Compute - Virtual Machines

See also

Prerequisites for VM operations:

Note

In the current Compute section, Virtual Machines and Auto Scaling workflows are covered in this guide.

Warning

Live VM migration is not currently supported as a standard platform feature. Plan workload movement using controlled stop/start or cutover workflows.

Before You Begin - Prerequisites and Readiness Validation

This section validates all prerequisites before creating your first VM. Complete this checklist to ensure smooth VM provisioning.

See also

Cross-check related setup guides:

  • Getting Started for installation media and onboarding baseline

  • Infrastructure for zone/pod/cluster/host readiness

  • Network for optional custom network design (not required for first VM in standard setup)

  • Compute Policies for compute/disk profiles and access options

  • Storage for VM templates and boot images

Important

Seeing 0 VMs after initial Karios deployment is expected. Internal system VMs are not shown in the user VM dashboard.

Important

Live VM migration is not currently supported as a standard Karios platform feature. Plan maintenance and host recovery with controlled stop/start, reboot, HA restart behavior, or workload cutover workflows instead.

New-User Flow at a Glance

Use this simple sequence for first success:

  1. Validate prerequisites (instance profile, host up, VM template/boot image ready, instance storage up).

  2. Open Virtual Machines and click Create VM.

  3. Choose VM Template or Boot Image based on your requirement.

  4. Track VM state until Running.

  5. Open VM details and validate Details, Console, Metrics, and Karios Shield.

  6. Take a snapshot before high-risk changes.

  7. Use lifecycle actions (start/stop/reboot/delete) only after impact check.

Mandatory Requirements for VM Creation

Use this as the minimum go/no-go check before opening Create VM.

Required item

Minimum state

UI path

Action if missing

VM permissions

VM_VIEW and VM_MANAGE

User Icon -> My Profile

Request VM access from an administrator before continuing.

Instance profile

At least one enabled profile

Compute Policies -> Instance Profiles

Click + Create and save a usable CPU/RAM profile.

Host capacity

At least one host in Up state

Infrastructure -> Hosts

Onboard or repair a host until at least one host is Up.

Instance Storage

At least one storage pool in Up state

Storage -> Instance Storage

Add or repair an instance storage pool and confirm free capacity.

VM image

At least one VM template or boot image in Ready state

Storage -> VM Templates or Storage -> Boot Images

Register a VM template or upload a boot image, then wait for Ready.

Guest network

At least one usable guest network

Network -> Guest Networks or default Bootstrap network

Use the default Bootstrap guest network or create/repair a guest network before provisioning.

Note

For first-VM workflows in standard Bootstrap deployments, the default guest network is usually sufficient.

Quick Setup Path for Missing Prerequisites

Follow this order when one or more required items are missing:

  1. Open Compute Policies -> Instance Profiles and create one enabled instance profile.

  2. Open Infrastructure -> Hosts and confirm at least one host is Up.

  3. Open Storage -> Instance Storage and confirm at least one pool is Up with enough free capacity.

  4. Open Storage -> VM Templates or Storage -> Boot Images and make sure at least one image is Ready.

  5. Confirm a guest network is available, either from Bootstrap defaults or in Network -> Guest Networks.

  6. Return to Compute -> Virtual Machines and click Create VM.

Quick Start Validation

For experienced users, confirm these four items:

  1. [ ] At least one instance profile exists

  2. [ ] At least one host is Up

  3. [ ] At least one VM template Ready OR boot image uploaded

  4. [ ] Instance Storage shows Up state

Simple check steps:

Step 1 - Instance profile

Go to: Compute Policies -> Instance Profiles

What to confirm:

  • At least one row exists in the table.

  • Profile is enabled/active.

  • CPU and RAM values are visible (example: 1 CPU, 1 GB RAM).

Quick fix if missing:

  • Click + Create.

  • Set a name (example: small-1c-1g).

  • Set CPU/RAM (minimum test profile: 1 CPU, 1024 MB).

  • Save and confirm the profile appears enabled.

Full guide: Instance Profiles

Step 2 - Host availability

Go to: Infrastructure -> Hosts

What to confirm:

  • At least one host row exists.

  • Host state is Up (not Error or Disconnected).

  • Host has available CPU and memory.

Quick fix if missing:

  • Add/onboard a host through infrastructure workflow.

  • Resolve host connectivity/BMC issues until at least one host is Up.

  • Recheck host resource availability.

Full guide: Infrastructure

Step 3 - VM Template or Boot Image

Go to: Storage -> VM Templates or Storage -> Boot Images

What to confirm:

  • At least one image status is Ready.

  • Image is available in the zone where you will deploy the VM.

  • OS type matches your first workload.

Quick fix if missing:

  • Fast path: register/select a VM template and confirm Ready status.

  • Alternative: upload a boot image and use boot-image provisioning (longer path).

  • Reopen the table and confirm Ready before starting VM creation.

Full guides: VM Templates, Boot Images

Step 4 - Instance Storage

Go to: Storage -> Instance Storage

What to confirm:

  • At least one storage pool row exists.

  • Pool state is Up.

  • Free capacity is sufficient for planned VM size.

Quick fix if missing:

  • If no pool exists, click + Add Storage Pool and complete required fields.

  • If pool exists but is not Up, verify storage endpoint/network path and refresh state.

  • Continue only when at least one pool is Up.

Full guide: Instance Storage

Note

Manual networking configuration is not required for initial VM creation in standard Bootstrap deployments. Use Network docs when you need custom VLAN/CIDR/public segmentation.

If all four are checked, skip to Section 2 (Navigating the VM Dashboard).

If any are unchecked, continue with detailed validation below.

