Refactor web-app structure and add chart and CI tests

stackhpc · Oct 29, 2024 · 27887b9 · 27887b9
1 parent ad11e2c
commit 27887b9
Show file tree

Hide file tree

Showing 24 changed files with 379 additions and 266 deletions.
diff --git a/.github/workflows/build-push-artifacts.yml b/.github/workflows/build-push-artifacts.yml
@@ -1,7 +1,6 @@
 # Adapted from https://github.com/stackhpc/azimuth/blob/master/.github/workflows/build-push-artifacts.yaml
 name: Publish artifacts
 # Run the tasks on every push
-# TODO: Add path filtering to only run on relevant changes
 on: push
 jobs:
   # Job to run change detection
@@ -24,9 +23,10 @@ jobs:
         id: filter
         with:
           base: ${{ github.ref_name }}
+          # TODO: Make image filters more granular
           filters: |
             images:
-              - 'images/**'
+              - 'web-apps/**'
             chart:
               - 'chart/**'
 
@@ -35,12 +35,11 @@ jobs:
     name: Build and push images
     runs-on: ubuntu-latest
     needs: changes
-    if: ${{ needs.changes.outputs.images == 'true' || github.ref_type == 'tag' }}
+    # if: ${{ needs.changes.outputs.images == 'true' || github.ref_type == 'tag' }}
     strategy:
       matrix:
         include:
-          # - component: api
-          - component: ui
+          - component: chat-interface
     steps:
       - name: Check out the repository
         uses: actions/checkout@v4
@@ -56,46 +55,45 @@ jobs:
         id: image-meta
         uses: docker/metadata-action@v5
         with:
-          images: ghcr.io/stackhpc/azimuth-llm-${{ matrix.component }}-base
+          images: ghcr.io/stackhpc/azimuth-llm-${{ matrix.component }}
           # Produce the branch name or tag and the SHA as tags
           tags: |
             type=ref,event=branch
             type=ref,event=tag
             type=sha,prefix=
 
       - name: Build and push image
-        uses: stackhpc/github-actions/docker-multiarch-build-push@allow-continue-after-scan
+        uses: azimuth-cloud/github-actions/docker-multiarch-build-push@update-trivy-action
         with:
-          cache-key: ${{ matrix.component }}-base
-          context: ./images/${{ matrix.component }}-base
+          cache-key: ${{ matrix.component }}
+          context: ./web-apps/${{ matrix.component }}
           platforms: linux/amd64,linux/arm64
           push: true
           tags: ${{ steps.image-meta.outputs.tags }}
           labels: ${{ steps.image-meta.outputs.labels }}
-          fail_on_high_severity_cve: false
 
-  # Job to build and publish Helm chart
-  build_push_chart:
-    name: Build and push Helm chart
-    runs-on: ubuntu-latest
-    # Only build and push the chart if chart files have changed
-    needs: [changes]
-    if: ${{ needs.changes.outputs.chart == 'true' || github.ref_type == 'tag' }}
-    steps:
-      - name: Check out the repository
-        uses: actions/checkout@v4
-        with:
-          # This is important for the semver action to work correctly
-          # when determining the number of commits since the last tag
-          fetch-depth: 0
+  # # Job to build and publish Helm chart
+  # build_push_chart:
+  #   name: Build and push Helm chart
+  #   runs-on: ubuntu-latest
+  #   # Only build and push the chart if chart files have changed
+  #   needs: [changes]
+  #   if: ${{ needs.changes.outputs.chart == 'true' || github.ref_type == 'tag' }}
+  #   steps:
+  #     - name: Check out the repository
+  #       uses: actions/checkout@v4
+  #       with:
+  #         # This is important for the semver action to work correctly
+  #         # when determining the number of commits since the last tag
+  #         fetch-depth: 0
 
-      - name: Get SemVer version for current commit
-        id: semver
-        uses: stackhpc/github-actions/semver@master
+  #     - name: Get SemVer version for current commit
+  #       id: semver
+  #       uses: azimuth-cloud/github-actions/semver@master
 
