For users who just want to run OpenShift Lightspeed, please refer to the OpenShift Lightspeed Repository. This documentation provides instructions needed for setting up and using the service.
A Kubernetes operator for managing Red Hat OpenShift Lightspeed.
You'll need an OpenShift 4.15+ cluster to run against.
Important
Officially, the Operator only supports OpenAI, Azure OpenAI, WatsonX, RHELAI and RHOAI as large language model (LLM) providers, but technically, if you have an OpenAI API compatible model server (Ollama, VLLM, MLX), it should work.
Note: Your controller will automatically use the current context from your kubeconfig file (i.e. whatever cluster oc cluster-info shows).
- Deploy the controller to the cluster:
make deployAlternatively, to build the Docker image and push it to a personal repository, then deploy the operator into the cluster, use the following commands:
IMG="docker.io/username/ols-operator:0.10" make docker-build docker-push
IMG="docker.io/username/ols-operator:0.10" make deploy- Create a secret containing the API Key for Watsonx, OpenAI, Azure OpenAI. The key for API key is
apitoken.
Tip
Watsonx example
apiVersion: v1
data:
apitoken: <base64 encoded API Key>
kind: Secret
metadata:
name: watsonx-api-keys
namespace: openshift-lightspeed
type: OpaqueTip
OpenAI example
apiVersion: v1
data:
apitoken: <base64 encoded API Key>
kind: Secret
metadata:
name: openai-api-keys
namespace: openshift-lightspeed
type: OpaqueTip
Azure OpenAI apitoken example
apiVersion: v1
data:
apitoken: <base64 encoded API Key>
kind: Secret
metadata:
name: azure-openai-api-keys
namespace: openshift-lightspeed
type: OpaqueTip
Azure OpenAI user-assigned identity
apiVersion: v1
data:
client_id: <base64 encoded client id>
client_secret: <base64 encoded client secret>
tenant_id: <base64 encoded tenant id>
kind: Secret
metadata:
name: azure-api-keys
namespace: openshift-lightspeed
type: OpaqueThese apitoken or client_secret values can be updated if the user wishes to change them later. The same applies to all the TLS and CA certs related to individual components. They get reflected automatically across the system.
- Create an
OLSConfigcustom resource
apiVersion: ols.openshift.io/v1alpha1
kind: OLSConfig
metadata:
name: cluster
spec:
llm:
providers:
- type: openai
credentialsSecretRef:
name: openai-api-keys
models:
- name: gpt-3.5-turbo
name: openai
url: https://api.openai.com/v1
- type: watsonx
credentialsSecretRef:
name: watson-api-keys
models:
- name: ibm/granite-13b-chat-v2
name: watsonx
url: https://us-south.ml.cloud.ibm.com
- type: azure_openai
credentialsSecretRef:
name: azure-openai-api-keys
models:
- name: gpt-3.5-turbo
name: my_azure_openai
url: "https://myendpoint.openai.azure.com/"
ols:
conversationCache:
redis:
maxMemory: 2000mb
maxMemoryPolicy: allkeys-lru
type: redis
defaultModel: gpt-3.5-turbo
defaultProvider: openai
logLevel: INFO
deployment:
replicas: 1- The Operator will reconcile the CustomResource (CR) and create all the necessary resources for launching the
Red Hat OpenShift Lightspeedapplication server.
To delete the CRDs from the cluster:
make uninstallUnDeploy the controller from the cluster:
make undeploy- Create a namespace
openshift-lightspeed
oc create namespace openshift-lightspeed- Install the CRDs into the cluster:
make install- Run your controller (this will run in the foreground, so switch to a new terminal if you want to leave it running):
make run-
Create a secret containing the API Key for BAM or OpenAI. The key for API key is
apitoken. -
Create an
OLSConfigcustom resource -
The Operator will reconcile the CustomResource (CR) and create all the necessary resources for launching the
Red Hat OpenShift Lightspeedapplication server.
➜ oc get configmaps -n openshift-lightspeed
NAME DATA AGE
kube-root-ca.crt 1 33m
lightspeed-console-plugin 1 29m
olsconfig 1 21m
openshift-service-ca.crt 1 33m
➜ oc get services -n openshift-lightspeed
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
lightspeed-app-server ClusterIP 172.31.165.151 <none> 8443/TCP 22m
lightspeed-console-plugin ClusterIP 172.31.158.29 <none> 9443/TCP 29m
lightspeed-operator-controller-manager-service ClusterIP 172.31.63.140 <none> 8443/TCP 24m
➜ oc get deployments -n openshift-lightspeed
NAME READY UP-TO-DATE AVAILABLE AGE
lightspeed-app-server 1/1 1 1 23m
lightspeed-console-plugin 2/2 2 2 30m
lightspeed-operator-controller-manager 1/1 1 1 25m
➜ oc get pods -n openshift-lightspeed
NAME READY STATUS RESTARTS AGE
lightspeed-app-server-97c9c6d96-6tv6j 2/2 Running 0 23m
lilightspeed-console-plugin-7f6cd7c9fd-6lp7x 1/1 Running 0 30m
lightspeed-console-plugin-7f6cd7c9fd-wctj8 1/1 Running 0 30m
lightspeed-operator-controller-manager-69585cc7fc-xltpc 1/1 Running 0 26m
➜ oc logs lightspeed-app-server-f7fd6cf6-k7s7p -n openshift-lightspeed
2024-02-02 12:00:06,982 [ols.app.main:main.py:29] INFO: Embedded Gradio UI is disabled. To enable set enable_dev_ui: true in the dev section of the configuration file
INFO: Started server process [1]
INFO: Waiting for application startup.
