Deadline rendering farm construction record

The working mode of the rendering farm: multiple machines form a cluster for rendering. For example, there are two rendering machines A and B in the cluster. The server will be assigned to tasks A and B at the same time. A renders the first frame, and B renders the second frame. When A is completed, the server will continue to allocate the third frame. Similarly, after the B server is completed, the server will allocate the fourth frame, and so on.
The rendering farm is divided into server and nodes. The server can also be added to the cluster as a node. Of course, the node can also submit rendering tasks.
Suggestions: The server can be installed separately, and the main plug-ins and custom modifications can be stored separately. When uploading batch rendering tasks, you don't need to take up rendering node resources.
Basic deployment mode:

necessary conditions:
1. Hardware: at least 2 computers
2. Each computer needs to install the same version of the software, and the installation path should be unified (not uniformly modify the default path of the software on the server)
3. At least one cluster A master (Deadline Repository), Master can be the CLInet at the same time

1. Install Master:
1. The default installation is complete DeadlineRepository-*****-windows-installer
2. Configure MongoDB certification path

3. Firewall open port 27100
4. Share DeadlineRepository and DeadlineDatabase to give Node read-only permission.
2. Install Node:
1. Default installation path

2.

Fill in the repository folder path of the deadline server\DeadlineServerHostname\DeadlineRepository10

Fill in the database certificate
\DeadlineServerHostname\DeadlineRepository10 \DeadlineDatabase10\certs\Deadline10Client.pfx

select license free
and keep the default installation completed

---

---

3. Add the version number of the rendering software to Deadline.
Take Maya as an example.
1. Before the operation, back up the DeadlineRepository on the master.
2. Modify the MayaSubmission.py file under \DeadlineServerHostname\DeadlineRepository10\scripts\Submission, find the value of scriptDialog.AddComboControlToGrid, and add the new version number:

3. Modify the Houdini.param file under \DeadlineServerHostname\DeadlineRepository10\plugins\MayaBatch, and add the value corresponding to the new version number to the drop-down list control:

---

4. Add scripts under Slaves
to add refresh group policy script as an example
1. Back up the DeadlineRepository on the master before operation
2. Add the UpdateGroupPolicy.py file in \DeadlineServerHostname\DeadlineRepository10\scripts\Slaves
3. The specific content is as follows

##------------------------------------------------------------
## Gpupdate .py
##
##
## Update Group Policy on the selected machines.
##------------------------------------------------------------

from System.IO import *
from Deadline.Scripting import *

#from DeadlineUI.Controls.Scripting.DeadlineScriptDialog import DeadlineScriptDialog

def __main__():
    # Get the selected slave info settings.
    selectedSlaveInfoSettings = MonitorUtils.GetSelectedSlaveInfoSettings()
    # Get the list of selected machine names from the slave info settings.
    machineNames = SlaveUtils.GetMachineNameOrIPAddresses(selectedSlaveInfoSettings)
    if len(machineNames) > 0:
        for machineName in machineNames:
            SlaveUtils.SendRemoteCommand(machineName, "Execute cmd /C  gpupdate  /Target:User  /force")

4. Modify the name displayed by the script in the configure script menus under DeadlineMonitor