Deploy demo (#1837)

* Add GH Actions workflow for demo deployment

* update demo ec2 instance type

* remove redundant docker-compose build

* add custom demo command and env vars

* deploy demo on updates to workflow resources

.github/workflows/demo.yaml

@@ -0,0 +1,40 @@
+name: Demo
+
+on:
+  workflow_dispatch:
+  push:
+    branches:
+      - master
+    paths:
+      - ".github/workflows/demo**"
+
+env:
+  AWS_REGION: eu-west-1
+
+permissions:
+  id-token: write
+  contents: read
+
+jobs:
+  deploy:
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/checkout@v2
+
+      - name: AWS Authentication
+        uses: aws-actions/configure-aws-credentials@master
+        with:
+          aws-region: ${{ env.AWS_REGION }}
+          role-to-assume: ${{ secrets.DEMO_AWS_DEPLOY_ROLE }}
+
+      - name: Deploy demo
+        env:
+          CF_KEY_NAME: ${{ secrets.DEMO_CF_KEY_NAME }}
+          CF_IMAGE_ID: ${{ secrets.DEMO_CF_IMAGE_ID }}
+          CF_IAM_INSTANCE_PROFILE: ${{ secrets.DEMO_CF_IAM_INSTANCE_PROFILE }}
+        run: |
+          aws cloudformation deploy \
+            --template-file .github/workflows/demo/ec2-autoscaling-group.yaml \
+            --stack-name haystack-demo-production-instance \
+            --parameter-overrides CommitShortSHA=${{ github.sha }} KeyName=${CF_KEY_NAME} ImageId=${CF_IMAGE_ID} IamInstanceProfile=${CF_IAM_INSTANCE_PROFILE} \
+            --capabilities CAPABILITY_IAM

.github/workflows/demo/docker-compose.demo.yml

@@ -0,0 +1,16 @@
+version: "3"
+services:
+  haystack-api:
+    restart: always
+    environment:
+      CONCURRENT_REQUEST_PER_WORKER: 16
+    command: "/bin/bash -c 'sleep 10 && gunicorn rest_api.application:app -b 0.0.0.0 -k uvicorn.workers.UvicornWorker --workers 3 --timeout 180'"
+
+  elasticsearch:
+    restart: always
+  ui:
+    restart: always
+    environment:
+      DEFAULT_DOCS_FROM_RETRIEVER: 7
+      DEFAULT_NUMBER_OF_ANSWERS: 5
+      DISABLE_FILE_UPLOAD: 1

.github/workflows/demo/ec2-autoscaling-group.yaml

@@ -0,0 +1,158 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/awslabs/goformation/v5.2.11/schema/cloudformation.schema.json
+
+
+Parameters:
+  Project:
+    Description: A project name that is used for resource names
+    Type: String
+    Default: haystack-demo
+
+  Environment:
+    Description: An environment name that is suffixed to resource names
+    Type: String
+    Default: production
+
+  VPCStack:
+    Description: VPC stack name
+    Type: String
+    Default: haystack-demo-production-vpc
+
+  CommitShortSHA:
+    Description: Commit short reference that triggered this deployment
+    Type: String
+
+  InstanceType:
+    Description: EC2 instance type
+    Type: String
+    Default: p3.2xlarge
+
+  ImageId:
+    Description: AMI to use for the EC2 instance
+    Type: String
+
+  IamInstanceProfile:
+    Description: IAM instance profile to attach to the EC2 instance
+    Type: String
+
+  KeyName:
+    Description: EC2 key pair to add to the EC2 instance
+    Type: String
+
+Resources:
+  AutoScalingGroup:
+    Type: AWS::AutoScaling::AutoScalingGroup
+    CreationPolicy:
+      ResourceSignal:
+        Count: "1"
+        Timeout: PT45M
+    UpdatePolicy:
+      AutoScalingRollingUpdate:
+        MinInstancesInService: "1"
+        MaxBatchSize: "1"
+        PauseTime: PT45M
+        WaitOnResourceSignals: true
+        SuspendProcesses:
+          - HealthCheck
+          - ReplaceUnhealthy
+          - AZRebalance
+          - AlarmNotification
+          - ScheduledActions
+    Properties:
+      LaunchConfigurationName: !Ref InstanceConfiguration
+      VPCZoneIdentifier:
+        - !ImportValue
+          "Fn::Sub": "${VPCStack}-PublicSubnet1"
+        - !ImportValue
+          "Fn::Sub": "${VPCStack}-PublicSubnet2"
+      MaxSize: "2"
+      DesiredCapacity: "1"
+      MinSize: "1"
+      TargetGroupARNs:
+        - !ImportValue
+          "Fn::Sub": "${VPCStack}-DefaultTargetGroup"
+      Tags:
+        - Key: Name
+          Value: !Sub ${Project}-${Environment}
+          PropagateAtLaunch: true
+        - Key: Environment
+          Value: !Ref Project
+          PropagateAtLaunch: true
+        - Key: Project
+          Value: !Ref Environment
+          PropagateAtLaunch: true
+        - Key: CommitShortSHA
+          Value: !Ref CommitShortSHA
+          PropagateAtLaunch: true
+
+  InstanceConfiguration:
+    Type: AWS::AutoScaling::LaunchConfiguration
+    Properties:
+      LaunchConfigurationName: !Sub ${Project}-${Environment}-${CommitShortSHA}
+      InstanceType: !Ref InstanceType
+      ImageId: !Ref ImageId
+      IamInstanceProfile: !Ref IamInstanceProfile
+      KeyName: !Ref KeyName
+      AssociatePublicIpAddress: true
+      SecurityGroups:
+        - !ImportValue
+          "Fn::Sub": "${VPCStack}-InstanceSecurityGroup"
+      EbsOptimized: true
+      BlockDeviceMappings:
+        - DeviceName: /dev/sda1
+          Ebs:
+            VolumeSize: 200
+
+      UserData:
+        Fn::Base64: !Sub |
+          #!/bin/bash -ex
+          mkdir -p /opt/aws/bin
+          wget https://s3.amazonaws.com/cloudformation-examples/aws-cfn-bootstrap-py3-latest.tar.gz
+          python3 -m easy_install --script-dir /opt/aws/bin aws-cfn-bootstrap-py3-latest.tar.gz
+
+          trap '/opt/aws/bin/cfn-signal --exit-code 1 --stack ${AWS::StackId} --resource AutoScalingGroup --region ${AWS::Region}' ERR
+
+          echo "Deploying Haystack demo, commit ${CommitShortSHA}"
+
+          echo 'APT::Periodic::Update-Package-Lists "0";
+          APT::Periodic::Unattended-Upgrade "0";' > /etc/apt/apt.conf.d/20auto-upgrades
+
+          apt update
+          apt install -y curl git ca-certificates curl gnupg lsb-release
+          curl -fsSL https://download.docker.com/linux/ubuntu/gpg | gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg
+          echo \
+              "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/docker-archive-keyring.gpg] https://download.docker.com/linux/ubuntu \
+              $(lsb_release -cs) stable" | tee /etc/apt/sources.list.d/docker.list > /dev/null
+          apt update
+          apt install -y docker-ce docker-ce-cli containerd.io
+
+          # Install Docker compose
+          curl -L "https://github.com/docker/compose/releases/download/1.29.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/bin/docker-compose
+          chmod +x /usr/bin/docker-compose
+
+          # Install Nvidia container runtime
+          curl -s -L https://nvidia.github.io/nvidia-container-runtime/gpgkey | \
+            apt-key add -
+          distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
+          curl -s -L https://nvidia.github.io/nvidia-container-runtime/$distribution/nvidia-container-runtime.list | \
+            tee /etc/apt/sources.list.d/nvidia-container-runtime.list
+          apt-get update
+          apt-get install -y nvidia-container-runtime
+
+          # Setup and start Docker
+          groupadd docker || true
+          usermod -aG docker $USER || true
+          newgrp docker || true
+          systemctl unmask docker
+          systemctl restart docker
+
+          # Exposes the GPUs to Docker
+          docker run --rm --gpus all ubuntu nvidia-smi
+
+          # Clone and start Haystack
+          git clone --branch deploy-demo https://github.com/askainet/haystack.git /opt/haystack
+          cd /opt/haystack
+          export COMPOSE_FILE=docker-compose-gpu.yml:.github/workflows/demo/docker-compose.demo.yml
+          docker-compose pull
+          docker-compose up -d
+
+          /opt/aws/bin/cfn-signal --exit-code $? --stack ${AWS::StackId} --resource AutoScalingGroup --region ${AWS::Region}

docker-compose-gpu.yml

@@ -32,7 +32,7 @@ services:
   elasticsearch:
     # This will start an empty elasticsearch instance (so you have to add your documents yourself)
     #image: "elasticsearch:7.9.2"
-    # If you want a demo image instead that is "ready-to-query" with some indexed articles 
+    # If you want a demo image instead that is "ready-to-query" with some indexed articles
     # about countries and capital cities from Wikipedia:
     image: "deepset/elasticsearch-countries-and-capitals"
     ports:

docker-compose.yml

@@ -22,7 +22,7 @@ services:
   elasticsearch:
     # This will start an empty elasticsearch instance (so you have to add your documents yourself)
     #image: "elasticsearch:7.9.2"
-    # If you want a demo image instead that is "ready-to-query" with some indexed articles 
+    # If you want a demo image instead that is "ready-to-query" with some indexed articles
     # about countries and capital cities from Wikipedia:
     image: "deepset/elasticsearch-countries-and-capitals"
     ports:

ui/webapp.py

@@ -36,7 +36,7 @@ def main():
 
