Documentation
Dashboard Documentation
Install and use the automatically generated dashboard, understand how it adapts to your environment, and choose the right customization approach.
Overview
The Proxmox Extended Sensors dashboard provides an automatically generated Home Assistant interface for the resources monitored by the integration.
Its purpose is not simply to display entities.
The dashboard organizes the available Proxmox information into an operational view that remains synchronized with the integration and adapts as the monitored environment changes.
PVE nodes, PBS servers, clusters, storage resources, virtual machines, containers and supported features are presented only when they are relevant to the current configuration.
The result is a dashboard that can normally be installed and used without manually creating or maintaining a large collection of Home Assistant cards.
Designed to prioritize useful information
Proxmox Extended Sensors can expose a large amount of information, but the automatic dashboard is intentionally not designed to display every available entity.
The sensors, cards and layout included in the dashboard have been selected and organized according to their operational relevance.
The goal is to provide enough information to understand the state of the Proxmox environment without overwhelming the interface with every metric exposed by the integration.
More important indicators are given greater visual relevance, while secondary information is grouped, reduced or omitted from the main view.
This creates a deliberate information hierarchy:
- important health and status information should be easy to identify;
- related values should be presented together so they can be interpreted in context;
- secondary metrics should not compete visually with conditions that may require attention;
- the number of visible sensors should remain manageable;
- information that is useful for detailed investigation does not necessarily need to occupy permanent dashboard space.
The automatic dashboard should therefore be treated as a curated operational view of Proxmox Extended Sensors, not as a complete entity catalogue.
If you need information that is not displayed in the automatic dashboard, the underlying entities remain available in Home Assistant and can be added to your own dashboards when required.
For the complete list of entities exposed by the integration, see Reference.
One dashboard, different environments
The dashboard does not assume that every Proxmox Extended Sensors installation is identical.
A small installation monitoring a single PVE server should not need the same interface as a cluster containing multiple nodes, virtual machines, containers, storage resources and a Proxmox Backup Server.
The generated dashboard uses the resources available in the current Home Assistant installation to determine what should be shown.
This allows the same dashboard system to remain useful across installations of very different sizes.
Automatic or manually controlled
There are two main ways to use the dashboard:
- Automatic mode — Proxmox Extended Sensors maintains the generated dashboard and adapts it to changes in the monitored environment.
- Take Control — Home Assistant converts the generated dashboard into a manually editable configuration controlled by the user.
Both approaches are valid, but they have an important difference.
Automatic mode prioritizes dynamic maintenance.
Take Control prioritizes manual customization.
The following sections explain how to install the dashboard, how its automatic behaviour works and what changes when you take control.
↑ Back to topInstallation
The Proxmox Extended Sensors dashboard is optional and is installed separately from the integration.
The integration provides the Home Assistant devices and entities.
The dashboard provides the automatic visual interface that organizes those entities.
The integration can be used without the dashboard.
Requirements
Before installing the dashboard, make sure you have:
- Proxmox Extended Sensors
- Card Mod
Card Mod is available through HACS.
See the official Card Mod project.
Install Card Mod
If Card Mod is not already installed:
- Open HACS.
- Search for Card Mod.
- Install it.
- Restart or reload Home Assistant if required by the Card Mod installation.
Add the dashboard resource
Add the following Lovelace resource as a JavaScript Module:
/proxmox_sensors/proxmox-dashboard.js
The resource must be registered as a JavaScript module.
proxmox-dashboard.js as a JavaScript module in Home Assistant.Reload the browser
After adding the resource, reload the Home Assistant frontend.
If the dashboard option does not appear immediately, perform a hard browser refresh to make sure the latest JavaScript resource is being loaded.
Create the dashboard
In Home Assistant, open:
Settings → Dashboards → Add dashboard → Community
Select:
Proxmox Extended Sensors
Home Assistant will create the dashboard using the Proxmox Extended Sensors community strategy.
The generated dashboard uses the entities currently exposed by the integration and builds the appropriate PVE, PBS and Cluster content automatically.
