Skip to content

Pre-installation Checklist

Before proceeding with your on-prem installation with or without installation support from Cognigy side, make sure the criteria from the following checklist are met:

  1. Read through the installation instructions in README.md files in both helm charts (MongoDB and Cognigy.AI) in advance and complete necessary preparations.

  2. Prerequisites for MongoDB Helm Chart installation are met (see README.md file in MongoDB Helm Chart for details).

  3. Prerequisites for Cognigy.AI Helm Chart installation are met (see README.md file in Cognigy.AI Helm Chart for details).

  4. Configuration sections for both Helm charts are accomplished and values.yaml files for both charts are prepared and prepopulated with necessary data (see README.md files in MongoDB and Cognigy.AI Helm Charts for details).

  5. Team has access to the kubernetes cluster via kubectl.

  6. kubectl and helm utilities are installed with required versions.

  7. Kubernetes cluster has enough resources: at least 6 x worker nodes (8 CPU, 16 GB RAM, x86_64 architecture, 100 GB root SSD storage).

  8. Kubernetes worker nodes are distributed across 3 Availability Zones (AZ) for high-availability setup if possible.

  9. Cognigy.AI license key is available.

  10. Cognigy Container image registry credentials are known.

  11. Domains / DNS names for Cognigy.AI URLs are created and available.

  12. Wildcard TLS certificate(s) are ready.

  13. IP whitelisting for Cognigy’s container registry (cognigy.azurecr.io:443) and billing server (billing.cognigy.ai:443) is complete.

  14. Network access is allowed for HTTP/HTTPS requests.

  15. Websocket protocol is supported and can pass through firewall(s).

  16. Storage requirements are met by the underlying cloud infrastructure:

    • Custom StorageClass is provisioned for MongoDB.
    • Block storage (disks) can be provisioned dynamically for MongoDB and Redis.
    • (AWS only): EFS shares for flow-modules and functions are provisioned in advance.
    • (AWS only): EFS shares for flow-modules and functions are reachable from the EKS cluster.
    • (Azure only): Azure File shares can be provisioned dynamically.