-      - name: Publish Helm charts
-        uses: stackhpc/github-actions/helm-publish@master
-        with:
-          token: ${{ secrets.GITHUB_TOKEN }}
-          version: ${{ steps.semver.outputs.version }}
-          app-version: ${{ steps.semver.outputs.short-sha }}
+  #     - name: Publish Helm charts
+  #       uses: azimuth-cloud/github-actions/helm-publish@master
+  #       with:
+  #         token: ${{ secrets.GITHUB_TOKEN }}
+  #         version: ${{ steps.semver.outputs.version }}
+  #         app-version: ${{ steps.semver.outputs.short-sha }}
diff --git a/.github/workflows/test-pr.yml b/.github/workflows/test-pr.yml
@@ -0,0 +1,50 @@
+name: Test pull request
+on:
+  pull_request:
+jobs:
+  chart_validation:
+    runs-on: ubuntu-latest
+    env:
+      CLUSTER_NAME: chart-testing
+      RELEASE_NAME: ci-test
+    steps:
+      - name: Checkout the repo
+        uses: actions/checkout@v4
+
+      - name: Install Helm
+        uses: azure/setup-helm@v4
+
+      - name: Set up chart testing
+        uses: helm/chart-testing-action@v2
+
+      - name: Run chart linting
+        run: ct lint --config ct.yaml
+
+      - name: Run helm template with default values
+        run: helm template ci-test .
+        working-directory: chart
+
+      - name: Build all web app images
+        run: ./build.sh
+        working-directory: web-apps
+
+      - name: Create Kind Cluster
+        uses: helm/kind-action@v1
+        with:
+          cluster-name: ${{ env.CLUSTER_NAME }}
+
+      - name: Load local docker images into kind cluster
+        run: |
+          IMAGE_NAMES=$(docker image ls | grep ghcr.io/stackhpc/azimuth-llm- | awk '{print $1}')
+          kind load docker-image $IMAGE_NAMES -n ${{ env.CLUSTER_NAME }}
+
+      - name: Add Helm repos for dependencies
+        run: |
+          helm repo add stakater https://stakater.github.io/stakater-charts
+
+      # https://github.com/helm/charts/blob/master/test/README.md#providing-custom-test-values
+      # Each chart/ci/*-values.yaml file will be treated as a separate test case with it's
+      # own helm install/test process.
+      - name: Run chart install and test
+        run: ct install --config ct.yaml
+
diff --git a/.gitignore b/.gitignore
@@ -8,11 +8,8 @@ __pycache__/
 
 # Ignore local dev helpers
 test-values.y[a]ml
-chart/web-app/settings.yml
-gradio-client-test.py
 **venv*/
 
-
 # Helm chart stuff
 chart/Chart.lock
-chart/charts
+chart/charts
diff --git a/chart/.helmignore b/chart/.helmignore
@@ -30,7 +30,7 @@ venv/
 __pycache__/
 images/
 .hf-token.secret
-hu-poc/
 test-values.yaml
 web-app/settings.yml
-web-app/example-settings.yml
+web-app/example-settings.yml
+ci/
diff --git a/chart/Chart.yaml b/chart/Chart.yaml
@@ -1,6 +1,10 @@
 apiVersion: v2
 name: azimuth-llm
 description: HuggingFace LLM model serving along with a simple web interface.
+maintainers:
+  - name: "Scott Davidson"
+    email: [email protected]
+    url: https://github.com/sd109
 
