VFX pipelines with Thinkbox Deadline

Learn how to use Thinkbox Deadline, a powerful render management software used to streamline rendering workflows, with Cudo Compute.

Thinkbox Deadline is a powerful render management software used to streamline rendering workflows. It can manage multiple projects from different applications and distribute rendering jobs across your hardware. It provides pre-built render scripts for more than 80 VFX software, such as AutoCAD, Cinema 4D, Maya, Blender, and Houdini. It enables the scheduling of rendering jobs by connecting on-premise hardware with cloud computing. Furthermore, it offers virtual cloud workstations management, licensing management, and even automatic procurement of hourly licenses for rendering engines.

To set up Thinkbox Deadline, a central server is needed to host the database and a shared file system. After that, a client application is installed on workstations and rendering nodes, allowing them to connect and submit rendering jobs seamlessly.

From Deadline 10.1.23 licences are not required. Additionally, Thinkbox began deprecating the Balancer and the Cloud panel, as well as the five supporting Cloud Plugins that work with these features (Amazon, Azure, Google, OpenStack, and vCenter).

Quick start guide

  1. Prerequisites
  2. Setting up a Deadline repository on Ubuntu
  3. Setting up a local workstation
  4. Setting up a CPU/GPU rendering node


Setting up a Deadline repository on Ubuntu

  • Choose a VM add enough storage to hold your input and output media
  • Use the Ubuntu 22.04 + Docker image (in CLI tool type -image ubuntu-2204-docker)

ssh into the vm and:

wget https://thinkbox-installers.s3.us-west-2.amazonaws.com/Releases/Deadline/10.2/4_10.2.1.0/Deadline-
tar -xzf Deadline-

Perform an unattended installation of deadline:

Note change password to a secure memorable password, you will need it for the remote server client later Also replace the IP address of your server

./DeadlineRepository- \
--mode unattended --installmongodb true --certgen_password password --dbhost your-server-ip --dbssl true

Deadline will be installed to the following path /opt/Thinkbox/DeadlineRepository10

In order for Deadline to function correctly, it is essential that the Repository is visible to all connecting machines. This particular section outlines the process of sharing the Repository folder and adjusting its permissions to guarantee that Clients are granted the required access. Specifically, the Clients must be granted read access to both the Repository root and its subdirectories.

adduser --disabled-password --gecos "" deadlineuser
chown -R deadlineuser:nogroup /opt/Thinkbox/DeadlineRepository10
chmod -R 500 /opt/Thinkbox/DeadlineRepository10
cd /opt/Thinkbox/DeadlineRepository10
chmod -R 700 jobs/ jobsArchived/ reports/
cd ~

Next you can set up Samba as your network sharing protocol.

apt-get update
apt-get install samba -y

To enable Samba to share your Deadline repository paste this whole text block into your command line:

cat <<EOT >> /etc/samba/smb.conf
path = /opt/Thinkbox/DeadlineRepository10
writeable = Yes
guest ok = No
create mask = 0700
force create mode = 0700
force directory mode = 0700
unix extensions = No

Create a user and enter a password

smbpasswd -a deadlineuser
service smbd restart

sudo mount -t cifs -o username=deadlineuser,password=deadline // /mnt/repo

Using the file share

To access your shared directory on the server via your local computer (replace the password): Linux

sudo mkdir /mnt/repo
sudo mount -t cifs -o username=deadlineuser,password=samba-password //your-server-ip/DeadlineRepository /mnt/repo


mkdir repo
mount_smbfs //deadlineuser@your-server-ip/DeadlineRepository ./repo

Installing the remote connection server

The remote connection server allows for secure remote access to the Deadline database, enabling users to monitor and manage rendering jobs from outside the local network. It will be used for your local clients to connect to.

cd ~
mkdir -p /opt/Thinkbox/Deadline10/lib/python3/lib/python3
touch /opt/Thinkbox/Deadline10/lib/python3/lib/python3/lib
./DeadlineClient- \
--mode unattended --enable-components proxyconfig --disable-components webservice_config \
--repositorydir /opt/Thinkbox/DeadlineRepository10 --dbsslcertificate /opt/Thinkbox/DeadlineDatabase10/certs/Deadline10Client.pfx \
--dbsslpassword mypassword --enabletls true --launcherdaemon true

Setting up a local workstation

To install a client on your local Windows, Mac, or Linux computer follow these instructions: https://docs.thinkboxsoftware.com/products/deadline/10.2/1_User%20Manual/manual/quick-install-client.html

Configure your client to use a remote connection (not direct) with the following address your-server-ip:4433 use TLS and supply the certificate from the server /opt/Thinkbox/Deadline10/certs/Deadline10RemoteClient.pfx (you will need to copy it to your local computer)

You can copy it to your local device like this (Linux):

scp root@your-server-ip:/opt/Thinkbox/Deadline10/certs/Deadline10RemoteClient.pfx .

Setting up a CPU/GPU rendering node

  • Choose a VM with an NVIDIA GPU and Configure
  • Use the Ubuntu 22.04 + NVIDIA drivers + Docker image (in CLI tool type -image ubuntu-nvidia-docker)


  • Choose a CPU only VM
  • Use the Ubuntu 22.04 + Docker image (in CLI tool type -image ubuntu-2204-docker)

On your local computer run the following command to copy the client certificate from the server to the rendering node:

scp root@your-server-ip:/opt/Thinkbox/Deadline10/certs/Deadline10RemoteClient.pfx .
scp Deadline10RemoteClient.pfx root@your-worker-ip:/root

SSH into the rendering node, and mount the Samba file share in your worker:

ssh root@your-worker-ip
sudo mkdir /mnt/repo
sudo mount -t cifs -o username=deadlineuser,password=samba-password //your-server-ip/DeadlineRepository /mnt/repo

This fixes a bug in Deadline which prevent unattended mode from completing.

cd ~
mkdir -p /opt/Thinkbox/Deadline10/lib/python3/lib/python3
touch /opt/Thinkbox/Deadline10/lib/python3/lib/python3/lib

Install the deadline worker client:

wget https://thinkbox-installers.s3.us-west-2.amazonaws.com/Releases/Deadline/10.2/4_10.2.1.0/Deadline-
tar -xzf Deadline-
./DeadlineClient- \
--mode unattended
--repositorydir /mnt/repo --connectiontype remote --proxycertificate Deadline10RemoteClient.pfx
-- restartstalled true --slavestartup true --launcherdaemon true --proxyrootdir your-server-ip:4433

Deadline should now be running on your rendering node, if you open up the Deadline Monitor on your local computer you should be able to see the rendering node and send rendering jobs to it after installing your desired rendering engine on it.