Detailed Prerequisites Validation

Use this section only if the 4 quick checks above pass but VM provisioning is still blocked.

Access Checks

Path: User Icon -> My Profile

Confirm:

  • Your role includes VM_VIEW and VM_MANAGE.

  • You can see the VM menu and Create VM button in VM dashboard.

If not met:

  • Permission issue: request VM Admin or higher role from your administrator.

Host and Capacity Checks

Path: Infrastructure -> Hosts and Storage -> Instance Storage

Confirm:

  • At least one host is Up (not Error/Disconnected).

  • At least one instance storage pool is Up.

  • Instance Storage has enough free capacity for planned VM size.

If not met:

Image Path Checks

Path: Storage -> VM Templates / Storage -> Boot Images and Storage -> Image Storage

Confirm:

  • At least one VM template or boot image is Ready.

  • Image Storage is healthy when using templates or boot images.

  • Image is available in the target deployment zone.

If not met:

Optional Security Check

Path: Compute Policies -> SSH Keypairs

  • Optional: at least one keypair is present for secure Linux access.

  • If missing, create one in SSH Keypairs.

Final Preflight Checklist

Run this final preflight before clicking Create VM and before final submission with Provision VM:

[ ] VM_MANAGE permission present
[ ] One host Up
[ ] One instance storage pool Up
[ ] One instance profile enabled
[ ] One VM template or boot image Ready

Common Missing-Prerequisite Symptoms

Symptom

Likely Missing Prerequisite

Create VM panel has empty dropdowns

Instance profile or VM template/boot image missing

No suitable host found error

Host offline, insufficient resources, or storage full

VM stuck in Creating for an extended period

Storage issue or platform initialization delay

VM has no IP address

Network not in Setup state or DHCP issue

Cannot select VM template

Image Storage unavailable or VM template not Ready

If You See Any Empty Dropdowns

Warning

If any dropdown in the Create VM panel is empty, that prerequisite is missing.

Troubleshooting:

  1. Empty Instance Profile -> Create profile: Instance Profiles

  2. Empty VM Template -> Register VM template: VM Templates

  3. Empty Boot Image -> Upload boot image: Boot Images

Manual network creation is optional for first VM in standard setup.

Note

If all prerequisites are met, continue to Section 2. Navigating the VM Dashboard to validate dashboard access and filters, then continue to Section 3. Creating Your First Virtual Machine.

Tip

Bookmark this checklist for future VM deployments or troubleshooting.

3. Creating Your First Virtual Machine

3.1. Launch the Create VM Panel

When to Use:

Start here when prerequisites are complete and you are ready to provision a VM.

Purpose:

Open the Create VM flow and confirm the Provision Virtual Machine panel loads with the required fields.

Steps:

  1. Open VM dashboard.

  2. Click Create VM at top-right.

  3. Confirm the Provision Virtual Machine panel opens.

Expected Outcome:

  • The Provision Virtual Machine panel opened from Create VM is visible and ready for input.

If this fails:

  1. Refresh the VM dashboard and confirm the Create VM button is visible and enabled in the current account/project scope.

  2. Recheck the VM creation prerequisites above, especially instance profile, host, Instance Storage, and image readiness.

  3. Check Observability Events/Alerts for create-form or permission errors before retrying.

Creation Form Views (UI Reference)

Use these two form views before filling fields:

VM creation form with VM Template selected

Create VM panel - VM Template path.

VM creation form with Boot Image selected

Create VM panel - Boot Image path.

3.2. Common Fields (VM Template and Boot Image)

Field

Required

Visible When

Description

VM Name

Yes

Always

VM identifier used in dashboard and operations.

Zone

Yes

Always

Deployment zone for compute/network/storage placement.

Instance Profile

Yes

Always

Compute sizing profile (CPU and memory).

Type

Yes

Always

Creation source selection: VM Template or Boot Image.

VM Template

Yes

Type = VM Template

Prebuilt VM image used for fast provisioning.

Boot Image

Yes

Type = Boot Image

Installer media used for manual OS installation.

Volume Profile

Yes

VM Template and Boot Image

Primary disk performance/size profile.

Network

No

VM Template and Boot Image

Guest network attachment for VM connectivity.

SSH Keypair

No

VM Template and Boot Image

Key-based access bootstrap for Linux VMs.

VM User Data

No

VM template workflow ()

First-boot automation script/cloud-init configuration.

Field Tips (New User)

  • VM Name: use descriptive names such as web-prod-01 or dev-api-01.

  • Zone: choose based on locality/compliance and available capacity.

  • Instance Profile: align with workload demand; over-sizing wastes capacity.

  • Volume Profile: choose performance tier based on I/O needs.

  • Network: optional for first VM in standard setup; keep default unless custom network behavior is required.

  • SSH Keypair: keep private key secure; key loss can block access.

  • VM User Data: use for repeatable first-boot setup; test scripts before production.

Submission Actions

  • Cancel: closes dialog with no resource allocation.

  • Provision VM: starts provisioning and returns to VM list for status tracking.

Expected Outcome:

  • VM appears in dashboard and transitions through Starting/Creating to Running (or Error if prerequisites fail).

3.3. Type Selection - How Form Changes

  • VM Template: preconfigured OS image path.

  • Boot Image: installer-based path for custom OS setup through Console.

3.4. VM Template vs Boot Image (Quick Comparison)

Attribute

VM Template

Boot Image

Boot readiness

Boots directly to installed OS.

Boots installer media; manual OS install required.

Best for

New users and standardized deployments.

Custom installs and advanced use.

Unique field

VM User Data

Boot image selection

3.4.1. Decision Gate (What to choose now)

  • If a VM template is available and in Ready state: use VM Template path.

