labManager-master 1.0.5

System for managing multi-station multi-tenant lab setups - master.

labManager v1.0.5

System for managing multi-station multi-tenant lab setups - master

This package is part of the labManager tools, a collection of tools for managing behavioral research setups. Please see here for more information.

Installation

pip install labManager-master
# or
pip install labManager-master[GUI] # ①

1. use the GUI extra to install labManager-master’s GUI.

Example script

An example script for running the labManager master script is provided here.

Configuration

The below shows the schema used for parsing the configuration file for labManager master, using StrictYAML’s notation. An example configuration file is also available.

'network': s.Str(),                                     # Network on which to discover clients, e.g. 10.0.1.0/24

'service_discovery_protocol':                           # Protocol to use for client discovery, MDNS or SSDP
    s.Enum(['MDNS','SSDP']),

s.Optional('MDNS'): s.Map({
    'service': s.Str(),                                 # Service name to discover when using MDNS, e.g.,
}),                                                     # _master._labManager._tcp.local.

s.Optional('SSDP'): s.Map({
    'device_type': s.Str(),                             # Device type to announce and listen for when using SSDP, e.g.,
}),                                                     # urn:schemas-upnp-org:device:labManager

s.Optional('projects'): s.Map({                         # Table of alias names for projects (to enable showing more friendly
                                                        # names)
    'name_table': s.MapPattern(s.Str(), s.Str()),       # Example entry: `0000-01: Demo environment`
}),

s.Optional('base_image_name_table'):                    # Table of alias names for disk images (to enable showing more friendly
                                                        # names)
    s.MapPattern(s.Str(), s.Str()),                     # Example entry: `station_base: Windows station`
s.Optional('SMB'): s.Map({                              # If users have access to a central storage facility using an SMB,
                                                        # configuration about the server and how project shares are named on it
    'server': s.Str(),                                  # Server FQDN or IP address
    'domain': s.Str(),                                  # Domain in which users are found (may be overridden by LDAP reply)
    'projects': s.Map({                                 # Project-to-SMB share mapping config
        'format': s.Str(),                              # Regex to match shares that are for projects
        s.Optional('remove_trailing', default=''):      # Characters to remove from end of project share name to map the to
            s.Str(),                                    # project names
    }),
    'mount_share_on_client': s.Bool(),                  # Boolean indicating whether the project share should be mounted as a
                                                        # network drive on client machines once the client connects to this
                                                        # master.
    'mount_drive_letter': s.Str(),                      # Drive letter for mapping the network drive, if
                                                        # `mount_share_on_client` is enabled
    'mount_only_known_clients': s.Bool(),               # Issue command to mount the network share only for known clients (i.e.
}),                                                     # clients listed in the clients configuration section below), not for
                                                        # other machines that run a labManager client (to prevent snooping of
                                                        # the user's credentials)

s.Optional('admin'): s.Map({                            # Configuration about the labManager admin-server
    'server': s.Str(),                                  # Server FQDN or IP address
    'port': s.Int(),                                    # Server port
}),

s.Optional('toems'): s.Map({                            # Configuration about the Theopenem instance
    'server': s.Str(),                                  # Server FQDN or IP address
    'port': s.Int(),                                    # Server port
    'images': s.Map({                                   # Project-to-disk image mapping config
        'format': s.Str(),                              # Regex to match disk images that belong to a project
    }),
    s.Optional('pre_upload_script'): s.Str(),           # Script that will be configured to run in Theopenem's LIE imaging
                                                        # environment when uploading a disk image (at the BeforeImaging stage)
    s.Optional('image_info_script'): s.Str(),           # Script that will be configured to run in Theopenem's LIE imaging
                                                        # environment when deploying a disk image (at the AfterFileCopy stage)
    s.Optional('image_info_script_partition'):          # Partition on the disk image for which the `image_info_script` should
        s.Int(),                                        # run.
}),

s.Optional('login'): s.Map({
    'hint': s.Str(),                                    # login hint to show in the labManager master GUI
}),

s.Optional('clients'): s.Seq(                           # Configuration for known clients, e.g., fixed stations in a lab setup
    s.Map({
        'name': s.Str(),                                # Name by which station should be known. Example entry: STATION01
        'MAC' : s.CommaSeparated(s.Str()),              # One or multiple MAC addresses of the station. Example entry:
    })                                                  # 0C:9D:92:1F:E6:04, F4:E9:D4:73:6F:EC, F4:E9:D4:73:6F:ED
),

s.Optional('tasks'): s.Seq(                             # preconfigured tasks to be shown in the labManager master GUI
    s.Map({
        'name': s.Str(),                                # Name to show in task GUI, should be descriptive
        'type': s.Enum(task.types),                     # Task type, one of labManager.common.task.Type
        s.Optional('payload', default=''): s.Str(),     # Command or other payload to execute
        s.Optional('payload_type', default='text'):     # Payload type, either as text to directly execute or path to a file to
            s.Enum(['text','file']),                    # load the payload from
        s.Optional('cwd', default=''): s.Str(),         # CWD in which to execute the command
        s.Optional('env', default=''): s.Str(),         # environment variables to set when executing the command
        s.Optional('interactive', default=False):       # Boolean indicating whether this is an interactive task. If true,
            s.Bool(),                                   # users can send commands to the task as it is running (e.g., use cmd
                                                        # as a remote shell)
        s.Optional('python_unbuffered', default=False): # if true, appends the -u flag to commands running the python
            s.Bool(),                                   # executable to put it in unbuffered mode, so that any output is
    }),                                                 # directly written to stdout/stderr and can be remotely monitored. Does
),                                                      # nothing for task types other than task.Type.Python_module and
                                                        # task.Type.Python_script

Standalone deployment

One simple way to make a standalone install of the app is to download WinPython, e.g. the latest 3.10.x version. I think this is recommended instead of using a system-wide or user installation of Python, so that users of the computer are unlikely to accidentally interfere with the Python distribution that runs the management tools.

Do as follows:

1. Download the dot version, not the full package, e.g. Winpython64-3.10.11.1dot.
2. Run the downloaded exe, which unzips the WinPython files.
3. Take the python folder from the unzipped files (e.g. python-3.10.11.amd64), you do not need the rest. This is your python distribution. Put it where you want on the disk.
4. Open a command prompt in the root of the Python installation. Install the wanted labManager packages into it using, e.g., .\python.exe -m pip install labManager-master.
5. Finally use the python.exe in the folder to execute your script, such the example script to launch this tool.

Acknowledgements

This project was made possible by funding from the LMK foundation, Sweden.