Every surface
in the console.
Grouped the way the console groups them, so this page is a map of what you will actually meet after you sign in rather than a second sales pitch.
The homepage argues the case for each of these. This page tells you where they live.
Read the case →Inventory
The daily view. What exists, how it is doing, and what to do about it.
What it watches →Aggregate counts across every node, a system alerts panel, node or project views, power actions and one-click terminal access.
Live CPU, memory, storage and network with minimum, average and maximum, top consumers and per-node status.
Every registered cluster with its status, API endpoint, platform and assigned zone, plus its lifecycle actions.
Every container and virtual machine across all nodes and projects, with live addresses and assigned profiles.
Clustering
What the cluster itself reports, rather than what the dashboard remembers.
Many clusters, one plane →Cluster member state, architecture and failure domains, including which member currently holds the database role.
Per-cluster server metadata and the hardware resources that cluster reports.
Asynchronous cluster tasks still running, with precise timestamps and execution state.
Warnings raised by the cluster, categorised by severity and linked to the resource they name.
A real shell inside the instance, as a floating card that survives navigating the rest of the dashboard.
A graphical console for virtual machines, with ISO attach, power control and boot-to-firmware.
Open sessions with protocol, node, instance, duration and transfer rate, plus a searchable history.
Reusable configuration templates applied at instance creation, with the instances currently using each one.
Networking
Every interface and every address, mapped to what holds it.
Everything else it manages →Create, edit and delete virtual networks, each with its own detail page and live state.
Named network access control rules, applied to instances and to networks.
Every assigned IPv4 and IPv6 address, mapped to its instance, MAC address and project.
Storage and images
Capacity you can see before it runs out, and images that arrive intact.
Everything else it manages →Provision and inspect storage pools with capacity reported per pool, across local and distributed backends.
Create, resize, attach and delete custom volumes, with snapshots and custom ISO attachments.
What each cluster holds, imports from a remote image server, and resumable uploads with crash recovery.
Client certificates, unrestricted tokens and pending access requests, tracked across the whole estate.
Node-scoped groups with resource-level permission assignment, from full administration down to read-only.
Named policies made of rules — view, operate, console, snapshot, file transfer — granted per cluster, project or instance.
Your own TLS client certificate for a target cluster and project, at the level and expiry you choose.
Cluster certificate status, issuer, expiry and days remaining, with renewal and regeneration in place.
Registered clusters on a map, with aggregate cores, memory, storage and GPUs summed per zone.
Per node: server and API version, authentication state, kernel, architecture, storage backend and firewall driver, alongside CPU, memory, storage and the GPU cards physically present with their PCI addresses.
System preferences, and cache management for the frontend and backend stores.
API keys with scope, level and expiry, plus an authenticated passthrough for your own tools.
A Model Context Protocol endpoint an AI assistant can query, bound to the same key scope and audit trail as any other caller.
Identity federation and role binding across the organizations you manage.
Logos, background, theme and accent colour, applied to everyone in the organization at once.
Host CPU, memory, partitions and network, plus the backend process footprint and runtime.
Thirty surfaces,
one browser session.
Nothing here is a separate product or a paid add-on. It is one self-hosted control plane in front of the clusters you already run.