CluebaseVMS

Ubuntu Server/Desktop 22.04 / 24.04 / 26.04 Pre-Configuration Guide for High-Load VMS
This guide describes the recommended initial configuration of Ubuntu systems for stable operation of a high-load video surveillance platform based on Cluebase VMS running in Docker containers.
The recommendations apply to:
Ubuntu Server/Desktop 22.04 LTS
Ubuntu Server/Desktop 24.04 LTS
Ubuntu Server/Desktop 26.04 LTS
The system is expected to operate under:
High network throughput
Intensive disk I/O
Large numbers of simultaneous video streams
NVIDIA GPU acceleration
Continuous 24/7 workloads
1. Fully Update the System
Update package indexes and install all available updates.
sudo apt update && sudo apt dist-upgrade -y && sudo apt autoremove -y && sudo apt autoclean && sudo snap refresh
Reboot after updates:
sudo reboot
2. Disable Automatic System Updates
Automatic updates may restart services unexpectedly and affect VMS stability.
Disable unattended upgrades:
sudo systemctl stop unattended-upgrades && sudo systemctl disable unattended-upgrades
Disable periodic APT updates:
sudo nano /etc/apt/apt.conf.d/20auto-upgrades
Set:
APT::Periodic::Update-Package-Lists "0";
APT::Periodic::Unattended-Upgrade "0";
Save the file.
3. Install Additional System Packages
Install useful administration and monitoring utilities:
sudo apt install -y \
mc \
htop \
curl \
unzip \
apt-transport-https \
ca-certificates \
gnupg \
lsb-release \
gnupg2 \
iotop \
net-tools \
iftop \
nvme-cli \
smartmontools \
jq
Package purpose:
Package
Purpose
mc
File manager
htop
Process monitoring
iotop
Disk I/O monitoring
iftop
Network traffic monitoring
smartmontools
Disk health monitoring
nvme-cli
NVMe SSD management
4. Check Secure Boot Status
Secure Boot must be disabled in BIOS/UEFI before installing proprietary NVIDIA drivers.
Check status:
mokutil --sb-state
Possible output:
SecureBoot enabled
or
SecureBoot disabled
If Secure Boot is enabled:
Reboot the server
Enter BIOS/UEFI
Disable:
Secure Boot
Fast Boot (recommended)
Save settings and reboot
5. Install NVIDIA Driver and NVIDIA Container Toolkit
Install NVIDIA drivers only if the server has an NVIDIA GPU.
Recommended driver installation:
sudo ubuntu-drivers autoinstall
Reboot:
sudo reboot
Verify driver installation:
nvidia-smi
Expected result:
GPU model displayed
Driver version visible
No errors
Install NVIDIA Container Toolkit only after installing Cluebase VMS or Docker
Official documentation:
NVIDIA Container Toolkit Install Guide
Configure the production repository:
curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
  && curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
    sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
    sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list
Update the packages list from the repository:
sudo apt update
Install the NVIDIA Container Toolkit packages:
export NVIDIA_CONTAINER_TOOLKIT_VERSION=1.19.0-1
  sudo apt-get install -y \
      nvidia-container-toolkit=${NVIDIA_CONTAINER_TOOLKIT_VERSION} \
      nvidia-container-toolkit-base=${NVIDIA_CONTAINER_TOOLKIT_VERSION} \
      libnvidia-container-tools=${NVIDIA_CONTAINER_TOOLKIT_VERSION} \
      libnvidia-container1=${NVIDIA_CONTAINER_TOOLKIT_VERSION}
Configure runtime:
sudo nvidia-ctk runtime configure --runtime=docker
Restart Docker:
sudo systemctl restart docker
6. Enable TCP BBR Congestion Control
BBR improves network throughput and reduces latency under heavy traffic.
Open sysctl configuration:
sudo nano /etc/sysctl.conf
Add:
net.core.default_qdisc=fq
net.ipv4.tcp_congestion_control=bbr
Apply settings:
sudo sysctl -p
Verify:
sysctl net.ipv4.tcp_congestion_control
Expected output:
net.ipv4.tcp_congestion_control = bbr
7. Configure Static IP Address
Ubuntu 22.04/24.04/24.06 uses Netplan.