  • If a VM template is unavailable but a boot image is ready: use Boot Image path and complete OS install in Console.

  • If neither is available: stop and complete media prerequisites first.

3.5. Step-by-Step (VM Template Path)

When to Use:

Use this path for provisioning with prebuilt operating system images.

Purpose:

Deploy a VM that can reach Running quickly with minimal manual setup.

Steps:

  1. Click Create VM.

  2. Enter VM Name.

  3. Select Zone.

  4. Select Instance Profile.

  5. Keep Type as VM Template.

  6. Select VM Template.

  7. Select Volume Profile.

  8. Optional: Select Network if required for your deployment.

  9. Optional: SSH Keypair.

  10. Optional: VM User Data.

  11. Click Provision VM.

  12. Confirm state becomes Running.

Expected Outcome:

  • VM reaches Running state.

  • Newly created VM appears automatically in the VM dashboard list.

Post-provision state handling:

  • Starting / Creating: monitor state transition.

  • Running: proceed to validation and access.

  • Error: check host/storage/platform state and active alerts.

  • If VM is not visible in the current view, clear filters or switch All States selection.

If this fails:

  1. Confirm the target snapshot still exists for the correct VM volume and that the VM is in the required state for restore.

  2. Validate the rollback target carefully and ensure no newer guest changes are still expected to be preserved.

  3. Check Observability Events/Alerts for the VM name, snapshot name, and restore timestamp, then retry after correction.

3.6. Step-by-Step (Boot Image Path)

When to Use:

Use this path when you must install the guest OS manually from installer media.

Purpose:

Provision a VM and complete operating system installation through Console.

Steps:

  1. Click Create VM.

  2. Enter VM Name.

  3. Select Zone.

  4. Select Instance Profile.

  5. Set Type to Boot Image.

  6. Select Boot Image.

  7. Select Volume Profile.

  8. Optional: Select Network if required for your deployment.

  9. Optional: SSH Keypair.

  10. Optional: VM User Data.

  11. Click Provision VM.

  12. Open Console and complete OS installation.

Expected Outcome:

  • VM is provisioned and boots the boot image installer; OS setup must be completed in Console.

  • Newly created VM appears automatically in the VM dashboard list.

Post-provision state handling:

  • Starting / Creating: monitor state transition.

  • Running: open Console to run installer.

  • Error: validate boot-image/media/zone selections (and custom network selection if used), then check active alerts.

  • If VM is not visible in the current view, clear filters or switch All States selection.

Warning

Boot-image VM is not ready for application use until OS installation is finished in Console.

If this fails:

  1. Confirm the VM is in the required power state for snapshot creation and refresh the Snapshot tab before retrying.

  2. Verify the selected volume is still attached, supports snapshots, and has enough backend capacity for the new restore point.

  3. Check Observability Events/Alerts for the VM name, target volume, and action timestamp, then retry after correction.

4. Managing VM Lifecycle (Start, Stop, Reboot, Delete)

VM actions are available from the VM detail page action toolbar and state-aware action dialogs.

When to Use:

Use this section whenever you need to change VM power state, decommission a VM, or resize compute profile.

Purpose:

Perform lifecycle actions safely with state checks and impact awareness.

Steps:

  1. Open Virtual Machines and click the target VM name.

  2. Confirm the current VM state in the header.

  3. Click the required action icon (Start, Stop, Reboot, Delete, or Scale Instance).

  4. Confirm the action dialog.

  5. Watch state transition until the VM returns to a stable state.

Note

Common actions are shown as icon buttons in the VM details header.

VM details action toolbar with lifecycle icons

VM header action toolbar.

What this screenshot shows:

  • Action icons for scaling, start, stop, reboot, delete, and recovery/state operations

  • Icon availability reflects current VM state

Icon-only action tip:

  • Hover each icon and confirm tooltip text before clicking (for example Start, Stop, Reboot, Delete, Scale Instance).

Scale Instance dialog with instance profiles

Scale Instance dialog.

What this screenshot shows:

  • Instance profile options with CPU and memory values

  • OK confirmation action to apply sizing change

Lifecycle Actions

Action

What It Does

When To Use

Start

Powers on stopped VM.

Resume service.

Stop

Graceful shutdown, keeps config/disks.

Maintenance or planned pause.

Reboot

Graceful reboot.

Apply config/package updates.

Delete / Destroy

Moves VM to destroyed state.

Decommission VM.

Scale Instance

Changes VM instance profile (CPU/memory profile).

Right-size VM after performance review.

Warning

Delete/Destroy is destructive. Confirm backup or snapshot coverage before proceeding.

Warning

Recoverability after delete depends on platform expunge/retention policy and can not be guaranteed.

Expected Outcome:

  • VM state or VM sizing changes according to selected action.

If this fails:

  1. Confirm the target snapshot still exists for the correct VM volume and that the VM is in the required state for restore.

  2. Validate the rollback target carefully and ensure no newer guest changes are still expected to be preserved.

  3. Check Observability Events/Alerts for the VM name, snapshot name, and restore timestamp, then retry after correction.

5. Running and Non-Running States

When to Use:

Use this view when you need to quickly split healthy running workloads from VMs that need intervention.

Purpose:

Speed up triage by grouping VMs by runtime state before troubleshooting.

Steps:

  1. Open Virtual Machines.

  2. Use the state card/filter for Active or All States.

  3. Focus on Stopped, Error, or Destroyed rows first.

Expected Outcome:

  • You can separate running VMs from stopped/error/destroyed VMs and move to the correct follow-up action.

