feat: update node quickstart script with gpu auto detect

[ndrpp]

Fixes # .

Changes proposed in this PR:

• update node quickstart script to include UPnP setup
• automatically detect gpus and add them to docker compute environments

[alexcos20]

Intel:

root@liveu-buntu:/repos/ocean/ocean-node# ./scripts/ocean-node-quickstart.sh
Do you have your private key for running the Ocean Node [ y/n ]: y
Enter your private key: XXXXXXXXXXX
Please provide the wallet address to be added as Ocean Node admin account: XXXXXXX
Provide the HTTP_API_PORT value or accept the default (press Enter) [8000]:
Provide the P2P_ipV4BindTcpPort or accept the default (press Enter) [9000]:
Provide the P2P_ipV4BindWsPort or accept the default (press Enter) [9001]:
Provide the P2P_ipV6BindTcpPort or accept the default (press Enter) [9002]:
Provide the P2P_ipV6BindWsPort or accept the default (press Enter) [9003]:
Enable UPnP (useful in case you can no set up port forwarding)? [ y/n ]: y
Provide the public IPv4 address or FQDN where this node will be accessible: 8.8.8.8
Do you want to run docker C2D jobs on your Ocean Node [ y/n ]: y
########################################################
### Docker Engine Compute Environments Configuration ###
########################################################
Check 'ComputeEnvironment' definition for more details on the format
_____________________________________________________

Do you want to add a specific docker environment configuration?
  (Hint: You can enter multiple in JSON format) [ y/n ]: y
Enter the array of docker environment(s):
Setting default DOCKER_COMPUTE_ENVIRONMENTS configuration
Checking for GPUs...
Detected 1 GPU type(s). Updating configuration...
GPUs added to Compute Environment resources.
[ { "socketPath": "/var/run/docker.sock", "resources": [ { "id": "disk", "total": 10 }, { "id": "intel-corporation-arrow-lake-p-intel-graphics", "description": "Intel Corporation Arrow Lake-P [Intel Graphics]", "type": "gpu", "total": 1, "init": { "deviceRequests": { "DeviceIDs": [ "0000:00:02.0" ], "Capabilities": [ [ "gpu" ] ] } } } ], "storageExpiry": 604800, "maxJobDuration": 36000, "minJobDuration": 60, "fees": { "1": [ { "feeToken": "0x123", "prices": [ { "id": "cpu", "price": 1 } ] } ] }, "free": { "maxJobDuration": 360000, "minJobDuration": 60, "maxJobs": 3, "resources": [ { "id": "cpu", "max": 1 }, { "id": "ram", "max": 1 }, { "id": "disk", "max": 1 } ] } } ]
Docker Compose file has been generated successfully.

You are now ready to start your Ocean Node.

1) If further customization is required, edit the docker-compose.yml file.
For all available configurations, refer to the environment variables documentation:
https://github.com/oceanprotocol/ocean-node/blob/main/docs/env.md

2) Start your Ocean Node by running the command:
docker-compose up -d

3) Allow and forward the following incoming TCP ports through the firewall to the Ocean Node host:
HTTP API Port: 8000
P2P IPv4 TCP Port: 9000
P2P IPv4 WebSocket Port: 9001
P2P IPv6 TCP Port: 9002
P2P IPv6 WebSocket Port: 9003

which is not correct, as it's not including the device.
Correct config looks like:

{
resources:
[
{id:"GPU-abcdef1234567890",
type:"gpu",
total:1,
init":{
      advanced:{
         Devices:["/dev/dri"],
         GroupAdd:["render"]
       }
 }
 }]

In case of multiple gpus, each one has it's own set of devices, like '/dev/dri/card1' and /dev/dri/renderD128 for card 1, '/dev/dri/card2' and /dev/dri/renderD129 for card2

[alexcos20]

see my comments

[alexcos20]

same for AMD, you are not passing the devices

[alexcos20]

Nvidia: it does not detect anything:

root@gpu-1:/repos/ocean2/ocean-node# git checkout feat/158-node-install-script-updates
Branch 'feat/158-node-install-script-updates' set up to track remote branch 'feat/158-node-install-script-updates' from 'origin'.
Switched to a new branch 'feat/158-node-install-script-updates'
root@gpu-1:/repos/ocean2/ocean-node# ./scripts/ocean-node-quickstart.sh
Do you have your private key for running the Ocean Node [ y/n ]: y
Enter your private key: XXXXXXX
Please provide the wallet address to be added as Ocean Node admin account: XXXXXXXX
Provide the HTTP_API_PORT value or accept the default (press Enter) [8000]:
Provide the P2P_ipV4BindTcpPort or accept the default (press Enter) [9000]:
Provide the P2P_ipV4BindWsPort or accept the default (press Enter) [9001]:
Provide the P2P_ipV6BindTcpPort or accept the default (press Enter) [9002]:
Provide the P2P_ipV6BindWsPort or accept the default (press Enter) [9003]:
Enable UPnP (useful in case you can no set up port forwarding)? [ y/n ]: y
Provide the public IPv4 address or FQDN where this node will be accessible: 8.8.8.8
Do you want to run docker C2D jobs on your Ocean Node [ y/n ]: y
########################################################
### Docker Engine Compute Environments Configuration ###
########################################################
Check 'ComputeEnvironment' definition for more details on the format
_____________________________________________________

Do you want to add a specific docker environment configuration?
  (Hint: You can enter multiple in JSON format) [ y/n ]: y
Enter the array of docker environment(s):
Setting default DOCKER_COMPUTE_ENVIRONMENTS configuration
Skipping GPU detection (script not found or jq missing).
[ { "socketPath": "/var/run/docker.sock", "resources": [ { "id": "disk", "total": 10 } ], "storageExpiry": 604800, "maxJobDuration": 36000, "minJobDuration": 60, "fees": { "1": [ { "feeToken": "0x123", "prices": [ { "id": "cpu", "price": 1 } ] } ] }, "free": { "maxJobDuration": 360000, "minJobDuration": 60, "maxJobs": 3, "resources": [ { "id": "cpu", "max": 1 }, { "id": "ram", "max": 1 }, { "id": "disk", "max": 1 } ] } } ]
Docker Compose file has been generated successfully.

You are now ready to start your Ocean Node.

1) If further customization is required, edit the docker-compose.yml file.
For all available configurations, refer to the environment variables documentation:
https://github.com/oceanprotocol/ocean-node/blob/main/docs/env.md

2) Start your Ocean Node by running the command:
docker-compose up -d

3) Allow and forward the following incoming TCP ports through the firewall to the Ocean Node host:
HTTP API Port: 8000
P2P IPv4 TCP Port: 9000
P2P IPv4 WebSocket Port: 9001
P2P IPv6 TCP Port: 9002
P2P IPv6 WebSocket Port: 9003

root@gpu-1:/repos/ocean2/ocean-node# nvidia-smi
Mon Jan 19 05:56:33 2026
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 580.65.06              Driver Version: 580.65.06      CUDA Version: 13.0     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce GTX 1060 3GB    On  |   00000000:01:00.0 Off |                  N/A |
|  0%   35C    P8              6W /  120W |       3MiB /   3072MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+
root@gpu-1:/repos/ocean2/ocean-node#