 # A chart can be either an 'application' or a 'library' chart.
 #
@@ -32,4 +36,4 @@ dependencies:
   - name: reloader
     version: 1.0.63
     repository: https://stakater.github.io/stakater-charts
-    condition: ui.enabled
+    condition: ui.enabled
diff --git a/chart/ci/web-apps-only-values.yaml b/chart/ci/web-apps-only-values.yaml
@@ -0,0 +1,13 @@
+api:
+  enabled: false
+ui:
+  image:
+    repository: ghcr.io/stackhpc/azimuth-llm-chat-interface
+    tag: latest
+    imagePullPolicy: Never
+  service:
+    zenith:
+      enabled: false
+  appSettings:
+    hf_model_name: AMead10/c4ai-command-r-08-2024-awq
+    backend_url: https://llm3-compute.cms.hu-berlin.de
diff --git a/chart/templates/test/end-to-end.yml b/chart/templates/test/end-to-end.yml
@@ -0,0 +1,23 @@
+{{- if .Values.api.enabled -}}
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: gradio-client-response
+  annotations:
+    "helm.sh/hook": test
+spec:
+  template:
+    spec:
+      containers:
+      - name: gradio-client-test
+        image: ghcr.io/stackhpc/azimuth-llm-chat-interface
+        imagePullPolicy: IfNotPresent
+        command:
+        - python
+        - gradio-client-test.py
+        - http://{{ .Values.ui.service.name }}.{{ .Release.Namespace }}.svc
+      restartPolicy: Never
+  # Allow plenty of retries since downloading
+  # model weights can take a long time.
+  backoffLimit: 10
+{{- end -}}
diff --git a/chart/templates/test/web-app.yml b/chart/templates/test/web-app.yml
@@ -0,0 +1,22 @@
+{{- if not .Values.api.enabled -}}
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: web-app-liveness
+  annotations:
+    "helm.sh/hook": test
+spec:
+  template:
+    spec:
+      containers:
+      - name: web-app-test
+        image: nicolaka/netshoot:v0.12
+        imagePullPolicy: IfNotPresent
+        command:
+        - nc
+        - -vz
+        - {{ .Values.ui.service.name }}.{{ .Release.Namespace }}.svc
+        - "80"
+      restartPolicy: Never
+  backoffLimit: 3
+{{- end -}}
diff --git a/chart/templates/ui/app-config-map.yml b/chart/templates/ui/app-config-map.yml
@@ -6,7 +6,6 @@ metadata:
   labels:
     {{- include "azimuth-llm.labels" . | nindent 4 }}
 data:
-{{ (.Files.Glob "web-app/*").AsConfig | nindent 2 }}
-  settings.yml: |
+  overrides.yml: |
     {{- .Values.ui.appSettings | toYaml | nindent 4 }}
-{{- end -}}
+{{- end -}}
diff --git a/chart/templates/ui/deployment.yml b/chart/templates/ui/deployment.yml
@@ -23,25 +23,20 @@ spec:
       containers:
       - name: {{ .Release.Name }}-ui
         image: {{ printf "%s:%s" .Values.ui.image.repository .Values.ui.image.version }}
+        imagePullPolicy: {{ .Values.ui.image.imagePullPolicy }}
         ports:
         - name: ui
-          containerPort: 7680
-        workingDir: /etc/web-app
+          containerPort: 7860
         volumeMounts:
         - name: app
           mountPath: /etc/web-app
-        command:
-        - python
-        args:
-        - {{ .Values.ui.entrypoint }}
-        - {{ .Values.huggingface.model }}
         env:
         - name: PYTHONUNBUFFERED
           value: "1"
         tty: true # Make stdout from python visible in k8s logs
         readinessProbe:
           tcpSocket:
-            port: 7680
+            port: 7860
           periodSeconds: 5
       volumes:
         - name: app

diff --git a/chart/values.yaml b/chart/values.yaml
@@ -100,8 +100,11 @@ ui:
         - Arial
   # Container image config
   image:
-    repository: ghcr.io/stackhpc/azimuth-llm-ui-base
-    version: "0.3.0"
+    # repository: ghcr.io/stackhpc/azimuth-llm-ui-base
+    # version: "0.3.0"
+    repository: azimuth-llm-chat-interface
+    version: latest
+    imagePullPolicy: IfNotPresent
   # Service config
   service:
     name: web-app

diff --git a/chart/web-app/config.py b/chart/web-app/config.py