If this fails:

  1. Clear filters, refresh the VM list, and confirm the state cards reflect the latest runtime state.

  2. Verify you are in the expected account/project scope and that the target VM is not transitioning between states.

  3. Check Observability Events/Alerts for recent VM lifecycle changes or list-load errors before retrying.

5.1. Running

Running VMs consuming host resources.

5.2. Non-Running

Stopped, Error, or Destroyed VMs.

Tip

Use the All States dropdown to switch directly to the target runtime state.

Expected Outcome:

  • You can quickly locate VM by runtime state and continue actions.

6. VM Detail View - Exploring Your VM

Open details by clicking VM name.

When to Use:

Open VM details after provisioning and before any high-impact action.

Purpose:

Use one page to validate identity, state, tabs, and operational controls for the selected VM.

Steps:

  1. Open Control Center -> Compute -> Virtual Machines.

  2. Click the VM Name.

  3. Confirm header state and available action icons.

  4. Navigate tabs (Details, Console, Snapshot, Volumes, Metrics, Karios Shield) based on your task.

Expected Outcome:

  • The selected VM detail page opens with current state, header actions, and tab navigation visible.

If this fails:

  1. Return to the VM list and reopen the VM by name to confirm you selected the intended workload.

  2. Verify the VM still exists, is visible in the current scope, and is not stuck in a transition state.

  3. Check Observability Events/Alerts for VM-detail or permission-related errors before retrying.

6.1. Page Actions (Header)

The VM detail header shows actions based on current VM state.

Action

Visible When

Description

Start

VM is Stopped

Powers on VM and returns it to service.

Stop

VM is Running

Performs graceful shutdown.

Reboot

VM is Running

Reboots the guest OS.

Delete

Available unless blocked by transition/policy

Permanently removes the VM and associated resources based on retention/expunge policy.

Note

Action buttons can be disabled while VM is transitioning (for example Starting, Stopping, Rebooting, Destroying).

6.2. Tab Overview (What Each Tab Is For)

Tab

Purpose

Details

Validate identity, hardware, OS/hypervisor, network, protection, and storage metadata.

Console

Browser-based VNC access for first boot, installer flow, and emergency troubleshooting.

Snapshot

Create, list, and restore rollback points.

Volumes

Review attached disks and perform attach/detach actions.

Metrics

Monitor CPU, memory, disk, and network trends for performance analysis.

Karios Shield

Review VM security posture and protection settings.

6.3. New User Navigation Flow

  1. Open target VM details from dashboard.

  2. Check current VM state in header before any action.

  3. Validate identity/network/storage quickly in Details tab.

  4. Use Console only if direct guest access is needed.

  5. Use Snapshot before high-risk updates.

  6. Use Metrics together with VM state and alerts when troubleshooting.

  7. Use Karios Shield for security posture checks.

6.4. Deletion Confirmation (From Detail Page)

Before confirming delete, verify:

  • Correct VM name/ID

  • Current runtime state

  • Attached storage volumes listed in the confirmation dialog

  • Snapshot/backup coverage and data retention expectations

Warning

Delete is destructive and cannot be undone. If recovery is required, confirm rollback artifacts before proceeding.

6.5. Quick Troubleshooting (Detail Page)

Symptom

Likely Cause

First Check

Action button disabled

VM in transitional state

Confirm stable state and refresh

Console unavailable/blank

Console proxy/session issue

Check console services/system VM health

Snapshot action not available

Policy/state restriction

Confirm VM state and snapshot capability

Metrics empty

VM stopped or just started

Recheck after VM is running and telemetry refresh completes

Action fails without a clear error message

Upstream infra issue

Cross-check host/storage/network state and active alerts

Expected Outcome:

  • You can access all operational tabs for the selected VM.

6.6. Open VM Detail Help Panel

  1. Open a VM details page.

  2. Click the help icon (?) in the top-right corner.

  3. Review page-specific help for actions and tabs.

  4. Close the help panel.

VM detail help panel

In-app help panel for the VM Detail page.

Expected Outcome:

  • You can review action guidance and field descriptions inline while operating the VM.

7. Details Tab - VM Configuration at a Glance

When to Use:

Use this tab first after VM creation and before lifecycle or recovery actions.

Purpose:

Validate core VM identity, hardware sizing, network values, and storage attachment.

VM details information panel

Details tab - configuration and identity.

What this screenshot shows:

  • VM header with state and tab navigation

  • Basic Information, Operating System & Boot, and Hardware Specifications cards

  • Network interface and security/protection settings in the right panel

Steps:

  1. Open VM details.

  2. Stay on Details tab.

  3. Compare identity, hardware, network, and storage values against expected configuration.

Expected Outcome:

  • VM identity, hardware, network, and ownership values are visible for comparison against the expected configuration.

If this fails:

  1. Refresh the detail page and reopen the Details tab to rule out a stale UI session.

  2. Confirm the VM is still reachable from the management plane and not in a state transition.

  3. Check Observability Events/Alerts for metadata-load or API errors tied to the VM before retrying.

7.1. Basic Information

Identity and metadata for the selected VM.

Field

Description

Name

Display name of the VM.

ID

Unique system identifier (UUID) used in API calls and logs.

Instance Name

Internal hypervisor instance name (for example i-2-42-VM).

Host

Physical host currently running this VM.

7.2. Hardware Specifications

Compute resources allocated to this VM.

Field

Description

CPU Count

Number of virtual CPU cores allocated.

CPU Speed

Clock speed cap per vCPU in MHz.

Memory

RAM allocated to the VM in MB.

Dynamically Scalable

Whether CPU/memory can be scaled without reboot (guest support dependent).

GPU Count

Number of GPU devices attached/passed through to the VM.

