K-Forge - User Guide (Node Details, Hardware, Power and Access)

0. Quick Overview

This guide covers the K-Forge screens you use around the BMO lifecycle: access and permissions, the dashboard controls, node details for management and server nodes, hardware inventory, BIOS configuration, firmware updates, and power actions.

K-Forge is the module for bare-metal server onboarding and lifecycle management.

The BMO lifecycle itself is documented on its own pages:

Dashboard reference:

Common workflows in this guide:

  • Validate management node details first

  • Validate server details from Hardware Inventory

  • Review BIOS configuration and firmware versions on the management node

  • Run node power actions

  • Open node Console to monitor jobs

How to read this guide:

  • Sections 4 through 9 are task-based runbooks you use during the lifecycle or for maintenance; they are not lifecycle stages.

  • Section 10 is outcome-based troubleshooting.

  • For onboarding a node for the first time, start with BMO - Bare-Metal Onboarding and return here to validate node details and hardware.

Common access checks

Use these checks when K-Forge does not open or the dashboard looks incomplete:

  1. Confirm your current role can access K-Forge and run node lifecycle actions.

  2. Refresh Control Center once and reopen K-Forge from the left navigation.

  3. If cards or rows are still missing, verify that at least one node has been added and that your scope includes the target facility.

1. Overview

K-Forge is where bare-metal nodes are onboarded and managed.

1.1. Core Terms (Read First)

  • BMC (Baseboard Management Controller): out-of-band management interface on the physical server used for remote power, console, and hardware control.

  • BMO (Bare Metal Orchestration): taking a physical server with a reachable BMC through to a working Karios node, and back again. Run from K-Forge.

  • Discovered: server is onboarded and reachable via BMC, but its hardware has not been revealed yet.

  • Ready: server has completed hardware reveal and is ready to be configured.

  • Configure: applies Ceph storage, the SDN network, and the rest of the node configuration; the node then becomes Configured and ready for production use.

For the lifecycle order, see BMO - Bare-Metal Onboarding.

1.2. What You Can Do from the K-Forge Dashboard

From the dashboard and node pages, you can:

  • Open node details by lifecycle state

  • Run power actions (power on, power off, power cycle, force off)

  • Review management node Hardware Inventory, BIOS Config, and Firmware Updates

  • Validate server hardware inventory

  • Open node console directly for troubleshooting and boot visibility

Adding nodes, hardware reveal, and configure are covered in BMO - Bare-Metal Onboarding; removing a node is covered in Decommission.

Note

Hardware Reveal is mandatory before a node can be configured. Rerun it after BIOS, firmware, or hardware changes before final inventory validation.

2. Prerequisites

Before using K-Forge, confirm:

  • You have BMC IP, username, password

  • Target host is powered on and reachable on the BMC/management network before Add Host

Warning

Invalid BMC details can prevent K-Forge workflows.

2.1. Required Access (RBAC)

Before running K-Forge workflows, confirm with your administrator that your account can access K-Forge and run these actions:

  • Add Host

  • Hardware Reveal

  • Configure

  • BIOS Configuration

  • Firmware Updates

  • Console

  • Decommission

3. Access K-Forge

When to Use:

Use this first before any node onboarding or maintenance workflow.

Purpose:

Open the K-Forge dashboard from the left navigation.

Steps:

  1. Log in to Control Center.

  2. Select K-Forge in the left navigation.

  3. Open the K-Forge dashboard.

Control Center left navigation showing K-Forge entry

Access K-Forge from the Control Center navigation.

What this screenshot shows:

  • Left sidebar navigation with the K-Forge entry icon.

What you can do from this screen:

  • Open K-Forge dashboard from navigation as the first step in all K-Forge workflows.

Expected Outcome:

  • K-Forge dashboard opens with summary cards and node table.

If this fails:

See Common access checks before retrying K-Forge access.

4. Understand the Dashboard (What to do first)

4.0. First Login Reality (Blank Dashboard)

When to Use:

Use this when K-Forge opens with no node rows visible.