INFO: Application startup complete.
INFO: Uvicorn running on http://0.0.0.0:8080 (Press CTRL+C to quit)If you have updated the API definitions, you must update the CRD manifests with the following command
make manifestsTo run the unit tests, we can run the following command
make testWhen using Visual Studio Code, we can use the debugger settings below to execute the test in debug mode
{
"version": "0.2.0",
"configurations": [
{
"name": "Launch Integration test ",
"type": "go",
"request": "launch",
"mode": "debug",
"program": "${workspaceFolder}/internal/controller",
"args": [
// "--ginkgo.v", # verbose output from Ginkgo test framework
],
"env": {
"KUBEBUILDER_ASSETS": "${workspaceFolder}/bin/k8s/1.27.1-linux-amd64"
},
},
]
}To run the end to end tests with a Openshift cluster, we need to have a running operator in the namespace openshift-lightspeed.
Please refer to the section Running on the cluster.
Then we should set 2 environment variables:
- $KUBECONFIG - the path to the config file of kubenetes client
- $LLM_TOKEN - the access token given by the LLM provider, here we use OpenAI for testing.
Then we can launch the end to end test by
make test-e2eWhen using Visual Studio Code, we can use the debugger settings below to execute the test in debug mode
{
"version": "0.2.0",
"configurations": [
{
"name": "Launch E2E test ",
"type": "go",
"request": "launch",
"mode": "debug",
"program": "${workspaceFolder}/test/e2e",
"args": [
// "--ginkgo.v", # verbose output from Ginkgo test framework
],
"env": {
"KUBECONFIG": "/path/to/kubeconfig",
"LLM_TOKEN": "sk-xxxxxxxx"
},
},
]
}NOTE: Run make --help for more information on all potential make targets
To update the catalog index from a Konflux snapshot, we need to connect to Konflux using oc login command:
- Go to the Developer Sandbox web portal https://registration-service-toolchain-host-operator.apps.stone-prd-host1.wdlc.p1.openshiftapps.com/
- Copy the proxy login command on the top right corner. It should look like this
oc login --token=$TOKEN --server=https://api-toolchain-host-operator.apps.stone-prd-host1.wdlc.p1.openshiftapps.com - Append our workspace to the server URL
oc login --token=$TOKEN --server=https://api-toolchain-host-operator.apps.stone-prd-host1.wdlc.p1.openshiftapps.com/workspaces/crt-nshift-lightspeed/ - Login using that command
Now we can use the script hack/snapshot_to_catalog.sh to update the catalog index. It takes 3 parameters snapshot_to_catalog.sh -s <snapshot-ref> -c <catalog-file>:
- snapshot-ref: required, the snapshot reference to use, example: ols-bnxm2
- catalog-file: optional, the catalog index file to update, default: lightspeed-catalog-4.16/index.yaml
To generate catalog index file lightspeed-catalog-4.16/index.yaml from the snapshot ols-bnxm2
➜ lightspeed-operator ✗ ./hack/snapshot_to_catalog.sh -s ols-bnxm2 -c lightspeed-catalog-4.16/index.yaml
Update catalog lightspeed-catalog-4.16/index.yaml from snapshot ols-bnxm2
using opm from /home/hasun/GitRepo/lightspeed-operator/bin/opm
using yq from /usr/bin/yq
Catalog will use the following images:
BUNDLE_IMAGE=registry.redhat.io/openshift-lightspeed-beta/lightspeed-operator-bundle@sha256:b9387e5900e700db47d2b4d7f106b43d0958a3b0d3d4f4b68495141675b66a1c
OPERATOR_IMAGE=registry.redhat.io/openshift-lightspeed-beta/lightspeed-rhel9-operator@sha256:4bb81dfec6cce853543c7c0e7f2898ece23105fe3a5c5b17d845b1ff58fdc92a
CONSOLE_IMAGE=registry.redhat.io/openshift-lightspeed-beta/lightspeed-console-plugin-rhel9@sha256:4f45c9ba068cf92e592bb3a502764ce6bc93cd154d081fa49d05cb040885155b
SERVICE_IMAGE=registry.redhat.io/openshift-lightspeed-beta/lightspeed-service-api-rhel9@sha256:794017379e28cfbbd17c8a8343f3326f2c99b8f9da5e593fa5afd52258d0c563
BUNDLE_IMAGE_ORIGIN=quay.io/redhat-user-workloads/crt-nshift-lightspeed-tenant/ols/bundle@sha256:b9387e5900e700db47d2b4d7f106b43d0958a3b0d3d4f4b68495141675b66a1c
Bundle version is 0.1.0
Validation passed for lightspeed-catalog-4.16/index.yaml
You'll need the following tools to develop the Operator:
- OperatorSDK, version 1.33
- git
- go, version 1.21
- docker, version 17.03+.
- oc or oc and access to an OpenShift cluster of a compatible version.
- golangci-lint, version v1.54.2
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot]")
![openshift-merge-bot[bot] avatar](https://avatars.githubusercontent.com/in/412865?v=4 "openshift-merge-bot[bot]")
![red-hat-konflux[bot] avatar](https://avatars.githubusercontent.com/in/296509?v=4 "red-hat-konflux[bot]")
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent