > API Credentials in the RightScale Cloud Management dashboard (this is deprecated in favor of Flexera One and will be removed in a future version)
     • Your API refresh token available from Flexera One > User Settings > API Credentials
   • Flexera One - Whether the Refresh Token is from the Flexera One platform instead of the legacy RightScale dashboard (the value should be true for Flexera One or false otherwise)
2. Environment variables - These are meant to be used by build systems such as Travis CI. The following vars must be set: FPT_LOGIN_ACCOUNT_ID, FPT_LOGIN_ACCOUNT_HOST, FPT_LOGIN_ACCOUNT_REFRESH_TOKEN. If the refresh token is from the Flexera One platform, you should set FPT_LOGIN_ACCOUNT_FLEXERA=true as well. These variables are equivalent to the ones described in the YAML section above.

Usage

Be sure to read Writing a Custom Policy and check out the brief Tutorial using this tool to develop a simple policy.

The following Policy related commands are supported:

fpt check <file>...
  Check syntax for a Policy Template.

fpt upload <file>...
  Upload Policy Template.

fpt run [<flags>] <file> [<options>...]
  Uploads and applies the PolicyTemplate.

  Execution of the policy will then be followed. Execution log will be tailed and followed and incident printed out.

  Options are user-supplied values for "parameters" defined in the PolicyTemplate language. Options must be in the form of "<name>=<value>". For arrays,
  values must be a comma separated list.

  Example: fpt run max_snapshots.pt regions=us-east-1,us-west-2 max_snapshot_count=100

  -C, --credentials=CREDENTIALS ...
                         Credentials is the map of name and credential used to launch the policy. Credentials must be of the form "--credentials <name1>=<value1> --credentials <name2>=<value2>".
  -n, --no-log           Do not print policy execution log.
  -k, --keep             Keep applied policy running at end, for inspection in UI. Normally policy is terminated at the end.
  -r, --run-escalations  If set, escalations will be run. Normally dry_run is set to avoid running any escalations.

fpt retrieve_data [<flags>] <file> [<options>...]
  Retrieve data from a Policy Template.

  Executes a policy once and retrieve generated datasources, saving them to disk.

  Example: fpt retrieve_data my_policy.pt --names instances

  -C, --credentials=CREDENTIALS ...
                               Credentials is the map of name and credential used to launch the policy. Credentials must be of the form "--credentials <name1>=<value1> --credentials <name2>=<value2>".
  -n, --names=NAMES ...        Names of resources/datasources to retrieve. By default, all datasources will be retrieved.
  -o, --output-dir=OUTPUT-DIR  Directory to store retrieved datasources.

fpt script [<flags>] <file> [<parameters>...]
  Run the body of a script locally.

  Example: fpt script max_snapshots.pt --result snapshots volumes=@ec2_volumes.json instances=@ec2_instances.json max_count=50

  -o, --out="out.json"  Script output file. Defaults to out.json
  -r, --result=RESULT   Name of variable holding final result to extract. Required if supplying a raw JavaScript.
  -n, --name=NAME       Name of script to run, if multiple exist.

Contributors

This tool is maintained by Douglas Thrift (douglaswth), Peter Schroeter (psschroeter), Avinash Bhashyam (avinashbhashyam-rs)

License

The fpt source code is subject to the MIT license, see the LICENSE file.