Fast Install

The steps, tips and tricks to get up and running as fast as possible.

When to use this page: This module is a condensed reference for experienced users who want to set up the entire lab environment quickly without the explanatory content of the individual modules. If this is your first time through the workshop, follow the full modules instead — they provide important context about what each step does and why.

Install Fast

{login_command}

git clone https://github.com/burrsutter/fantaco-redhat-one-2026
cd fantaco-redhat-one-2026/

There are a few key variables that have been pre-populated for you as part of this lab environment:

Persist the most important env variables to my_env_vars.sh therefore if they become lost, you can just simply source $HOME\fantaco-redhat-one-2026\my_env_var.sh

cat << 'EOF' > my_env_vars.sh
export VLLM_API_URL='{litellm_api_base_url}'
export VLLM_API_TOKEN='{litellm_virtual_key}'
export LLAMA_STACK_BASE_URL=http://llamastack-distribution-vllm-service.agentic-{user}.svc:8321
export INFERENCE_MODEL=vllm/qwen3-14b
export API_KEY="not-applicable"
export EMBEDDING_MODEL=sentence-transformers/nomic-ai/nomic-embed-text-v1.5
export EMBEDDING_DIMENSION=768
export CUSTOMER_MCP_SERVER_URL=https://$(oc get routes -l app=mcp-customer -o jsonpath="{range .items[*]}{.status.ingress[0].host}{end}")/mcp
export FINANCE_MCP_SERVER_URL=https://$(oc get routes -l app=mcp-finance -o jsonpath="{range .items[*]}{.status.ingress[0].host}{end}")/mcp
export CHAT_URL=http://$(oc get routes -l app=simple-agent-chat-ui -o jsonpath="{range .items[*]}{.status.ingress[0].host}{end}")
export LANGFUSE_HOST=https://$(oc get route langfuse -o jsonpath='{.spec.host}')
EOF

source $HOME/fantaco-redhat-one-2026/my_env_vars.sh

echo "VLLM_API_URL="$VLLM_API_URL
echo "VLLM_API_TOKEN="$VLLM_API_TOKEN
echo ""
echo "LLAMA_STACK_BASE_URL="$LLAMA_STACK_BASE_URL
echo "INFERENCE_MODEL="$INFERENCE_MODEL
echo "API_KEY="$API_KEY
echo ""
echo "EMBEDDING_MODEL"=$EMBEDDING_MODEL
echo "EMBEDDING_DIMENSION="$EMBEDDING_DIMENSION
echo ""
echo "CUSTOMER_MCP_SERVER_URL="$CUSTOMER_MCP_SERVER_URL
echo "FINANCE_MCP_SERVER_URL="$FINANCE_MCP_SERVER_URL
echo ""
echo "CHAT_URL"=$CHAT_URL
echo ""
echo "LANGFUSE_HOST"=$LANGFUSE_HOST

Note: Some of these are filled in as you work through the steps

https://llm-prod.apps.maas.redhatworkshops.io/v1
sk-331A-A8c_DxQiHnUPX1Ulg

oc apply -f llama-stack-scripts/llamastack-configmap.yaml

oc apply -f - <<'EOF'
apiVersion: llamastack.io/v1alpha1
kind: LlamaStackDistribution
metadata:
  annotations:
    openshift.io/display-name: llamastack-distribution-vllm
  name: llamastack-distribution-vllm
  labels:
    opendatahub.io/dashboard: 'true'
spec:
  replicas: 1
  server:
    containerSpec:
      command:
        - /bin/sh
        - '-c'
        - llama stack run /etc/llama-stack/run.yaml
      env:
        - name: VLLM_TLS_VERIFY
          value: 'false'
        - name: MILVUS_DB_PATH
          value: ~/.llama/milvus.db
        - name: FMS_ORCHESTRATOR_URL
          value: 'http://localhost'
        - name: VLLM_MAX_TOKENS
          value: '4096'
        - name: VLLM_API_URL
          value: '{litellm_api_base_url}'
        - name: VLLM_API_TOKEN
          value: '{litellm_virtual_key}'
        - name: LLAMA_STACK_CONFIG_DIR
          value: /opt/app-root/src/.llama/distributions/rh/
      name: llama-stack
      port: 8321
      resources:
        limits:
          cpu: '2'
          memory: 12Gi
        requests:
          cpu: 250m
          memory: 500Mi
    distribution:
      name: rh-dev
    userConfig:
      configMapName: llama-stack-config
EOF

echo "LLAMA_STACK_BASE_URL="$LLAMA_STACK_BASE_URL
echo "INFERENCE_MODEL="$INFERENCE_MODEL
echo "API_KEY="$API_KEY

oc get pods -w

NAME                                           READY   STATUS     RESTARTS   AGE
llamastack-distribution-vllm-9b647c5b5-xhfxs   0/1     Init:0/1   0          43s

./install-lls-fantaco.sh

oc get pods -w

NAME                                            READY   STATUS    RESTARTS   AGE
fantaco-customer-main-7fd4ddb666-4q8dh          1/1     Running   0          55s
fantaco-finance-main-75ffddb44b-t8rrg           1/1     Running   0          55s
langgraph-fastapi-569d6d554-cjnxl               1/1     Running   0          52s
llamastack-distribution-vllm-85866c49c6-k769c   1/1     Running   0          3m4s
mcp-customer-6bd8bcfc7b-kcjg5                   1/1     Running   0          53s
mcp-finance-8cc684b8d-d254r                     1/1     Running   0          53s
postgresql-customer-ff78dffdf-jrlhp             1/1     Running   0          55s
postgresql-finance-689d97894f-hks84             1/1     Running   0          55s
simple-agent-chat-ui-6d7794dc6b-5j6fs           1/1     Running   0          52s