Purpose:

Start node onboarding from an empty dashboard and reach Discovered state.

For first-time users, the K-Forge dashboard can be empty.

This is expected when no BMC endpoints are registered yet.

What to do next:

  1. Click Add Host on the dashboard and complete the form. The full procedure, including the BMC details it needs, is in BMO - Bare-Metal Onboarding.

  2. The node appears in Discovered and K-Forge starts the hardware reveal automatically, moving it to Ready without further action.

Tip

If no nodes are visible, your first required action is adding BMC details using the form.

Expected Outcome:

If this fails:

  • Confirm BMC IP, username, and password are valid.

  • Confirm BMC endpoint is reachable on management network.

  • Retry Add Host and refresh dashboard.

4.1. Read Summary Cards

When to Use:

Use this immediately after opening K-Forge dashboard.

Purpose:

Understand current lifecycle distribution before running actions.

Steps:

  1. Open Control Center -> K-Forge.

  2. Review each summary card value.

  3. Use card values to choose the first node set to process.

  • All States: total nodes currently visible in K-Forge.

  • Discovered: node is reachable and registered, but its hardware has not been revealed yet.

  • Ready: hardware reveal is complete and the node is ready to be configured.

  • In Progress: background workflow is currently running.

  • Configured: node integrated and operational.

Karios Forge dashboard

K-Forge dashboard with lifecycle summary and nodes table.

What this screenshot shows:

  • Summary cards for lifecycle visibility: All States, Discovered, Ready, In Progress, Configured

  • Lifecycle filter cards that control table scope

  • Nodes table with key columns (for example BMC IP, Vendor, Stage, Health, Console, Actions)

What you can do from this screen:

  • Select a lifecycle card to filter rows by stage.

  • Open node details from the node row (or eye icon).

  • Open node console from the Console column.

  • Start the stage-appropriate action from the row Action menu.

Expected Outcome:

  • You can identify which lifecycle stage requires action first.

If this fails:

  • Refresh dashboard data and recheck the summary cards.

  • Confirm nodes are visible in K-Forge and reload the page.

4.2. Use Lifecycle Filters (Cards)

When to Use:

Use this when you need to focus on one lifecycle stage at a time.

Purpose:

Filter the node table to the exact lifecycle stage you want to process.

Steps:

  1. From the dashboard summary cards (shown in 4.1), select one lifecycle card: All States, Discovered, Ready, In Progress, or Configured.

  2. Confirm the table now shows only nodes in that stage.

  3. Run the stage-appropriate action for those rows.

Expected Outcome:

  • Table rows match the selected lifecycle filter.

If this fails:

  • Clear and re-apply the filter.

  • Refresh the dashboard and retry.

4.2.1. Use Node Power Action Menu (If Available)

When to Use:

Use this when you need to run node power operations from the node details header.

Purpose:

Run safe power actions for the selected node.

Steps:

  1. Open K-Forge dashboard and open the target node details page.

  2. In the top-right power/action area, open the Action menu.

  3. Select the required power operation.

  4. Click Execute.

K-Forge node details showing power action menu

Node details power action menu.

What this screenshot shows:

  • Power status indicator in node details header.

  • Action dropdown with power operations.

  • Execute button to run selected operation.

What you can do from this screen:

  • Select and execute node power operations from the menu.

  • Validate node power behavior during lifecycle operations.

Expected Outcome:

  • Selected power operation is submitted for the node.

If this fails:

  • Refresh node details and retry from the same action menu.

  • Verify BMC connectivity for the target node.

4.2.2. Open Node Console from Dashboard

When to Use:

Use this when you need live boot/output visibility for a node before, during, or after lifecycle actions.

Purpose:

Open node console directly from the K-Forge dashboard without leaving table workflow.

Steps:

  1. Open K-Forge dashboard.

  2. Locate the node row in the table.

  3. Click the icon in the Console column.

  4. Observe boot/output state and return to dashboard for the next action.

Expected Outcome:

  • Console session opens for the selected node.

If this fails:

  • Refresh dashboard and retry from the same node row.