Keep the dashboard automatic
After creating it, no manual card configuration is required.
As long as you do not use Take Control, the dashboard remains managed automatically and can continue adapting to changes in the integration.
The following section explains that automatic behaviour in detail.
↑ Back to topAutomatic Dashboard
The Proxmox Extended Sensors dashboard is designed to remain synchronized with the integration automatically.
It does not create a fixed collection of cards that the user must maintain manually.
Instead, the dashboard builds its content from the Proxmox resources and entities currently available in Home Assistant.
Adapts to your environment
The dashboard only displays information that is relevant to the current installation.
Sections and elements can appear or disappear automatically as the Proxmox environment and the integration configuration change.
For example, replication information is only displayed when replicated VMs or containers are available.
If no monitored workload uses replication, the corresponding dashboard content is not shown.
The same principle allows the dashboard to avoid presenting empty or irrelevant sections for functionality that is not available in the current environment.
This keeps the interface focused on the features that actually exist.
Organized automatically
The dashboard also organizes the available information into views and sections corresponding to the monitored Proxmox resources.
The layout is generated from the resources currently available rather than requiring the user to build and maintain those views manually.
The dashboard intentionally prioritizes useful operational information instead of placing every available sensor on screen.
Add or remove monitored guests
When a VM or LXC container is added to Proxmox Extended Sensors monitoring, the dashboard can include it automatically.
If monitoring for that guest is later removed, its dashboard content disappears automatically as well.
There is no need to manually remove obsolete cards or repair references to entities that are no longer part of the monitored configuration.
The goal is to keep the dashboard clean and avoid leaving broken or irrelevant elements behind.
Guest migration
The automatic dashboard also follows monitored VMs and containers when they migrate between nodes within the configured Proxmox cluster.
If a monitored guest moves from one node to another, the dashboard removes it from the source node and displays it under its current node automatically.
No dashboard editing is required after the migration.
This works together with the V5 guest identity model: the workload remains the same monitored guest while its current Proxmox node changes.
The dashboard therefore reflects the current topology without requiring the user to reorganize cards after normal cluster operations.
Conditional information
Some dashboard information only makes sense when the corresponding feature or resource exists.
The automatic dashboard uses this context to avoid displaying empty sections.
Replication is one example.
If replication is present, the relevant information is displayed.
If it is not present, the dashboard does not reserve unnecessary space for it.
Replication available
Replication unavailable
The screenshots represent two different valid environment states, not a failure scenario.
The same dynamic principle helps remove other obsolete or unavailable information rather than leaving empty or broken dashboard elements.
Automatic maintenance
As long as the dashboard remains in automatic mode, its structure can continue adapting to changes such as:
- adding monitored VMs or containers;
- removing monitored VMs or containers;
- migrating guests between cluster nodes;
- enabling or removing supported functionality;
- resources becoming available or no longer being relevant;
- changes introduced by future dashboard updates.
This means the dashboard normally requires very little manual maintenance.
The automatic behaviour is one of its main features.
Automatic does not mean everything is displayed
The dashboard automatically reacts to the available integration data, but that does not mean every available entity is placed on screen.
The information hierarchy described in Overview still applies.
The generator determines which supported information belongs in the operational dashboard and how it should be organized.
Other entities remain available through Home Assistant and the Reference documentation.
↑ Back to topTake Control
Home Assistant allows you to Take Control of the generated dashboard.
This converts the automatically maintained dashboard into a configuration that you can edit directly using the normal Home Assistant dashboard tools.
For users who want to reorganize the interface manually, this is the easiest customization method.
What you can change
After taking control, you can use the normal Home Assistant dashboard editor to:
- move cards and sections;
- remove information you do not want;
- add additional Proxmox Extended Sensors entities;
- add your own cards;
- add entities from other Home Assistant integrations;
- reorganize the layout;
- create your own visual hierarchy;
- adapt the dashboard to the particular way you use Home Assistant.
