Overview#

This guide provides detailed instructions for deploying GritivaCore in a production environment. It is intended for sysadmins or DevOps engineers who have shell access to at least one Linux VM with a public IP, root access, and a Cloudflare account.

Prerequisites#

• Hardware Sizing: For a Pro deployment, recommend at least 2 vCPU and 4GB RAM per VM. Each scope adds approximately 50MB, and each gmesh peer adds around 5MB.
• Pre-flight Checklist:
  • Outbound port 443 must be open.
  • Kernel version must be >= 5.15.
  • Ensure systemd, iproute2, and nftables are present.

Deployment Process#

Step 1: Preparation#

1. Install Agent on First VM: Run the following command:

   curl -fsSL https://get.gritiva.com/agent | sudo bash -s -- --token <ENROLL_TOKEN>

Step 2: Execution#

2. Verify Enrollment: Check the enrollment status with:

   gritivactl status

   Ensure it shows connected: yes within 30 seconds.

3. Create Your First Scope: Execute:

   gritivactl scope create my-app --cpu 2 --mem 4G

4. Attach a Public Domain: Use the command:

   gritivactl domain add app.example.com --scope my-app --port 8080

5. Add a Second VM and Join It: Repeat the curl install command on the second VM. The panel will auto-discover it.

Step 3: Verification#

• Ensure all components are functioning correctly and monitor the following:
  • NRestarts
  • Mesh peer state
  • Scope OOM (Out of Memory)
  • Cloudflare API rate limits

Troubleshooting#

• Common issues may arise during deployment. Ensure all prerequisites are met and verify network configurations.

Rollback Procedures#

• In case of deployment failure, revert to the previous stable configuration using your version control system.

Best Practices#

• Pin agent version per environment.
• Use separate enrollment tokens for each VM.
• Configure systemd settings: CPUWeight=300, IOWeight=300, Nice=-5 (already included in install.sh post-PR-#55).
• Regularly monitor the deployment for performance and stability.