Tip

Use the Scale Instance action in the header toolbar to adjust CPU and memory without recreating the VM.

7.3. Operating System and Boot

OS and hypervisor configuration for the VM.

Field

Description

OS Type

Guest operating system display name.

Hypervisor

Virtualization technology running this VM.

Architecture

CPU instruction set architecture (for example x86_64 or aarch64).

7.4. Network Interfaces

One entry per virtual NIC attached to the VM.

Field

Description

MAC Address

Unique hardware MAC address of the VM interface.

IP Address

Private IP assigned to this NIC in the guest network.

Network

Guest network this NIC is connected to.

Gateway

Default gateway for this NIC subnet.

Note

VMs can have multiple NICs connected to different guest networks.

7.5. Network Configuration

VM-level network traffic and public connectivity details.

Field

Description

IP Address

Primary private IP of the VM.

Public IP

Public/floating IP associated with the VM, if assigned.

Network KB Read

Total network data read since VM start.

Network KB Write

Total network data written since VM start.

Received Bytes

Raw bytes received by the VM.

Sent Bytes

Raw bytes sent by the VM.

7.6. Security and Protection

VM protection and availability settings.

Field

Description

Delete Protection

Prevents accidental VM deletion when enabled.

HA Enabled

Starts the VM on another host if the current host fails.

Password Enabled

Indicates whether guest password reset is supported via platform controls.

Security Groups

Number of applied firewall/security-group rules.

Tip

Enable Delete Protection on production VMs and use HA Enabled for critical workloads.

7.7. Storage Information

Disk configuration and cumulative I/O details for this VM.

Field

Description

Root Device Type

Storage backend type for the root disk (for example NetworkFilesystem or RBD).

Volume Profile

Volume profile applied to attached data disks, if any.

Disk I/O Read

Total disk read operations since VM start.

Disk I/O Write

Total disk write operations since VM start.

Disk KB Read

Total data read from disk since VM start.

Disk KB Write

Total data written to disk since VM start.

7.8. Resource Allocation

Cloud assignment context for tenancy and placement.

Field

Description

Instance Profile

Compute profile controlling CPU and memory limits.

VM Template

Base image used to create this VM.

Domain

Domain this VM belongs to for multi-tenancy.

Account

Account owner of this VM.

Zone

Availability zone where this VM is deployed.

Host

Physical hypervisor host currently running this VM.

Expected Outcome:

  • Configuration and runtime metadata are validated before operations.

8. Console Tab - Direct CLI Access

When to Use:

Use Console when SSH/RDP is unavailable, when using boot-image install flow, or for emergency guest access.

Purpose:

Access the guest directly from the browser to complete install and validate boot behavior.

VM console access

Browser-based VM console.

What this screenshot shows:

  • Console tab open while VM is Running

  • Live guest output visible in the embedded console view

Console components:

  • VNC Console for browser-based guest access

  • Console session list/details when active console connections exist

Steps:

  1. Open VM details.

  2. Ensure VM is in Running state (start it if needed).

  3. Click Console tab.

  4. Confirm login prompt appears.

  5. Sign in with guest OS credentials.

Use Console for:

  • First boot verification

  • Boot image installer flow

  • Emergency troubleshooting

Why this matters:

  • Gives direct access when network services (SSH/RDP) are not ready

  • Required for boot-image-based OS installation flow

  • Helps diagnose boot and login issues quickly from one place

Expected Outcome:

  • Interactive VM terminal session is available in browser.

Warning

If console is blank, check platform console proxy/system VM health.

Console Troubleshooting

Symptom

Likely Cause

First Check

Blank console page

Console proxy/session service issue

Verify console proxy/system VM health

Keyboard input not accepted

Browser session focus/capture issue

Reopen console in a fresh session

Login prompt never appears

Guest not fully booted

Check VM state and recent lifecycle actions

If this fails:

  1. Confirm the VM is Running and that console proxy or required system VM services are healthy.

  2. Reopen the console in a fresh browser session and allow any blocked pop-up or session prompts.

  3. Check Observability Events/Alerts for console or system-VM errors before retrying.

9. Snapshot Tab - Backup and Recovery

When to Use:

Create snapshots before high-risk guest changes and restore snapshots when rollback is required.

Purpose:

Provide a rollback option for VM state and data.

VM snapshot tab with snapshot name input and create action

Snapshot creation workflow.

What this screenshot shows:

  • Snapshot tab showing per-volume snapshot controls

  • Snapshot name input and Create action for each attached volume

  • Running-state warning banner indicating snapshot creation/rollback requires VM stop first

9.1. Why Snapshots

  • rollback safety before changes

  • recovery from failed updates

  • change validation and testing

Why Snapshot is important:

  • Reduces risk before patching, upgrades, or configuration changes

  • Provides a fast rollback point if the VM becomes unstable

  • Supports safer experimentation in test and staging workflows

Snapshot vs Backup (important):

  • Snapshot: platform rollback on the storage path.

  • Backup: independent copy for recovery from host/storage/site failure.

  • Do not treat snapshots as full backup strategy for production workloads.

9.2. Create Snapshot

When to Use:

Use this before high-risk guest changes so you have a rollback point for the selected VM volume.

Purpose:

Create a point-in-time restore marker you can use if patching, upgrades, or config changes fail.

Steps:

  1. Stop VM.

  2. Open Snapshot tab.

  3. Select volume.

  4. Enter snapshot name.

  5. Click Create.

  6. Confirm latest snapshot entry.

Snapshot success indicators:

  • Snapshot name appears in list for selected volume

  • Snapshot entry updates

  • Restore point is available for rollback workflow

Expected Outcome:

  • New snapshot is listed for selected volume.