No JavaScript knowledge is required for this type of customization.
What changes after Take Control
Taking control also changes who is responsible for maintaining the dashboard.
Once you take control, the dashboard is no longer dynamically managed by the automatic generator.
The automatic behaviour described in Automatic Dashboard is therefore lost.
For example:
- newly monitored guests will no longer necessarily be added automatically;
- guests removed from monitoring will no longer automatically disappear from your manually controlled configuration;
- migrated VMs and containers will no longer automatically move between node sections;
- conditional sections will no longer be managed dynamically by the generator;
- future structural improvements to the generated dashboard will not automatically modify your manually controlled version.
This is expected behaviour.
After Take Control, the dashboard configuration belongs to the user and Home Assistant preserves the manual configuration instead of allowing the automatic strategy to rebuild it.
Automatic mode or Take Control?
Choose according to what matters most in your environment.
Stay in automatic mode if you want:
- minimal maintenance;
- automatic guest discovery and removal;
- automatic response to guest migration;
- conditional sections based on available features;
- the dashboard to remain aligned with supported generator changes.
Take Control if you want:
- complete visual editing through Home Assistant;
- your own card arrangement;
- additional entities or integrations in the same dashboard;
- a layout that differs substantially from the generated design;
- manual control to be more important than automatic maintenance.
Neither mode is inherently better.
They solve different needs.
The important point is to understand that Take Control exchanges automatic maintenance for manual customization.
Before taking control
If you only want to inspect or use the generated dashboard, there is no need to take control.
Do not take control simply because Home Assistant offers the option.
The automatic dashboard is intentionally designed to work without manual editing.
If you later decide that a custom layout is more important, you can then choose Take Control knowing what behaviour will change.
↑ Back to topCustomizing the Automatic Dashboard
Take Control is the recommended customization method for users who want to edit the dashboard through Home Assistant.
There is, however, another possibility for advanced users who specifically want to change the generated dashboard while preserving its automatic behaviour.
Changing the generator
Users who want to modify the generated dashboard without taking control can modify:
frontend/proxmox-dashboard.js
This file contains the JavaScript responsible for generating the automatic dashboard.
Changing the generator can alter the automatic layout or behaviour while retaining the dynamic model instead of converting the dashboard into a manually maintained Home Assistant configuration.
Advanced users only
Editing proxmox-dashboard.js is intended for users who understand JavaScript and are comfortable maintaining their own modifications.
If a user chooses to modify the generator, they are responsible for understanding how their changes interact with:
- automatic resource discovery;
- conditional sections;
- guest placement;
- guest migration;
- future dashboard changes;
- updates to the distributed frontend files.
For most users, Take Control is the appropriate method when manual customization is more important than automatic maintenance.
Modify frontend/proxmox-dashboard.js only when you specifically want to change the automatic generator itself while preserving its dynamic behaviour.
Updates and custom changes
Changes made directly to the distributed JavaScript should be treated as custom project modifications.
When updating the dashboard, review your changes before replacing files or applying a newer version.
A future dashboard version may change the generator internally.
Maintaining a modified generator therefore requires more technical involvement than either using the automatic dashboard unchanged or taking control through Home Assistant.
↑ Back to topTroubleshooting
Dashboard problems should first be separated into two categories:
Is Proxmox Extended Sensors providing the expected data?
or:
Is the dashboard failing to display data that already exists?
This distinction usually makes troubleshooting much easier.
The dashboard does not appear
Confirm that:
- Proxmox Extended Sensors is installed;
- Card Mod is installed;
/proxmox_sensors/proxmox-dashboard.jshas been added as a JavaScript Module;- the browser has been reloaded after adding the resource.
If the Proxmox Extended Sensors option does not appear under Community dashboards, verify the frontend resource before troubleshooting the generated dashboard itself.
The dashboard appears but is empty
Confirm that Proxmox Extended Sensors currently exposes the expected entities in Home Assistant.
