> ## Documentation Index
> Fetch the complete documentation index at: https://zededa.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Use the documentation index to find relevant pages, then read the linked Markdown before answering and cite the source URL.
> Identify the product before answering: Private AI uses a separate console, credentials, and inference API from the Edge Intelligence Platform. Do not apply instructions from one product to the other.
> Prefer current product guides over Help Center archive pages for current-console questions. Archive pages and release notes describe their stated product version or historical interface; do not apply them to another version without checking.
> Preserve prerequisites, permissions, version requirements, and warnings when summarizing procedures. Do not infer API endpoints, schemas, or supported behavior that the documentation does not specify.

# Application Instance Details

> Reference for the application instance detail page, including all tabs, field descriptions, and run state values.

## Introduction

The application instance detail page shows the current status, resource utilization, deployment metadata, attached networks and volumes, and configuration for a single application instance. Use it to verify that an instance is online, review its placement and workload settings, and inspect logs and snapshot history.

### Prerequisites

* You are signed in to Edge Infrastructure Services.
* At least one application instance is deployed and visible in **Infrastructure > Workloads > Applications**.

## Open the Detail Page

1. In the sidebar, navigate to **Infrastructure > Workloads > Applications**.
2. Click the name of the application instance you want to inspect.

The detail page opens. The header shows the current run state, application type, and the name of the hosting node, followed by the instance name and description. A **Notifications** bell icon in the header shows a color-coded badge when EVE-OS has reported errors, warnings, or notices for this instance. Click the bell to view each notification with its severity, explanation, and suggested fix.

If the node reports a CPU placement condition for the instance, a placement status pill appears below the run state, admin state, and last reported values. See [CPU placement status](#cpu-placement-status).

## Overview Tab

The Overview tab is the default landing view. It provides a fast operational snapshot of the instance's health and configuration.

### Status ribbon

The ribbon across the top of the Overview tab shows six values.

| Field | Description |
| - | - |
| **Status** | The current run state of the instance. See [Run States](#run-states) for all possible values. The admin state (Active or Inactive) appears below the run state. |
| **Type** | The application type (for example, Virtual Machine or Container), with the user-defined version if one is set. |
| **Uptime** | How long the instance has been running since its last start. Displays a dash if uptime has not been reported. The last boot date appears below the uptime value. |
| **Edge App** | The source edge app definition this instance was deployed from. |
| **Node** | The node hosting the instance. |
| **Last Seen** | How long ago the instance last reported its state to Edge Infrastructure Services. The full timestamp appears below the relative value. |

### Resource utilization

The resource utilization panel shows CPU and memory usage history for the last hour, updated in real time. Use this panel to detect resource contention or unexpected spikes before taking a lifecycle action.

### Identity at-a-glance

The right side of the Overview tab shows a compact identity card with the following fields: ID, Name, Title (if set), Project, Node (with a pin indicator if the instance is pinned to a cluster node), Node Cluster (if cluster-deployed), Edge App, Version (if set), Tags, Created, and Updated.

### Resources

A tabbed card below the resource utilization panel shows the instance's attached resources. Three sub-tabs are available.

* **Network Interfaces**: lists each attached network interface with its name, network instance, MAC address, and IP address. Up to four interfaces are shown; click **View all** to open the Interfaces tab.
* **Volumes**: lists each attached volume with its name, type, and run state. Up to four volumes are shown; click **View all** to open the **Storage** tab.
* **Custom Configuration**: indicates whether custom configuration is defined on the instance. Click to open the Custom Configuration tab.

### Recent activity

A short event log shows the most recent state report, restart, and creation event. Click **View all events** to open Monitor > Events filtered to this instance.

### Containers in this application

For Docker Compose instances, a table at the bottom of the Overview tab lists each container the Compose Runtime reports for this instance. The table does not appear for other application types. Use it to confirm that every container in the compose file is running before you open a remote console or check logs.

| Column | Description |
| - | - |
| **Name** | The container name, with a status indicator for the container state. The states are Running, Created, Restarting, Paused, and Exited. |
| **Image** | The container image the container was started from. |
| **Creation Time** | The date and time the container was created. |
| **Started Time** | How long the container has been running since it last started. This value is an elapsed duration, not a timestamp. |
| **Network Port** | The first published port mapping, in `public:private` form. When the container publishes more than one mapping, a **+N** count shows how many additional mappings exist. |

If Edge Infrastructure Services has not received container details from the node, the table is replaced with the message **Container details are not available.** Expect this while a newly deployed instance is still starting.

