Process manager
Python Shell HTML
Switch branches/tags
Nothing to show
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


Pycroraptor2 is a process manager, essentially intended to be a Python replacement for the Microraptor process manager. Pycroraptor2 is in a working state but is not yet a feature-complete replacement for Microraptor.

The Pycroraptor2 system includes several components:

  • The pyraptord process management daemon. Typically, an instance of pyraptord is the parent process for all managed processes running on a particular host.
  • The pyrterm client shell that provides a tty user interface for controlling processes managed by pyraptord.
  • A web interface for controlling processes managed by pyraptord. This component has not been implemented yet.

Pycroraptor2 also offers a ZeroRPC API. To quickly get started using this API, you can use the zerorpc command-line tool included in the ZeroRPC Python repo or use the zclient Python shell included in the geocamUtil repo.


(Fill this in)


pyraptord is currently known to work in Linux (tested on RedHat Enterprise Linux 6 and Ubuntu 12.04) and Mac OS X (tested on Snow Leopard 10.6.8).

It probably will not work in any version of Windows without extensive modifications.


Pycroraptor2 depends on geocamUtil:

git clone
cd geocamUtil
python install

Download or clone from the geocamPycroraptor2 repository on GitHub:

git clone


In order to run, Pycroraptor2 requires two configuration files:

  • The ports.json file specifies network endpoints for named services that are available through the 0MQ socket library. This file specifies the endpoint where pyraptord should listen for commands and tells pyrterm or zclient where to find pyraptord.
  • The pycroraptor.json file specifies the rest of the Pycroraptor2 configuration, including how to run processes and where to put log files.

Examples for these files can be found in the tests subdirectory of the Pycroraptor2 repo.

Pycroraptor2 is often used to manage persistent processes that are started on boot. To make that happen:

  • You need to install a boot script for Pycroraptor2 that brings up the pyraptord instance for that host. See the Boot Script section below.
  • In order to start processes at pyraptord start time, they should be placed in the startup group in pycroraptor.json.

Boot Script

Example boot scripts for pyraptord are available in the bin subdirectory of the Pycroraptor2 repo, for the following platforms:

  • RedHat Enterprise Linux 6 (boot_script_redhat_pyraptord)
  • Ubuntu Linux 12.04 (boot_script_ubuntu_pyraptord.conf)
  • Mac OS X 10.8.3 Mountain Lion (boot_script_macosx_gov.nasa.arc.irg.pyraptord.plist)

Look within each boot script to find platform-specific install instructions.


Examples of how to start and stop the pyraptord daemon:

  • On RedHat Enterprise Linux:

    sudo service pyraptord start
    sudo service pyraptord stop
    sudo service pyraptord restart
    sudo service pyraptord status
  • On Mac OS X:

    sudo launchctl load /Library/LaunchDaemons/gov.nasa.arc.irg.pyraptord.plist
    sudo launchctl unload /Library/LaunchDaemons/gov.nasa.arc.irg.pyraptord.plist
    man launchctl

To control Pycroraptor2, run -p ports.json using the same ports.json file you used with pyraptord, and try some of these example commands: