High-Level Summary
A new datumctl CLI plugin that provides operators and developers with a unified command-line interface to deploy, manage, and diagnose Galactic VPC and Edge networking resources. Instead of manually authoring and applying Kubernetes manifests for CRDs, NADs, BGP configurations, and DaemonSets, users interact with a single plugin to provision end-to-end VPC networks across multi-cloud clusters.
Motivation
Today, deploying a Galactic VPC requires operators to manually orchestrate five discrete steps: install Cosmos CRDs from an external repo, push kubeconfigs to worker nodes, apply FRR underlay DaemonSets, craft NetworkAttachmentDefinition YAML with base62-encoded VPC identifiers and SRv6 SIDs, and annotate pod deployments with Multus references. This workflow is error-prone, lacks validation, and has no built-in diagnostic path — operators debug misconfigurations by reading raw CRD state and kernel netlink output. A CLI plugin abstracts away manifest authoring, provides inline validation before deployment, and surfaces diagnostic commands for BGP peering, route reachability, and SRv6 SID state. This reduces the barrier to deploying multi-cloud VPC networks and gives operators a single tool for both provisioning and troubleshooting.
Goals
- Users can install the network CLI plugin via
datumctl install network (or equivalent plugin discovery mechanism).
- Users can define and deploy a Galactic VPC end-to-end with a single command, including CRD installation, NAD generation, BGP peer configuration, and DaemonSet deployment.
- Users can inspect the state of deployed VPCs, VPCAttachments, and BGP sessions via
datumctl network get-style subcommands.
- Users can run diagnostic commands to verify BGP peering health, SRv6 SID encoding, and cross-node route reachability without manually inspecting kernel state or GoBGP logs.
- Documentation exists covering plugin installation, VPC deployment workflows, and diagnostic command usage, publishable as part of the Datum Cloud operator guides.
Non-Goals
- Replacing the companion Cosmos operator — the plugin consumes VPC and VPCAttachment CRDs; it does not replace the operator's USID allocation or NAD generation logic.
- A GUI or web-based management interface for networking resources.
- IPAM pool management or subnet allocation logic — the plugin accepts IPAM parameters as input but does not own pool assignment.
- CNI plugin or galactic-router code changes — the plugin is a control-plane tool that drives existing Galactic components via CRDs and kubectl.
- Support for non-Galactic networking backends (e.g., Calico, Cilium).
High-Level Summary
A new
datumctlCLI plugin that provides operators and developers with a unified command-line interface to deploy, manage, and diagnose Galactic VPC and Edge networking resources. Instead of manually authoring and applying Kubernetes manifests for CRDs, NADs, BGP configurations, and DaemonSets, users interact with a single plugin to provision end-to-end VPC networks across multi-cloud clusters.Motivation
Today, deploying a Galactic VPC requires operators to manually orchestrate five discrete steps: install Cosmos CRDs from an external repo, push kubeconfigs to worker nodes, apply FRR underlay DaemonSets, craft NetworkAttachmentDefinition YAML with base62-encoded VPC identifiers and SRv6 SIDs, and annotate pod deployments with Multus references. This workflow is error-prone, lacks validation, and has no built-in diagnostic path — operators debug misconfigurations by reading raw CRD state and kernel netlink output. A CLI plugin abstracts away manifest authoring, provides inline validation before deployment, and surfaces diagnostic commands for BGP peering, route reachability, and SRv6 SID state. This reduces the barrier to deploying multi-cloud VPC networks and gives operators a single tool for both provisioning and troubleshooting.
Goals
datumctl install network(or equivalent plugin discovery mechanism).datumctl network get-style subcommands.Non-Goals