4.3. Open Dashboard Help

When to Use:

Use this when you need on-screen guidance for the K-Forge dashboard controls.

Purpose:

Open dashboard help content and review field/action definitions.

Steps:

  1. Open Control Center -> K-Forge dashboard.

  2. Click the help icon in the dashboard header.

  3. Review definitions for summary cards, filters, and table actions.

K-Forge dashboard help panel

Dashboard help view with operational guidance.

What this screenshot shows:

  • Help panel for dashboard cards, filters, and table actions.

  • On-screen definitions tied to dashboard UI elements.

What you can do from this screen:

  • Confirm the meaning of each dashboard card/filter before running lifecycle actions.

  • Validate action labels/intent before selecting row operations.

Expected Outcome:

  • Help content opens and explains dashboard elements before you proceed.

If this fails:

  • Refresh the page and reopen the dashboard.

5. Add a Host

Adding a host and taking it from Discovered to Configured is documented in BMO - Bare-Metal Onboarding.

Use that page for the Add Host form, hardware reveal, and configure; return here to validate node details and hardware inventory.

6. Review Management Node Details First

6.1. Path

Control Center -> K-Forge -> Dashboard -> Management server

Quick access:

K-Forge dashboard -> click the management node

Console quick access:

K-Forge table -> node row -> Console column icon

Use console when:

  • confirming boot device behavior

  • checking if node is responsive during power operations

6.2. Management Node Completion Check

Complete this check before reviewing server node details. The management node is the control-plane entry in K-Forge and should be validated first so server-node inventory is interpreted against the correct site and management context.

Steps:

  1. Open the management node from the K-Forge dashboard or left node list.

  2. Confirm the node is labeled as the management server.

  3. Confirm power state, BMC address, vendor, and health are expected.

  4. Complete the management-node Hardware Inventory, BIOS Config, and Firmware Updates checks below.

  5. Return to the server node only after the management node details are correct.

Expected Outcome:

  • Management node details are reviewed and no unresolved health or identity mismatch remains before server-node inventory validation.

If this fails:

  • Refresh K-Forge and reopen the management node.

  • Verify BMC connectivity and management-network reachability.

  • Resolve management-node identity or health issues before continuing.

6.3. Management Node Hardware Inventory

When to Use:

Use this after opening the management node details page and before changing BIOS or firmware settings.

Purpose:

Confirm the management node inventory is current before applying configuration or maintenance changes.

Steps:

  1. Open the management node details page.

  2. Open Hardware Inventory.

  3. Click Get latest hardware.

  4. Review device, network, storage, and component information.

Management node hardware inventory

Management node hardware inventory.

Expected Outcome:

  • Management node inventory is current and matches the expected hardware.

6.4. Management Node BIOS Config

When to Use:

Use this when management-node BIOS values must be reviewed or adjusted before final validation.

Purpose:

Confirm required virtualization and boot-related BIOS values on the management node.

Steps:

  1. Open the management node details page.

  2. Open BIOS Config.

  3. Review summary counts for total, enabled, disabled, and pending BIOS settings.

  4. Review the BIOS Configuration Management panel.

  5. Apply changes only during an approved maintenance window.

  6. Restart the server if BIOS changes require activation.

Management node BIOS configuration

Management node BIOS configuration.

Expected Outcome:

  • BIOS values are reviewed, required changes are applied, and the node returns to a stable state after any required reboot.

6.5. Management Node Firmware Updates

When to Use:

Use this when management-node BMC or BIOS firmware must be reviewed or updated.

Purpose:

Confirm current BMC/BIOS firmware versions and apply vendor-approved updates when required.

Steps:

  1. Open the management node details page.

  2. Open Firmware Updates.

  3. Use Firmware Downloads to access the vendor firmware source.

  4. Review current BMC and BIOS firmware model/version values.

  5. Upload only vendor-approved firmware for the exact server model.

  6. Refresh and verify firmware status and audit history after update.

Management node firmware updates

Management node firmware update overview.