Warning

Snapshot creation can be blocked while VM is running.

If this fails:

  1. Confirm the VM is in the required power state for snapshot creation and refresh the Snapshot tab before retrying.

  2. Verify the selected volume is still attached, supports snapshots, and has enough backend capacity for the new restore point.

  3. Check Observability Events/Alerts for the VM name, target volume, and action timestamp, then retry after correction.

9.3. Restore Snapshot

When to Use:

Use this when you must roll back a VM volume to a known-good snapshot state.

Purpose:

Restore VM volume data/state to the selected snapshot to recover from failed changes.

Steps:

  1. Open Snapshot tab.

  2. Locate target snapshot by name.

  3. Select restore action for that snapshot.

  4. Confirm restore prompt and impact.

  5. Confirm restore completes and VM state stabilizes.

Expected Outcome:

  • VM state/data returns to selected snapshot point.

Warning

Restoring a snapshot can overwrite newer guest changes. Validate rollback intent before confirming.

If this fails:

  1. Confirm the target snapshot still exists for the correct VM volume and that the VM is in the required state for restore.

  2. Validate the rollback target carefully and ensure no newer guest changes are still expected to be preserved.

  3. Check Observability Events/Alerts for the VM name, snapshot name, and restore timestamp, then retry after correction.

10. Volumes Tab - Storage Management

When to Use:

Use this tab when reviewing VM disk layout or attaching/detaching data disks.

Purpose:

Control VM storage safely without changing VM identity or compute profile.

VM volumes tab

Volumes tab showing ROOT and DATADISK volumes.

Volume types:

  • ROOT: OS disk, always attached

  • DATADISK: additional storage, attach/detach by policy

Why Volumes are important:

  • Separates OS disk from application/data disks for better management

  • Enables controlled storage expansion without rebuilding the VM

  • Improves recovery planning by isolating critical data volumes

Common actions:

  • Add volume (+)

  • View volume details

  • Detach DATADISK (if allowed)

Create and attach disk drawer from VM volumes tab

Create and Attach Disk drawer opened from the Volumes tab.

Steps:

  1. Attach: add DATADISK for application/data growth.

  2. Detach: remove DATADISK only after guest-side unmount and data validation.

  3. View Details: confirm size, type, and state before changes.

Expected Outcome:

  • Volume inventory and attach/detach state are visible and manageable per policy.

Warning

Do not detach ROOT volume. For DATADISK, always unmount inside guest OS first.

If this fails:

  1. Confirm you selected the correct volume and that guest-side unmount or application shutdown is complete before detach.

  2. Verify the volume state, storage pool health, and current VM state allow the requested disk action.

  3. Check Observability Events/Alerts for attach/detach or storage-side failures before retrying.

11. Metrics Tab - Performance Monitoring

When to Use:

Use this tab when a VM feels slow, resource limits are suspected, or you need trend validation after changes.

Purpose:

Correlate VM performance signals to workload behavior and right-sizing decisions.

Monitor:

  • CPU

  • memory

  • disk I/O

  • throughput

  • network behavior

VM metrics tab showing CPU and memory charts

Metrics tab with CPU and memory trend charts.

VM metrics tab showing disk IO throughput and network charts

Metrics tab with disk I/O operations, throughput, and network trend charts.

Tip

Use metrics timeline to correlate spikes with deployments or incidents.

Metric interpretation quick guide:

Signal

Typical Pattern

Action

CPU sustained high

High utilization over long window

Check workload saturation and resize/optimize if needed

Memory near limit

Frequent near-max memory usage

Review memory pressure and right-size VM

Disk I/O spikes

Sharp read/write bursts

Correlate with backup/jobs and storage tier capability

Network throughput anomalies

Unexpected in/out changes

Validate workload behavior and network path health

Expected Outcome:

  • Resource trends are visible for CPU, memory, disk, and network.

12. Karios Shield Tab - Security Scanning

When to Use:

Use this tab after VM provisioning and after remediation cycles to validate security posture.

Purpose:

Run VM-level scans, track findings history, and export evidence for audits.

VM Karios Shield metrics view

Karios Shield metrics with compliance and findings.

What this screenshot shows:

  • Metrics sub-tab with compliance score, compliance status, total findings, and scan status

  • Profile, scan ID, severity breakdown, report export buttons, and findings table

VM Karios Shield history view

Karios Shield scan history.

What this screenshot shows:

  • History sub-tab with scan rows and View Findings action

  • Compliance fields used for trend tracking

12.1. Karios Shield Screen Components (What You See)

Top actions and tabs:

  • Start Security Scan button (top-right): starts a new security scan for this VM

  • Metrics tab: current/latest scan summary and vulnerability list

  • History tab: previous scans with status/compliance and detail actions

Metrics summary cards:

  • Compliance Score (percentage)

  • Compliance Status (for example: GOOD)

  • Total Findings (with severity split)

  • Scan Status (for example: COMPLETED)

Metrics detail panels:

  • Profile, Scan ID, Target

  • Severity counts: Cat I, Cat II, Cat III

  • Reports buttons: PDF and HTML

  • Vulnerabilities table with columns like ID, Title, Severity, Status, Category, Remediation

History table:

  • Scan ID

  • Profile

  • Status

  • Compliance

  • Actions (View Findings)

12.2. Run Security Scan (Step-by-Step)

  1. Open VM details.

  2. Open Karios Shield tab.

  3. Click Start Security Scan.

  4. Confirm status updates to COMPLETED.

  5. Stay on Metrics to review results.

Expected Outcome:

  • New scan result appears in Metrics and is added to History.

12.2.1. Enabling VM Scanning by OS Type

Use this before running scans from the VM Karios Shield tab.