Find interface name:
ip a
Example interface:
ens18
Edit Netplan configuration:
sudo nano /etc/netplan/01-netcfg.yaml
Example configuration:
network:
  version: 2
  renderer: networkd
  ethernets:
    ens18:
      dhcp4: false
      addresses:
        - 192.168.1.100/24
      routes:
        - to: default
          via: 192.168.1.1
      nameservers:
        addresses:
          - 1.1.1.1
          - 8.8.8.8
Apply configuration:
sudo netplan apply
Verify:
ip a
ip route
8. Configure Additional fs.aio-max-nr Parameter
Increase asynchronous I/O limits for high-load applications.
Edit sysctl configuration:
sudo nano /etc/sysctl.d/99-aio.conf
Add:
fs.aio-max-nr = 1048576
Apply:
sudo sysctl --system
Verify:
sysctl fs.aio-max-nr
Additional Recommended Optimizations for High-Load Cluebase VMS Systems
Increase File Descriptor Limits
Large video systems require high open-file limits.
Edit:
sudo nano /etc/security/limits.conf
Add:
* soft nofile 1048576
* hard nofile 1048576
root soft nofile 1048576
root hard nofile 1048576
Also configure systemd:
sudo nano /etc/systemd/system.conf
Add or modify:
DefaultLimitNOFILE=1048576
Edit:
sudo nano /etc/systemd/user.conf
Add:
DefaultLimitNOFILE=1048576
Reload:
sudo systemctl daemon-reexec
Increase Network Buffers
Edit:
sudo nano /etc/sysctl.conf
Add:
net.core.rmem_max = 67108864
net.core.wmem_max = 67108864
net.core.rmem_default = 262144
net.core.wmem_default = 262144
net.ipv4.tcp_rmem = 4096 87380 67108864
net.ipv4.tcp_wmem = 4096 65536 67108864
net.core.netdev_max_backlog = 250000
Apply:
sudo sysctl -p
Set Performance CPU Governor
Recommended for dedicated VMS servers.
Install tools:
sudo apt install -y linux-tools-common linux-tools-generic
Set performance mode:
sudo cpupower frequency-set -g performance
Verify:
cpupower frequency-info
Enable TRIM for SSD/NVMe
sudo systemctl enable fstrim.timer
sudo systemctl start fstrim.timer
Verify:
systemctl status fstrim.timer
Recommended BIOS Settings
For stable 24/7 operation:
Disable:
Secure Boot
Fast Boot
CPU C-States (optional for latency-sensitive systems)
ASPM power saving (optional)
Enable:
Above 4G Decoding (for large GPUs)
Resizable BAR (if supported)
Performance power profile
Monitoring Recommendations
Useful commands:
CPU:
htop
Disk I/O:
iotop
GPU:
watch -n 1 nvidia-smi
Network:
iftop
Disk health:
sudo smartctl -a /dev/sda
Final Recommendations
For production Cluebase VMS servers:
Use Ubuntu Server instead of Desktop whenever possible
Use enterprise-grade SSD/NVMe drives
Use RAID with battery-backed cache for archive storage
Use dedicated NICs for camera traffic
Prefer 10G networking for large installations
Keep OS and archive storage separated
Avoid installing unnecessary GUI applications and services
Reboot only during maintenance windows
Use UPS power protection
Monitor disk temperatures and SMART status regularly

WebRTC Setup for Cluebase VMS
Local Network Usage
WebRTC works inside a local network by default.
No additional configuration is required.
A TURN server is only needed if remote access to Cluebase VMS is required over the Internet.
1. Install a TURN Server in Docker on a VPS
Create a 
compose.yaml file with the following content:
services:
  coturn:
    image: coturn/coturn
    container_name: coturn
    restart: always
    network_mode: 'host'
    command:
      - -n
      - --log-file=stdout
      - --listening-ip=0.0.0.0
      - --relay-ip=local_ip
      - --external-ip=public_ip
      - --min-port=40000
      - --max-port=59000
      - --no-auth
Replace the following values:
local_ip — local IP address of the VPS
public_ip — public IP address of the VPS
Start the TURN server:
docker compose up -d
2. Using the Same Server for Cluebase VMS and TURN
If the server where Cluebase VMS is installed has a public IP address, the TURN server can be installed on the same server.
In this case:
Install the TURN server on the same machine as Cluebase VMS
Configure port forwarding on the router
Ports that need to be forwarded in the router:
3478 TCP and UDP
40000-59000 UDP
If a firewall is enabled, make sure these ports are allowed in the firewall rules.