Expected Outcome:

  • Firmware versions and health are verified, and any approved update completes successfully before server-node validation continues.

If this fails:

  • Re-check firmware file compatibility with the exact management-node model.

  • Confirm the management node is healthy before retrying the update.

  • Retry only inside an approved maintenance window.

6.6. Management Node Details Actions Reference

Location / Action

What It Does

When To Use

Hardware Inventory -> Get latest hardware

Pulls the latest management-node inventory data.

After reveal, BIOS changes, firmware changes, or physical hardware changes.

BIOS Config -> Apply Changes

Saves selected BIOS values as pending settings.

After approved BIOS value changes.

BIOS Config -> reboot/power action

Restarts the node so BIOS updates take effect.

When the UI indicates a reboot is required.

Firmware Updates -> Refresh

Reloads firmware version and health from the node.

Before and after firmware work.

Firmware Updates -> upload/update

Uploads and applies vendor firmware binaries.

Approved BMC or BIOS firmware maintenance.

6.7. Typical Timing for Maintenance Actions

Action

Typical Time

Notes

BIOS apply + reboot

3-10 minutes

Management-node BIOS changes depend on vendor firmware behavior and POST time.

Firmware update

10-30+ minutes

Management-node firmware updates are vendor- and version-dependent; always use a maintenance window.

Hardware inventory refresh

1-3 minutes

Run after reveal, BIOS changes, firmware changes, or hardware changes.

Note

These are operator expectation ranges, not SLA guarantees. Firmware behavior or hardware validation can extend actual runtime.

7. Review Server Node Details

7.1. Path

Control Center -> K-Forge -> Node Details -> Hardware Inventory

Quick access:

K-Forge dashboard -> click target server node

7.2. Current Server Details Layout

The current server node details page exposes the Hardware Inventory tab only. It does not show separate BIOS Config or Firmware Updates tabs.

K-Forge server node details hardware inventory

Server node details with Hardware Inventory selected.

UI components in this screen:

  • Header: node breadcrumb, node name, power state, and favorite

  • Node summary: BMC IP and Vendor

  • Active tab: Hardware Inventory

  • Get latest hardware button: refreshes inventory from the selected node

  • Device Information: device ID, name, type, manufacturer, status, role, and U height

  • Location Information: site, location, rack, and position

  • Network Interfaces: BMC, bridge, and physical interface rows with MAC, status, description, and IP data

  • Storage Devices: discovered disks with role, manufacturer, and description

  • Other Inventory Items: additional discovered hardware components

What this screenshot shows:

  • Server node details currently focus on hardware inventory validation.

  • The available detail tab is Hardware Inventory.

  • The header provides power state visibility for the node.

What you can do from this screen:

  • Click Get latest hardware after Hardware Reveal or hardware replacement.

  • Confirm identity, BMC address, vendor, location, network interfaces, and storage devices.

Expected Outcome:

  • Node details match the physical server and site inventory records.

If this fails:

  • Refresh node details and click Get latest hardware again.

  • Verify BMC connectivity for the target node.

  • Re-run Hardware Reveal from the dashboard row action if inventory is stale.

7.3. Server Node Details Actions Reference (New User)

Use this section when you are already inside a node details page.

Location / Action

What It Does

When To Use

Expected Result

Dashboard Row -> Action -> Re-run Hardware Reveal

Runs the discovery boot again on a revealed node.

When a reveal returned incomplete or wrong hardware. The first reveal runs automatically when the host is added — see BMO - Bare-Metal Onboarding.

Inventory is collected again and the node returns to Ready.

Hardware Inventory -> Get latest hardware

Pulls latest inventory data from the node.

After reveal or hardware changes.

Updated component data appears.

Dashboard Row -> Action -> Decommission

Full reset: removes all configuration and wipes disks. See Decommission.

Resetting or recycling a host completely.

Node returns to Discovered and stays in inventory.

8. Run Hardware Reveal

Hardware reveal is part of the onboarding flow and is documented in BMO - Bare-Metal Onboarding.

Return here once the node reaches Ready to validate the collected inventory (Section 9).