source my_env_vars.sh
echo $CUSTOMER_MCP_SERVER_URL
echo $FINANCE_MCP_SERVER_URL

echo $CHAT_URL

Langfuse Setup

Keep the .venv folder inside of $HOME/fantaco-redhat-one-2026/

cd $HOME/fantaco-redhat-one-2026/

python -m venv .venv

source .venv/bin/activate

cd $HOME/fantaco-redhat-one-2026/langfuse-setup

source ./create-langfuse-url.sh

source ./create-secrets.sh

helm repo add langfuse https://langfuse.github.io/langfuse-k8s
helm repo update

helm install langfuse langfuse/langfuse \
    -f values-openshift-single-user.yaml \
    --set langfuse.nextauth.url=$LANGFUSE_HOST

oc get pods -w

NAME                                            READY   STATUS    RESTARTS      AGE
fantaco-customer-main-7fd4ddb666-c26fv          1/1     Running   0             55m
fantaco-finance-main-75ffddb44b-frlxb           1/1     Running   0             55m
langfuse-clickhouse-shard0-0                    1/1     Running   0             23m
langfuse-postgresql-0                           1/1     Running   0             23m
langfuse-redis-primary-0                        1/1     Running   0             23m
langfuse-s3-5fb6c8f845-s9shp                    1/1     Running   0             23m
langfuse-web-67f4c9849-rmkxt                    1/1     Running   2 (23m ago)   23m
langfuse-worker-69d997664c-bdr8f                1/1     Running   0             23m
langfuse-zookeeper-0                            1/1     Running   0             23m
langgraph-fastapi-569d6d554-65j95               1/1     Running   0             55m
llamastack-distribution-vllm-78ccd8fcbf-tp2ch   1/1     Running   0             59m
mcp-customer-6bd8bcfc7b-fmmt2                   1/1     Running   0             55m
mcp-finance-8cc684b8d-bln8l                     1/1     Running   0             55m
postgresql-customer-ff78dffdf-69p5n             1/1     Running   0             55m
postgresql-finance-689d97894f-jlmm8             1/1     Running   0             55m
simple-agent-chat-ui-6d7794dc6b-4cndl           1/1     Running   0             55m

langfuse-web likes to CrashLoop see trouble shooting section

echo $LANGFUSE_HOST

https://langfuse-agentic-user1.apps.cluster-m7dw5.dynamic.redhatworkshops.io

Create API Keys via the Langfuse GUI

Place the values in keys.txt. This is helpful if the env vars become lost due to browser refresh, timeout, cluster stop/sleep, etc.

vi keys.txt

source $HOME/fantaco-redhat-one-2026/langfuse-setup/convert-keys-txt-env-vars.sh

Verify you have the needed Langfuse env vars

echo $LANGFUSE_SECRET_KEY
echo $LANGFUSE_PUBLIC_KEY
echo $LANGFUSE_HOST

cd ~/fantaco-redhat-one-2026/langfuse-setup/langgraph-agent/backend

pip install -r requirements.txt

python 6-langgraph-langfuse-fastapi-chatbot.py

This can fail for a number of reasons, check the startup logging to see if all the env vars are correctly set.

Terminal 2

source $HOME/fantaco-redhat-one-2026/langfuse-setup/convert-keys-txt-env-vars.sh
echo $LANGFUSE_SECRET_KEY
echo $LANGFUSE_PUBLIC_KEY
echo $LANGFUSE_HOST

curl -sS -X POST http://localhost:8002/chat \
  -H "Content-Type: application/json" \
  -d '{
    "message": "what are Fran Wilson'\''s orders?",
    "user_id": "test-user",
    "session_id": "test-session-123"
}' | jq -r '.reply'

Evals

curl -sS -X POST http://localhost:8002/evaluate \
  -H "Content-Type: application/json" \
  -d '{
    "run_name": "after-prompt-update",
    "sync_dataset": true,
    "record_to_langfuse": true
  }' | jq .

Feedback

The Route was pre-created during cluster creation within the showroom namespace. This script just calculates it based on a pattern. You will need access to the chatbot GUI to leave feedback

For Showroom users:

$HOME/fantaco-redhat-one-2026/langfuse-setup/get-agent-route-as-student.sh

For Workbench users:

oc apply -f workbench-service-route.yaml

echo "https://$(oc get route chatbot-8002 -o jsonpath='{.spec.host}')"

Make sure you have the correct env vars

source $HOME/fantaco-redhat-one-2026/langfuse-setup/convert-keys-txt-env-vars.sh

List traces

source $HOME/fantaco-redhat-one-2026/langfuse-setup/list-traces.sh

What is the feedback of the last trace?

$HOME/fantaco-redhat-one-2026/langfuse-setup/get-trace-feedback.sh

Troubleshooting

OpenShift AI, aka RHODS

LlamaStack is still in developer preview and is often not backward compatible.

As Cluster Admin

oc get csv -A -o custom-columns=NAME:.metadata.name,VERSION:.spec.version | grep rhods | head -1

rhods-operator.3.2.0                      3.2.0

oc get llamastackdistribution -A -o custom-columns=NAME:.metadata.name,OPERATOR_VERSION:.status.version.operatorVersion,LLAMASTACK_VERSION:.status.version.llamaStackServerVersion

NAME                           OPERATOR_VERSION   LLAMASTACK_VERSION
llamastack-distribution-vllm   "0.4.0"            0.3.5.1+rhai0

Monitor that LLS pod

oc adm top pod -l app=llama-stack

NAME                                            CPU(cores)   MEMORY(bytes)
llamastack-distribution-vllm-6954d965d8-m29kr   2m           461Mi