3. Configure Cluebase VMS
Edit the 
.env file of Cluebase VMS.
Change:
ENABLE_STUN_TURN=0
to:
ENABLE_STUN_TURN=1
Add the following parameters:
STUN_SERVER_HOST=turn_server_ip
STUN_SERVER_PORT=3478
Replace 
turn_server_ip with the IP address of your TURN server.

Enabling Visual Assistant in Cluebase VMS
1. Install and Run Ollama with the LLaVA Model
Create a 
compose.yaml file with the following content:
volumes:
  ollama:
services:
  ollama:
    image: ollama/ollama
    container_name: ollama
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]
    environment:
      - OLLAMA_SCHED_SPREAD=1
    ports:
      - "11434:11434"
    volumes:
      - ollama:/root/.ollama
    restart: always
Start the Ollama container:
docker-compose up -d
After the container is running, download the 
llava model:
docker exec -it ollama ollama pull llava
2. Enable Ollama Support in the MySQL Database
Connect to the MySQL database container:
docker exec -it vms-db mysql -uroot -p
When prompted, enter the MySQL root password.
Select the 
vcloud database:
USE vcloud;
Enable Ollama integration by executing the following query:
UPDATE GeneralSettings SET ollamaStatus="ACTIVE";
Exit the MySQL console:
exit
Result
After completing these steps:
Ollama will be running in Docker
The 
llava visual model will be installed
Visual Assistant support will be activated in Cluebase VMS

Optimizing PTZ Performance in Cluebase VMS
To achieve the lowest possible latency and the most responsive PTZ (Pan-Tilt-Zoom) control, please ensure your system is configured with the following settings:
Set Camera Codec to H.264. Configure your camera's primary stream to use the H.264 video codec. While H.265 is efficient for storage, H.264 typically offers better compatibility and lower processing overhead for real-time control.
Use WebRTC Protocol. Ensure that WebRTC is selected as the streaming protocol in the Cluebase VMS interface. WebRTC is specifically designed for real-time communication and significantly reduces the delay compared to traditional protocols like WS or HLS.

Enabling NVIDIA GPU Support for the LPR Module in Cluebase VMS
Prerequisites
NVIDIA GPU support requires the NVIDIA drivers and NVIDIA Container Toolkit to be installed on the host system.
Install and configure NVIDIA Container Toolkit before proceeding:
NVIDIA Container Toolkit Documentation
You can verify that Docker has access to the NVIDIA runtime using:
docker info | grep -i nvidia
You can also verify GPU availability on the host system with:
nvidia-smi
1. Load the Docker image
Copy the 
roadar_lpr_websocket.tar.gz archive containing the Docker image to the server.
Load the Docker image using the following command:
sudo docker load -i roadar_lpr_websocket.tar.gz
2. Stop the existing 
vms-lpr container
Navigate to the directory where Cluebase VMS is installed and stop the 
vms-lpr container if it is currently running:
sudo docker compose down vms-lpr
3. Update the 
docker-compose.yml configuration
Open the 
docker-compose.yml file in a text editor.
Replace the following configuration block:
vms-lpr:
  image: vcloudaiorg/vcloudai-vms-lpr:latest
  restart: always
  container_name: vms-lpr
  network_mode: host
  extra_hosts:
    - host.docker.internal:host-gateway
  depends_on:
    - vms-server
  volumes:
    - ./static/lpr:/data
  environment:
    WS_HOST: host.docker.internal:${WS_SERVER_PORT}
    API_HOST: 127.0.0.1
    API_PORT: ${ROADAR_PORT}
    STATE_FILE: /data/state.json
    LICENSE_SERVICE_HOST: 127.0.0.1
    LICENSE_SERVICE_PORT: 32433
with the following GPU-enabled configuration:
vms-lpr:
  image: roadar_lpr_websocket
  restart: always
  container_name: vms-lpr
  network_mode: host
  extra_hosts:
    - host.docker.internal:host-gateway
  depends_on:
    - vms-server
  runtime: nvidia
  deploy:
    resources:
      reservations:
        devices:
          - driver: nvidia
            count: all
            capabilities: [gpu, compute, video]
  volumes:
    - ./static/lpr:/data
  environment:
    - WS_HOST=host.docker.internal:${WS_SERVER_PORT}
    - API_HOST=127.0.0.1
    - API_PORT=${ROADAR_PORT}
    - STATE_FILE=/data/state.json
    - LICENSE_SERVICE_HOST=127.0.0.1
    - LICENSE_SERVICE_PORT=32433
    - TRIAL_TIME=0
    - NVIDIA_VISIBLE_DEVICES=all
    - NVIDIA_DRIVER_CAPABILITIES=compute,video,utility
Save the file and exit the text editor.
4. Start the 
vms-lpr container
Run the following command to start the container with NVIDIA GPU support enabled:
sudo docker compose up -d vms-lpr
5. Verify NVIDIA GPU support inside the container
To verify that the container has access to the NVIDIA GPU, run:
sudo docker exec -it vms-lpr nvidia-smi
If GPU support is configured correctly, the command will display information about the installed NVIDIA GPU(s).

Connecting a NAS to Linux via iSCSI
If your vCloud server and NAS storage are on separate machines, the most reliable way to connect them is iSCSI.
iSCSI allows the NAS to present a disk over the network, and Linux sees it as a local block device (like 
/dev/sdb). This is ideal for video storage because it usually performs better than SMB/NFS for large sequential writes.
Architecture
+-------------------+        iSCSI        +-------------------+
|   vCloud Server   | <-----------------> |        NAS        |
| Ubuntu Server     |                     | Synology/TrueNAS  |
| 192.168.1.10      |                     | 192.168.1.20      |
+-------------------+                     +-------------------+
Step 1 — Prepare the NAS
You need to configure the NAS and create an iSCSI LUN. To do this, refer to the user manual for your NAS server.
Step 2 — Install iSCSI Client on Ubuntu
On the vCloud server:
sudo apt update
sudo apt install open-iscsi -y
Enable the service:
sudo systemctl enable --now iscsid
Step 3 — Discover the NAS Target
Replace 
192.168.1.20 with your NAS IP.
sudo iscsiadm -m discovery -t sendtargets -p 192.168.1.20
You should see something like:
192.168.1.20:3260,1 iqn.2026-07.local.synology:vcloud-storage
Step 4 — Configure Authentication (if CHAP is enabled)
Edit the node configuration:
sudo nano /etc/iscsi/iscsid.conf
Find and set:
node.session.auth.authmethod = CHAP
node.session.auth.username = vcloud
node.session.auth.password = StrongPassword123
Save the file.
Restart the service:
sudo systemctl restart iscsid
Step 5 — Log In to the iSCSI Target
sudo iscsiadm -m node --login
Expected output:
Login to [iface: default, target: iqn.2026-07.local.synology:vcloud-storage, portal: 192.168.1.20,3260] successful.
Step 6 — Verify the New Disk
List disks:
lsblk
Example:
sda    1.8T
├─sda1
└─sda2
sdb   10.0T
The iSCSI disk is usually 
sdb.
Step 7 — Create a Filesystem
⚠️ This will erase the iSCSI disk.
sudo mkfs.ext4 /dev/sdb
Step 8 — Create a Mount Point
sudo mkdir -p /mnt/vcloud-storage
Step 9 — Mount the Disk
sudo mount /dev/sdb /mnt/vcloud-storage
Check:
df -h
Example:
/dev/sdb        9.8T   24K  9.3T   1% /mnt/vcloud-storage
Step 10 — Make the Mount Persistent
Get the UUID:
sudo blkid /dev/sdb
Example:
/dev/sdb: UUID="a1b2c3d4-e5f6-7890-abcd-1234567890ef"
Edit 
/etc/fstab:
sudo nano /etc/fstab
Add:
UUID=a1b2c3d4-e5f6-7890-abcd-1234567890ef /mnt/vcloud-storage ext4 _netdev,nofail 0 2
Test:
sudo umount /mnt/vcloud-storage
sudo mount -a
If there are no errors, the configuration is correct.
Step 11 — Ensure iSCSI Reconnects After Reboot
Enable automatic login:
sudo iscsiadm -m node --op update -n node.startup -v automatic
Check:
sudo iscsiadm -m node
Useful Commands
Check iSCSI sessions
sudo iscsiadm -m session
Log out
sudo iscsiadm -m node --logout
Rediscover targets
sudo iscsiadm -m discovery -t sendtargets -p 192.168.1.20