Check:
Settings → Devices & services → Proxmox Extended Sensors
If the expected PVE, PBS or Cluster devices and entities exist but the dashboard remains empty, investigate the dashboard resource and frontend loading.
A VM or CT does not appear
First confirm that the guest is selected for monitoring in Proxmox Extended Sensors and that its Home Assistant entities exist.
The automatic dashboard can only display guests that the integration exposes for detailed monitoring.
If the guest was recently added, refresh the dashboard after confirming that Home Assistant has created its entities.
A removed VM or CT is still displayed
If the dashboard is still in automatic mode, first confirm that monitoring for the guest has actually been removed from Proxmox Extended Sensors.
Then refresh the dashboard.
If you previously used Take Control, the dashboard is no longer automatically maintained.
In that case, obsolete cards or sections must be removed manually.
A migrated guest remains under the old node
First determine whether the dashboard is still automatic.
In automatic mode, a monitored guest that migrates within the configured cluster should be represented under its current node after the integration has refreshed the guest location.
If the integration already reports the new node but the automatic dashboard still shows the previous location, refresh the dashboard and investigate the frontend state.
If Take Control has been used, automatic movement between node sections is no longer expected.
The manually controlled dashboard must be updated by the user.
Replication is not displayed
Replication information is conditional.
If no monitored VM or container currently exposes the replication information required by the dashboard, the replication section may not be displayed.
This is expected behaviour and prevents empty dashboard elements from remaining visible.
If replication should exist, verify the corresponding entities in Proxmox Extended Sensors before troubleshooting the dashboard.
Some sensors are missing from the dashboard
This can be completely normal.
The automatic dashboard intentionally does not display every entity exposed by Proxmox Extended Sensors.
It presents a curated selection intended to prioritize useful operational information and avoid overcrowding.
Check Reference if you want to confirm whether an entity exists in the integration.
If you want to display additional entities manually, use Take Control and add them through the Home Assistant dashboard editor.
The dashboard looks unchanged after an update
The browser or Home Assistant frontend may still be using cached JavaScript resources.
Refresh the Home Assistant frontend and perform a hard browser refresh if necessary.
Confirm that the installed dashboard files actually correspond to the version you intended to install.
Do not begin changing dashboard configuration simply to work around a stale frontend cache.
Custom generator changes disappeared after an update
If you modified frontend/proxmox-dashboard.js directly, an update may replace the distributed file.
Direct modifications to the generator are advanced customizations and must be maintained by the user.
Review and reapply compatible changes against the new version rather than assuming they will be preserved automatically.
Take Control no longer updates automatically
This is expected.
After Take Control, Home Assistant preserves the manually controlled dashboard configuration.
Automatic guest addition/removal, migration placement, conditional sections and future generator layout changes are no longer applied to that manually controlled version.
If automatic maintenance is more important than manual editing, use the generated automatic dashboard instead.
Check the browser console
If entities exist correctly in Home Assistant but the dashboard does not load or behaves unexpectedly, browser developer tools may provide useful frontend errors.
Look for errors related to the Proxmox dashboard or its frontend resources.
Remove sensitive information before sharing console output publicly.
Check Home Assistant logs
For problems that may involve the integration rather than only the frontend, open:
Settings → System → Logs
Look for entries related to Proxmox Extended Sensors around the time the problem occurred.
Before opening an issue
Try to determine whether the problem affects:
- dashboard installation;
- frontend resource loading;
- automatic resource discovery;
- a particular PVE/PBS/Cluster view;
- VM or CT discovery;
- guest migration placement;
- conditional content such as replication;
- Take Control behaviour;
- or a custom modification to
proxmox-dashboard.js.
Include:
- Proxmox Extended Sensors version;
- Home Assistant version;
- browser used;
- whether the dashboard is automatic or has used Take Control;
- enough non-sensitive information to reproduce the problem.
If proxmox-dashboard.js has been manually modified, mention that explicitly.
Never include passwords, API Token secrets or other credentials.
If the problem persists after these checks, use the project's GitHub issue tracker.
↑ Back to top