9. Validate Hardware Inventory

This step follows hardware reveal, which is documented in BMO - Bare-Metal Onboarding.

9.1. Path

Control Center -> K-Forge -> Dashboard -> Select Node -> Hardware Inventory

Quick access:

K-Forge dashboard -> click target node -> Hardware Inventory

When to Use:

Use this immediately after completing Hardware Reveal or replacing server hardware.

Purpose:

Confirm discovered hardware and network inventory matches the physical server.

Steps:

  1. Open the target node details page.

  2. Click Get latest hardware.

  3. Review device, location, network interface, storage device, and other inventory sections.

Use the server node details screen shown in Section 7.2.

What this screenshot shows:

  • Server node details with the Hardware Inventory tab selected.

  • Hardware sections shown include device information, location information, network interfaces, storage devices, and other inventory items.

What you can do from this screen:

  • Click Get latest hardware and review collected inventory for the selected node.

  • Validate hardware details after a reveal run.

UI components in this screen:

  • Get latest hardware button: pulls latest inventory from the node

  • Device Information section

  • Location Information section

  • Network interfaces section

  • Storage Devices section

  • Other Inventory Items section

Hardware Inventory actions (what and why):

  • Get latest hardware: reloads latest inventory data for the node. Why: use after Hardware Reveal or physical hardware changes.

9.2. What to Validate

  • Device identity and vendor are correct

  • Location, rack, and position are correct

  • Network interface details are correct

  • Storage device details are correct

  • Other inventory items are present as expected

Expected Outcome:

  • Inventory matches physical server.

Action checklist on this page:

  1. Click Get latest hardware after Hardware Reveal completes.

  2. Verify timestamp updated.

  3. Review device, location, network interface, storage device, and other inventory sections.

  4. If mismatch is found, return to Hardware Reveal and rerun discovery.

If this fails:

  • Re-run Hardware Reveal

  • Verify the physical device and BMC target are correct

  • Click Get latest hardware again after 5 minutes

10. Quick Troubleshooting by Outcome

10.1. Node cannot be added

  • Check BMC IP reachability

  • Check BMC credentials

  • Check management VLAN/firewall

10.2. Power actions fail

  • Validate BMC login directly

  • Retry after session refresh

  • Verify node/BMC is not in error state

  • Open node console from the Console column icon to verify host responsiveness

10.3. Reveal has no data

  • Select Re-run Hardware Reveal from the row Action menu

  • The first reveal runs automatically when a host is added; this action re-runs it on a node that already reports Ready

  • Use the console icon to confirm action progress/output

  • Refresh dashboard and confirm stage transition

10.4. Inventory incomplete

  • Re-run reveal

  • Verify the node details page points to the expected server

  • Click Get latest hardware and wait for refresh completion

11. Quick Action Map

  • Add host: Control Center -> K-Forge -> Add Host

  • Open node: Control Center -> K-Forge -> click node

  • Open console: K-Forge table -> Console column icon

  • Inventory: Node details -> Hardware Inventory

  • Management node BIOS: Node details -> BIOS Config

  • Management node firmware: Node details -> Firmware Updates

  • Onboarding flow: see BMO - Bare-Metal Onboarding

  • Decommission: see Decommission

12. First-Run Validation Checklist

Use this checklist after a node has been onboarded with BMO - Bare-Metal Onboarding, to confirm it is fit for normal operations.

12.1. Validation Sequence

  1. Confirm the node appears in Configured on the K-Forge dashboard.

  2. Open and complete the management node details checks from Sections 6.2 through 6.5.

  3. Open the server node details page from Section 7.2.

  4. Validate Hardware Inventory in Section 9.

  5. Open the Console column icon and verify node boot/output health.

  6. Confirm health is green and inventory matches hardware.

12.2. Done Criteria (New User Success)

A node is ready for normal operations when all are true:

  • Stage is Configured

  • Health is green/OK

  • Hardware Reveal completed successfully

  • Hardware Inventory matches expected components

To reset a node later, see Decommission.

→ Next: Network