Skip to content

cld2labs/ubuntu22.04-deployment-scripts#49

Open
HarikaDev296 wants to merge 21 commits intoopea-project:mainfrom
cld2labs:cld2labs/ubuntu22.04-deployment-scripts
Open

cld2labs/ubuntu22.04-deployment-scripts#49
HarikaDev296 wants to merge 21 commits intoopea-project:mainfrom
cld2labs:cld2labs/ubuntu22.04-deployment-scripts

Conversation

@HarikaDev296
Copy link

Adding ubuntu deployment scripts under third-party dell folder.

Copy link
Collaborator

@amberjain1 amberjain1 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed the general structure.
Will wait for validation to complete before reviewing the iac folder.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we only capture specific instructions that differ as compared to the main readme guides. For generic steps shall we start pointing to common setup and user-guides so that we do not duplicate documentation?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This documentation is done as per Dell's guidance

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shall we create a separate common document (or point to the one that is already present in docs folder) and only capture delta's here?

Copy link
Collaborator

@alexsin368 alexsin368 Feb 3, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I also agree. There are a lot of duplicate instructions. I think the changes are minimal such that we can have just one README for setting up both apisix and genai gateway. The testing inference portions seems more elaborate, so those can be split into two.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This documentation is done as per Dell's guidance

@preethivenkatesh
Copy link
Collaborator

preethivenkatesh commented Feb 3, 2026

@bagd0003 please add your review comments here

Co-authored-by: alexsin368 <109180236+alexsin368@users.noreply.github.com>
```
**Mount ISO**
```bash
./mount-iso.sh
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I had to change all .sh scripts to be executable before I could run them.

chmod 755 ./*.sh

Either add this step or change the permissions of the .sh files before commiting them.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Updated readme to run chmod +x mount.sh before executing the script

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@bagd0003 Heerak mentioned you had concerns in the way that the ISO is being downloaded from the public ubuntu url, followed by installation via terraform. Please recommend if there is a better way Dell prefers before we upstream this code

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is this related to ISO mount, already updated the script so that it can pick http and https?

-p Linux123! \
-t hf_xxxxxxxxxxxxx \
-g gaudi3 \
-m "1"
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does the user have to input a model ID or can the user select from the list of models in the menu? Better to make it flexible for the user to choose the desired model to deploy.

Updated user guide for Intel AI for Enterprise Inference to clarify setup instructions and improve readability.

```bash
cd ~
git clone https://github.com/opea-project/Enterprise-Inference.git
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

All these deployment scripts are inside the Enterprise-Inference repo already. Can we avoid re-downloading it?

Harika and others added 3 commits February 4, 2026 18:19
Updated README.md to include new instructions for mounting ISO and clarify mandatory Terraform variables.
Added instructions for changing file permissions and running the script.

```bash
wget -O deploy-enterprise-inference.sh \
https://raw.githubusercontent.com/cld2labs/Enterprise-Inference/refs/heads/cld2labs/ubuntu22.04-deployment-scripts/third_party/Dell/ubuntu-22.04/iac/deploy-enterprise-inference.sh
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Isnt cloning a better option? so its less hardcoded for better maintenance?

-m "1" \
-r
```

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The following info should be included here:

  1. the model ID numbers corresponding to the model names in the EI pre-validated models list
  2. how to put in a model from HF (for Gaudi, will also need to specify TP size)
  3. how to install another model or uninstall an existing model without tearing down the kubernetes deployment

MODELS="Enter Model number"
DEPLOYMENT_MODE="keycloak"
DEPLOY_OBSERVABILITY="off"
KEYCLOAK_CLIENT_ID="api"
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

change the default values for keycloak to match these values: https://github.com/opea-project/Enterprise-Inference/blob/main/core/inventory/inference-config.cfg

keycloak_client_id=my-client-id
keycloak_admin_user=your-keycloak-admin-user
keycloak_admin_password=changeme

please update these values anywhere else in the script or other files

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants