Skip to content

Deploy Speech Models to GKE #60

Deploy Speech Models to GKE

Deploy Speech Models to GKE #60

Workflow file for this run

name: Deploy Speech Models to Cloud RUN
# TODO: determine if changes to backend folder before pushing
on:
#push:
# branches: [ "main", "development" ]
# paths:
# - 'backend/modal/**'
workflow_dispatch:
inputs:
environment:
description: 'Select the environment to deploy to'
required: true
default: 'development'
branch:
description: 'Branch to deploy from'
required: true
default: 'main'
env:
SERVICE: models
REGION: us-central1
CONTAINER_NAME: speech-models
jobs:
deploy:
environment: ${{ (github.ref == 'refs/heads/development' && 'development') || (github.ref == 'refs/heads/main' && 'prod') }}
permissions:
contents: 'read'
id-token: 'write'
runs-on: ubuntu-latest
steps:
# To workaround "no space left on device" issue of GitHub-hosted runner
- name: Delete huge unnecessary tools folder
run: rm -rf /opt/hostedtoolcache
- name: Checkout
uses: actions/checkout@v4
- name: Google Auth
id: auth
uses: 'google-github-actions/auth@v2'
with:
credentials_json: ${{ secrets.GCP_CREDENTIALS }}
- run: gcloud auth configure-docker
- name: Build and Push Docker image
run: |
docker build -t gcr.io/${{ vars.GCP_PROJECT_ID }}/${{ env.SERVICE }}:${GITHUB_SHA::7} -f backend/modal/Dockerfile .
docker push gcr.io/${{ vars.GCP_PROJECT_ID }}/${{ env.SERVICE }}:${GITHUB_SHA::7}
- name: Connect to GKE cluster
run: |
curl https://packages.cloud.google.com/apt/doc/apt-key.gpg | sudo gpg --dearmor -o /usr/share/keyrings/cloud.google.gpg
echo "deb [signed-by=/usr/share/keyrings/cloud.google.gpg] https://packages.cloud.google.com/apt cloud-sdk main" | sudo tee -a /etc/apt/sources.list.d/google-cloud-sdk.list
sudo apt-get update && sudo apt-get install google-cloud-cli-gke-gcloud-auth-plugin -y
gcloud container clusters get-credentials ${{ vars.GKE_CLUSTER }} --region ${{ env.REGION }} --project ${{ vars.GCP_PROJECT_ID }}
- name: Deploy VAD to GKE cluster using Helm
run: |
helm -n ${{ vars.ENV }}-omi-backend upgrade --install ${{ vars.ENV }}-omi-vad ./backend/charts/vad -f ./backend/charts/vad/${{ vars.ENV }}_omi_vad_values.yaml --set "image.tag=${GITHUB_SHA::7}"
# - name: Create SSH Key
# run: |
# mkdir -p ~/.ssh
# echo "${{ secrets.SPEECH_MODELS_SSH_PRIVATE_KEY }}" > ~/.ssh/id_ed25519
# chmod 600 ~/.ssh/id_ed25519
# - name: Deploy Docker image
# run: |
# ssh -o StrictHostKeyChecking=no \
# ${{ secrets.SPEECH_MODELS_SSH_USERNAME }}@${{ secrets.SPEECH_MODELS_SSH_HOST }} \
# "set -x; \
# echo '[+] Pull latest Speech Models image...'; \
# docker pull gcr.io/${{ vars.GCP_PROJECT_ID }}/${{ env.SERVICE }}:${GITHUB_SHA::7}; \
# echo '[+] Remove current Speech Models container...'; \
# docker rm -f ${{ env.CONTAINER_NAME }}; \
# echo '[+
# ] Start new Speech Models container...'; \
# docker run -d --name ${{ env.CONTAINER_NAME }} -p 8080:8080 \
# --volume /var/lib/nvidia/lib64:/usr/local/nvidia/lib64 \
# --volume /var/lib/nvidia/bin:/usr/local/nvidia/bin \
# --device /dev/nvidia0:/dev/nvidia0 \
# --device /dev/nvidia-uvm:/dev/nvidia-uvm \
# --device /dev/nvidiactl:/dev/nvidiactl \
# -e OPENAI_API_KEY=${{ secrets.OPENAI_API_KEY }} \
# -e HUGGINGFACE_TOKEN=${{ secrets.HUGGINGFACE_TOKEN }} \
# gcr.io/${{ vars.GCP_PROJECT_ID }}/${{ env.SERVICE }}:${GITHUB_SHA::7}"
# - name: Deploy to Cloud Run
# id: deploy
# uses: google-github-actions/deploy-cloudrun@v2
# with:
# service: ${{ env.SERVICE }}
# region: ${{ env.REGION }}
# image: gcr.io/${{ vars.GCP_PROJECT_ID }}/${{ env.SERVICE }}
# If required, use the Cloud Run url output in later steps
# - name: Show Output
# run: echo ${{ steps.deploy.outputs.url }}