新手如何在Docker容器中启动Apache NiFi并自动运行已有模板

Hey there! Since you're new to both containerization and Apache NiFi, let's walk through a couple of straightforward methods to get your existing template running automatically every time your NiFi container boots up. I'll break this down into actionable steps so you can follow along easily.

Method 1: Use a Custom Startup Script with NiFi's REST API #

This method is flexible—great if you want to keep using your existing template and have control over the flow deployment process.

Step 1: Prepare Your Template & Script #

1. Make sure your NiFi template is saved as an XML file (e.g., my_workflow_template.xml).
2. Create a bash script named start_nifi_with_template.sh with the following content:

   #!/bin/bash
   
   # Start the NiFi service in the background
   /opt/nifi/scripts/start.sh &
   
   # Wait for NiFi to fully initialize (more reliable than fixed sleep)
   echo "Waiting for NiFi to start up..."
   until curl -s http://localhost:8080/nifi-api/system-diagnostics > /dev/null; do
     sleep 10
   done
   
   # Upload your template to NiFi
   curl -X POST -F "template=@/opt/nifi/templates/my_workflow_template.xml" http://localhost:8080/nifi-api/templates/upload
   
   # Fetch the template ID (requires jq for JSON parsing)
   TEMPLATE_ID=$(curl -s http://localhost:8080/nifi-api/templates | jq -r '.templates[] | select(.name=="YOUR_TEMPLATE_NAME") | .id')
   
   # Instantiate the template in the root process group
   curl -X POST -H "Content-Type: application/json" -d '{
     "templateId": "'"$TEMPLATE_ID"'",
     "originX": 100,
     "originY": 100
   }' http://localhost:8080/nifi-api/process-groups/root/template-instance
   
   # Start all components in the root process group
   curl -X PUT -H "Content-Type: application/json" -d '{
     "id": "root",
     "state": "RUNNING"
   }' http://localhost:8080/nifi-api/process-groups/root/state
   
   # Keep the container running by attaching to the NiFi process
   wait
   

   • Replace YOUR_TEMPLATE_NAME with the exact, case-sensitive name of your NiFi template.
   • The until loop ensures we only proceed once NiFi's API is responsive, avoiding race conditions.

Step 2: Build a Custom NiFi Docker Image #

The official NiFi image doesn't include jq (needed for JSON parsing), so we'll create a custom image:

1. Create a Dockerfile in the same directory as your template and script:

   FROM apache/nifi:latest
   
   USER root
   
   # Install jq for JSON parsing
   RUN apt-get update && apt-get install -y jq && rm -rf /var/lib/apt/lists/*
   
   # Copy your template and startup script into the container
   COPY my_workflow_template.xml /opt/nifi/templates/
   COPY start_nifi_with_template.sh /opt/nifi/scripts/
   
   # Make the script executable
   RUN chmod +x /opt/nifi/scripts/start_nifi_with_template.sh
   
   USER nifi
   
   # Set our custom script as the entrypoint
   ENTRYPOINT ["/opt/nifi/scripts/start_nifi_with_template.sh"]
   

2. Build the image:

   docker build -t my-nifi-auto-template .
   

3. Run the container:

   docker run -d -p 8080:8080 my-nifi-auto-template
   

Method 2: Mount a Preconfigured flow.xml.gz #

If you've already set up and tested your flow in a NiFi instance, you can directly use its native configuration file (flow.xml.gz):

1. From your existing NiFi instance, locate the flow.xml.gz file (usually in the nifi/conf/ directory).
2. Run the official NiFi container, mounting your custom flow.xml.gz:

   docker run -d \
     -p 8080:8080 \
     -v /path/to/your/flow.xml.gz:/opt/nifi/nifi-current/conf/flow.xml.gz \
     apache/nifi:latest
   

   • NiFi will load this configuration on startup, and if your flow was running when you exported flow.xml.gz, it will automatically resume operations.

Key Notes #

• Permissions: Ensure mounted files (template, script, flow.xml.gz) have permissions compatible with the nifi user inside the container.
• Startup Time: NiFi can take 2-5 minutes to initialize on first run—adjust the wait loop timing if needed.
• Template Updates: If you modify your template, re-run the Docker build process (for Method 1) or re-export flow.xml.gz (for Method 2) to apply changes.

内容的提问来源于stack exchange，提问作者R.Dyer-Evans