Ubuntu Instances

To enable automated security scanning on an Ubuntu VM, provision the VM with Ubuntu VM Scan in VM User Data.

Important

Scanning works only when Ubuntu VM Scan user data is selected during VM provisioning. If this option is not selected, the VM is created normally, but VM scanning capabilities are not enabled.

When to Use:

Use this when provisioning an Ubuntu VM that must support Karios Shield scanning.

Purpose:

Enable scan communication automatically by deploying Ubuntu with the required Ubuntu VM Scan user data.

Steps:

  1. Navigate to VM provisioning.

  2. Select Ubuntu as the operating system.

  3. In the VM User Data dropdown, select Ubuntu VM Scan.

  4. Complete remaining provisioning steps and deploy the VM.

  5. Run scan using Ubuntu_Baseline profile, or upload and use your own custom profile.

What Ubuntu VM Scan user data does:

  • Installs the QEMU Guest Agent.

  • Enables and starts the agent service during boot.

  • Allows the scanning system to communicate with the VM through the hypervisor.

Expected Outcome:

  • Ubuntu VM is provisioned with Ubuntu VM Scan user data and is ready for Karios Shield scan execution.

If this fails:

  1. Confirm Ubuntu VM Scan was selected during provisioning and that the VM was deployed from the intended template or boot image.

  2. Log in to the guest and verify qemu-guest-agent is installed, enabled, and running.

  3. Check Observability Events/Alerts for guest-agent or scan-communication errors before retrying.

Windows Instances

To enable automated security scanning on a Windows VM, install QEMU Guest Agent manually inside the VM. Unlike Ubuntu VMs, Windows VMs do not automatically install the guest agent during provisioning.

Important

Scanning works only after QEMU Guest Agent is installed and running inside the Windows VM. If the guest agent is not installed, scans fail because the scanning system cannot communicate with the VM.

Steps to enable scanning on a Windows VM:

  1. Provision the Windows VM normally.

  2. Log in to the VM using RDP or VM console.

  3. Install QEMU Guest Agent inside the VM.

  4. Confirm QEMU Guest Agent service is running.

  5. Run scan using Windows_Baseline profile, or upload and use your own custom profile.

Verifying installation:

  1. Open Services in Windows.

  2. Locate QEMU Guest Agent.

  3. Confirm service status is Running.

What QEMU Guest Agent enables:

  • Communication between the scanning system and VM through the hypervisor.

  • Safe execution of scanning operations.

  • Collection of required system information for scan output.

Note

Guest agent installation is required only once per VM. After installation, scanning continues to work after VM restart.

12.3. Actions in Metrics Tab

Use these actions directly from Karios Shield:

  • Click Start Security Scan and choose a validated profile.

  • Click PDF or HTML in Metrics to export reports.

  • Click Recommend in vulnerability rows to view remediation guidance.

Security Scan Profile dialog with profile selection and Start Security Scan action

Security scan profile selection dialog.

What this screenshot shows:

  • Existing Profiles and custom profile upload options

  • Profile list with selected baseline and Start Security Scan confirmation

Karios Shield findings dialog showing vulnerability list with severities

Scan findings modal.

What this screenshot shows:

  • Findings table for a selected scan ID

  • Vulnerability ID, title, and severity columns for remediation triage

What each action does:

  • Start Security Scan: initiates a new scan with selected profile.

  • PDF / HTML: gives evidence/report artifact for sharing and audits.

  • View Findings / Recommend: opens finding context for remediation.

Expected Outcome:

  • You can export reports and start remediation from finding-level guidance.

12.4. Actions in History Tab

  1. Open History sub-tab.

  2. Locate the scan row by scan ID.

  3. Click View Findings in the row.

What this action does:

  • Opens the findings context for that historical scan so you can compare security posture across scans.

Expected Outcome:

  • You can review older scan outputs and track compliance trend consistency.

12.5. New User Workflow for Karios Shield

  1. Run first scan with Start Security Scan.

  2. Review Compliance Score and Total Findings.

  3. Sort focus by severity (Cat I -> Cat II -> Cat III).

  4. Use Recommend on high-severity findings first.

  5. Apply remediation changes inside the VM.

  6. Re-run scan and compare new results in History using View Findings.

  7. Download PDF report for records.

Expected Outcome:

  • Security posture is measurable, remediations are traceable, and improvements are verifiable across scans.

Warning

Prioritize high-severity findings first and confirm remediation with a follow-up scan.

13. Auto Scaling

Auto Scaling automatically adjusts the number of virtual machines in a group based on real-time performance metrics, helping workloads keep the right amount of compute capacity as demand changes.

When to Use:

Use Auto Scaling when a workload should add or remove VM capacity automatically based on metric thresholds instead of manual Scale Instance actions.

Purpose:

Maintain a bounded group of VMs behind a load balancer rule, with scale-up and scale-down policies controlling when VM members are added or removed.

13.1. Auto Scaling Dashboard

Path: Control Center -> Compute -> Auto Scaling

The Auto Scaling dashboard lists configured VM scaling groups with their current state and member counts. From here you can create new groups, enable or disable existing groups, and open group details.

Auto Scaling dashboard with summary cards, empty state, and Create Auto Scaling Group action

Auto Scaling dashboard.

What this screenshot shows:

  • Total Groups, Enabled, and Disabled summary cards

  • Empty-state message when no auto scaling groups exist

  • Create Auto Scaling Group action

  • Auto Scaling selected in the Compute navigation

Summary cards at the top provide a quick snapshot:

Card

Meaning

Total Groups

Number of auto scaling groups configured in the environment.

Enabled

Groups that are actively monitoring metrics and scaling VMs.