     # Persistent state
     state = SessionState.get(
-        random_question=DEFAULT_QUESTION_AT_STARTUP, 
+        random_question=DEFAULT_QUESTION_AT_STARTUP,
         random_answer="",
         last_question=DEFAULT_QUESTION_AT_STARTUP,
         results=None,
@@ -51,7 +51,7 @@ def main():
     # Title
     st.write("# Haystack Demo - Explore the world")
     st.markdown("""
-This demo takes its data from a selection of Wikipedia pages crawled in November 2021 on the topic of 
+This demo takes its data from a selection of Wikipedia pages crawled in November 2021 on the topic of
 
 <h3 style='text-align:center;padding: 0 0 1rem;'>Countries and capital cities</h3>
 
@@ -63,18 +63,18 @@ Ask any question on this topic and see if Haystack can find the correct answer t
     # Sidebar
     st.sidebar.header("Options")
     top_k_reader = st.sidebar.slider(
-        "Max. number of answers", 
-        min_value=1, 
-        max_value=10, 
-        value=DEFAULT_NUMBER_OF_ANSWERS, 
-        step=1, 
+        "Max. number of answers",
+        min_value=1,
+        max_value=10,
+        value=DEFAULT_NUMBER_OF_ANSWERS,
+        step=1,
         on_change=reset_results)
     top_k_retriever = st.sidebar.slider(
-        "Max. number of documents from retriever", 
-        min_value=1, 
-        max_value=10, 
-        value=DEFAULT_DOCS_FROM_RETRIEVER, 
-        step=1, 
+        "Max. number of documents from retriever",
+        min_value=1,
+        max_value=10,
+        value=DEFAULT_DOCS_FROM_RETRIEVER,
+        step=1,
         on_change=reset_results)
     eval_mode = st.sidebar.checkbox("Evaluation mode")
     debug = st.sidebar.checkbox("Show debug info")
@@ -107,7 +107,7 @@ Ask any question on this topic and see if Haystack can find the correct answer t
             text-align: center;
         }}
         .haystack-footer h4 {{
-            margin: 0.1rem; 
+            margin: 0.1rem;
             padding:0;
         }}
         footer {{
@@ -132,7 +132,7 @@ Ask any question on this topic and see if Haystack can find the correct answer t
     # Search bar
     question = st.text_input("",
         value=state.random_question,
-        max_chars=100, 
+        max_chars=100,
         on_change=reset_results
     )
     col1, col2 = st.columns(2)
@@ -147,7 +147,7 @@ Ask any question on this topic and see if Haystack can find the correct answer t
     #state.get_next_question = col2.button("Random question")
     if col2.button("Random question"):
         reset_results()
-        new_row = df.sample(1)   
+        new_row = df.sample(1)
         while new_row["Question Text"].values[0] == state.random_question:  # Avoid picking the same question twice (the change is not visible on the UI)
             new_row = df.sample(1)
         state.random_question = new_row["Question Text"].values[0]
@@ -200,7 +200,7 @@ Ask any question on this topic and see if Haystack can find the correct answer t
                 answer, context = result["answer"], result["context"]
                 start_idx = context.find(answer)
                 end_idx = start_idx + len(answer)
-                # Hack due to this bug: https://github.com/streamlit/streamlit/issues/3190 
+                # Hack due to this bug: https://github.com/streamlit/streamlit/issues/3190
                 st.write(markdown(context[:start_idx] + str(annotation(answer, "ANSWER", "#8ef")) + context[end_idx:]), unsafe_allow_html=True)
                 source = ""
                 url, title = get_backlink(result)
@@ -213,7 +213,7 @@ Ask any question on this topic and see if Haystack can find the correct answer t
             else:
                 st.info("🤔 &nbsp;&nbsp; Haystack is unsure whether any of the documents contain an answer to your question. Try to reformulate it!")
                 st.write("**Relevance:** ", result["relevance"])
-                
+
             if eval_mode and result["answer"]:
                 # Define columns for buttons
                 is_correct_answer = None