Bring your own Kubernetes (BYOK) system requirements

Chef 360 Platform Server has the following requirements for a bring your own Kubernetes (BYOK) deployment.

Hardware

Chef 360 Platform supports single-node and multi-node deployments. Select a topology based on your availability and scalability requirements.

For production environments, run a benchmark test to determine your system’s requirements. The benchmark test should include the number of nodes you plan to enroll, job frequency, output size, job duration, and check-in frequency.

Single-node requirements

A single-node Chef 360 Platform deployment (hyperconverged non-HA) has the following minimum requirements. Adjust these values based on your specific usage patterns and workload. For sizing recommendations tailored to your environment, contact your Customer Architect or Customer Success Manager.

vCPU	Memory	Storage
16	32 GB	200 GB

Multi-node requirements

All nodes must meet or exceed the requirements for their assigned role. Minimum node counts and node sizing requirements must both be satisfied. Using fewer nodes with larger specifications doesn’t replace the required node count. Node roles must be deployed exactly as defined for each topology.

These requirements support reliable operation and high availability of the platform.

Multi-node systems have the following minimum requirements:

Topology	Roles	Nodes	vCPU	Memory (GB)	Disk (GB)
Hyperconverged-HA	Controller + Frontend + Backend	3	16	32	200
Tiered-HA	Controller + Backend	3	16	32	200
	Frontend	3 or more	8	16	50
Hyperscale-HA	Controller	3	4	16	50
	Frontend	3 or more	8	16	50
	Backend	3	16	32	200

Node sizing requirements can vary based on workload characteristics, scale expectations, performance objectives, availability requirements, and integration patterns. The requirements documented here represent a baseline configuration. Work with a Chef 360 Platform Architect to validate and refine node sizing and ensure your deployment meets the specific needs of your environment.

For more information about cluster topologies and adding nodes, see Cluster management.

File system requirements

Chef 360 Platform has the following file system requirements:

• A mounted XFS filesystem with the ftype=1 option. This is the default in recent RHEL versions.
• The /var directory isn’t mounted with the noexec option.

Ports

Chef 360 Platform requires the following ports for all deployments. Open the following ports if you are using default ports.

Ports for inbound connections:

Port	Description
22	SSH
5985-5986	WinRM
30000	Chef 360 Platform Console
31000	API Gateway
31050	RabbitMQ
31101	Mailpit (optional)

Ports for outbound connections:

Port	Description
443	For non-air gapped installations

Ports for multi-node deployment

Multi-node deployments require additional ports for node-to-node communication. Create firewall rules to allow bidirectional traffic between nodes on these ports.

Port	Description
2380	etcd server client API (TCP)
4789	Flannel VXLAN overlay network (UDP)
6443	Kubernetes API server (TCP)
9091	Prometheus metrics (TCP)
9443	Webhook server (TCP)
10249	kube-proxy (TCP)
10250	Kubelet API (TCP)
10256	kube-proxy health check and metrics (TCP)
30000	Admin Console (TCP), required for nodes joining the cluster

Fully qualified domain name

Chef 360 Platform Server requires a fully qualified domain name (FQDN) that’s RFC 1123 compliant and registered with the Domain Name System (DNS).

Disaster recovery and data backup

Chef 360 Platform’s built-in disaster recovery features aren’t supported in BYOK deployments. Your organization is responsible for backup and recovery strategies for both the Kubernetes cluster and Chef 360 Platform’s data.

Kubernetes cluster

Chef 360 Platform supports the following Kubernetes platforms:

• Amazon EKS: Elastic Kubernetes Service on AWS
• Azure AKS: Azure Kubernetes Service
• Red Hat OpenShift: Version 4.10 or later
• Generic Kubernetes: Version 1.29 or later (other managed or self-hosted clusters)

You must have the following administrative access on the cluster:

• kubectl version 1.30 or later with context set to target the cluster from your workstation
• Administrative access to create namespaces, CustomResourceDefinitions, and cluster-scoped roles or role bindings

Air-gapped cluster

For air-gapped deployments, the following network and infrastructure requirements apply.

Network isolation:

• The Kubernetes cluster has no direct internet connectivity.
• Cluster nodes can communicate with each other over the internal network.
• If the cluster’s workstation is also air-gapped, you must have a secure file-transfer mechanism to transfer artifacts to the workstation.

Infrastructure requirements:

• A private container registry accessible from all Kubernetes nodes. Supported options include Docker Registry, Harbor, or a similar registry. The registry must have sufficient storage capacity for Chef 360 Platform images.

• A local file server or object storage to host installation artifacts. Supported options include:

  • An HTTP/HTTPS file server
  • An S3-compatible object storage service
  • A network file share accessible from the installation host

Internet-connected cluster

If you’re deploying Chef 360 Platform on an internet-connected cluster, the cluster must have outbound connectivity to the following Chef 360 Platform services:

• Container registry: registry.chef360.chef.io
• Downloads portal: download.chef360.chef.io
• Proxy registry services: proxy.chef360.chef.io
• Application services: appservice.chef360.chef.io
• Docker Hub domains:
  • index.docker.io
  • cdn.auth0.com
  • *.docker.io
  • *.docker.com

If your environment uses a corporate proxy, ensure it’s configured to allow access to these endpoints. Configure proxy settings in:

• Docker daemon on cluster nodes
• Kubernetes cluster configuration
• Container runtime (containerd, cri-o, etc.)