Disabled

Groups that exist but are paused and will not trigger scaling actions.

13.2. Auto Scaling Group Table

The main table lists each group and its current operating limits.

Column

Description

Name

Unique identifier for the auto scaling group. Click a row to open the detail page for that group.

State

Enabled groups are active and scale based on policy thresholds. Disabled groups are paused and do not trigger scaling actions.

Load Balancer Rule

Load balancer rule associated with the group. Traffic is distributed across VMs managed by this group.

Min / Max Members

Lower and upper bounds for how many VMs the group can maintain.

Interval

How frequently, in seconds, the auto scaling engine checks metric counters against policy thresholds.

Available VMs

Current number of VMs actively running under this group.

13.3. Creating an Auto Scaling Group

Click Create Auto Scaling Group to open the three-step creation wizard.

Create Auto Scaling Group wizard open on VM Profile step

Create Auto Scaling Group wizard.

What this screenshot shows:

  • Three-step wizard: VM Profile, Scale Policies, and VM Group

  • Required Zone, Template, Compute Offering, and Expunge Instance Grace Period (s) fields

  • Next action to continue to scale policy configuration

Step 1 - VM Profile

Define what each scaled VM looks like:

  • Zone

  • Template

  • Compute offering

  • Expunge instance grace period before removing a scaled-down VM

  • Optional networking, storage, and key pair parameters

Step 2 - Scale Policies

Configure two policies:

  • Scale Up defines the metric counter, threshold, and timing that triggers adding a new VM.

  • Scale Down defines the conditions that remove a VM from the group.

Each policy references:

  • Counter, such as CPU or memory

  • Relational operator, such as GT or LT

  • Threshold value

  • Duration, in seconds, that the condition must hold

  • Quiet time, or cooldown period, between consecutive scaling actions

You can select an existing counter or create a new counter inline.

Step 3 - VM Group

Set the group name, link the group to a load balancer rule, and configure minimum members, maximum members, and polling interval.

13.4. Enabling and Disabling Groups

Use the toggle action on any row to enable or disable a group without deleting it. Disabling a group stops scaling evaluations while preserving the group configuration and existing VMs.

13.5. Deleting a Group

Deleting a group removes the scaling configuration. Existing VMs managed by the group are not automatically destroyed, but they are no longer managed by Auto Scaling.

14. Troubleshooting and Quick Reference

14.1. Common Issues and Fixes

Symptom

Likely Cause

Immediate Action

Empty Create VM dropdowns

Missing prerequisite objects

Validate profile/media/storage readiness

VM stuck Starting

Host/storage/media readiness issue

Check host/storage availability and active alerts

VM in Error

Provisioning/runtime failure

Review action error text and correlated alerts

Console blank

Console proxy/session issue

Verify console/system VM services

Snapshot action unavailable

Policy/state limitation

Confirm VM state and snapshot support

No VM templates/boot images selectable

Images not registered/ready

Register media and confirm Ready state

Karios Shield no output

Guest-side compatibility/profile issue

Validate agent/profile compatibility and scan status

Metrics empty

VM stopped or telemetry delay

Ensure VM is running and recheck after telemetry refresh

14.3. First VM Checklist (New User)

  • Host ready in Forge

  • Zone enabled

  • Instance profile exists

  • Volume profile exists

  • Network available

  • Instance Storage up and Image Storage reachable

  • VM template or boot image in ready state

14.4. Quick VM Creation Flow

VM Template path:

Compute -> Virtual Machines -> Create VM -> Name -> Zone -> Instance Profile -> VM Template -> Volume Profile -> Network -> Provision VM

Boot Image path:

Compute -> Virtual Machines -> Create VM -> Name -> Zone -> Instance Profile -> Boot Image -> Volume Profile -> Network -> Provision VM -> Console install

14.5. Key Reminders

  • 0 VMs initially is normal

  • VM Template path uses preconfigured OS images

  • Boot Image path requires manual OS installation via Console

  • Take snapshots before high-risk changes

  • Check state filters when VM is not visible

  • Use Metrics and active alerts together for troubleshooting

14.6. Escalation Data to Capture

If issue persists, collect the following before escalation:

  • VM name and ID

  • Current VM state and last known good state

  • Exact failing action and related alert details

  • Relevant error text, VM state, and active alert details

  • Host/zone/account context

  • Whether issue reproduces after retry/re-login

  • Screenshot of affected UI state

14.7. Provisioning Failure Recovery

If VM provisioning fails mid-process:

  1. Do not repeatedly re-run the same failed request without checking VM state and active alerts.

  2. Capture the exact error text shown in action dialogs/notifications.

  3. Correlate with alert timestamps in Observability -> Alerts.

  4. Validate dependency objects:

  • instance profile exists and has capacity

  • target network is operational

  • VM template/boot image is Ready

  • Instance Storage pool is Up

Recovery decision:

  • Capacity issue: pick smaller instance profile or add host resources.

  • Network/media issue: fix dependency, then provision new VM.

  • Unknown recurring failure: escalate with alert evidence + VM ID + related alert details.

Expected Outcome:

  • New users can recover from common issues and complete VM operations end-to-end.

→ Next: Nodes

Expected Outcome

After this section, you should be able to:

  • Create and manage VMs end-to-end

  • Create and manage Auto Scaling groups for VM capacity changes

  • Use detail tabs (Console, Snapshot, Volumes, Metrics, Karios Shield)

  • Troubleshoot common VM lifecycle issues

What To Do Next

  1. Provision one test VM using a VM template and confirm it reaches Running.

  2. Validate console access and baseline network connectivity from that VM.

  3. Continue to Storage to manage volumes and capacity safely.