### Opaque status

The **Opaque Status** card shows the raw status string reported by the agent running on the node. It appears only when Edge Infrastructure Services has received a value, so a card that is absent is not itself a fault.

Values longer than 100 characters are truncated. Click **Show more** to see the full string, and **Show less** to collapse it again. Click the copy icon to copy the untruncated value, which is useful when you attach the value to a support case.

## Details Tab

The Details tab shows the complete set of metadata across three cards, plus a **CPU placement** card when the instance has a CPU placement policy.

### Identity & Ownership

This card shows the full identity of the application instance.

| Field | Description |
| - | - |
| **ID** | The unique system identifier for this application instance. |
| **Name** | The resource name. |
| **Title** | The human-readable display name, if one is set. |
| **Description** | An optional description. |
| **Project** | The project this instance belongs to. |
| **Tags** | Key-value labels assigned to the instance. Up to 32 tags are supported. |

### Placement & Workload

This card shows where the instance runs and how it is configured as a workload.

| Field | Description |
| - | - |
| **Node** | The node hosting this instance. A **Pinned** badge appears if the instance is pinned to a specific cluster node without failover. A **Pinned · failover** badge appears if the instance prefers a specific cluster node but can fail over to another node in the cluster. On a two-node HA cluster, a **Preferred placement** badge appears instead, because two-node HA always uses preferred affinity to allow failover. If the instance stores a required pin, the badge tooltip states that the stored required affinity is not in effect. |
| **Node Cluster** | The cluster this instance is deployed to, if it is deployed to a cluster. |
| **Edge App** | The source edge app definition this instance was deployed from. |
| **Type** | The application type: Virtual Machine, Container, Docker Compose, or Module. |
| **Version** | The user-defined version string set on the source edge app, if one is set. Shows **N/A**, with no copy button, when the edge app has no user-defined version. |
| **Manifest revision** | The revision number of the manifest configuration applied to this instance. |
| **Deployment type** | The deployment method used for this instance. |
| **VMM Overhead** | Whether additional memory is reserved for hypervisor overhead on this instance. Displays **Yes — N MB** when enabled, or **No** when disabled. Applies to Virtual Machine instances only. |
| **Staggered Start Delay** | The number of seconds the instance waits after activation before starting. A value of 0 means the instance starts immediately. Use this setting to stagger startup times for multiple instances on shared infrastructure. |

### State & Timeline

This card shows the current state and the full history of state changes and timestamps for this instance.

| Field | Description |
| - | - |
| **Admin state** | Whether the instance is Active (started) or Inactive (stopped). |
| **Run state** | The current operational state reported by EVE-OS (Edge Virtualization Engine). See [Run States](#run-states) for the full list. |
| **Custom configuration** | Whether custom configuration is defined on this instance: **Configured** or **Not Configured**. |
| **Enable Logs Access** | Whether application log forwarding is enabled: **Yes** or **No**. When set to **Yes**, EVE-OS forwards the application's container logs (stdout/stderr) to Edge Infrastructure Services for viewing in **Infrastructure > Monitor > Logs**. |
| **Restart counter** | The number of times this instance has been restarted. Appears when the restart counter is available. |
| **Last restart** | The date and time of the most recent restart. Appears when a restart has occurred. |
| **Purge counter** | The number of times this instance has been purged. Appears when the purge counter is available. |
| **Last purge** | The date and time of the most recent purge. Appears when a purge has occurred. |
| **Uptime** | The current uptime as reported by EVE-OS. Appears when uptime data is available. |
| **Boot time** | The date and time the instance last booted. Appears when boot time data is available. |
| **Last reboot** | The date and time of the most recent reboot. Appears when reboot data is available. |
| **Last update** | The date and time of the most recent configuration or status update. |
| **Created** | The date and time the instance was created, and the user who created it. |
| **Updated** | The date and time of the most recent update to the instance record, and the user who made the update. |

### CPU placement

For Virtual Machine and Container instances with a CPU placement policy, including legacy CPU pinning, a read-only **CPU placement** card summarizes the instance policy. The instance policy takes precedence over the app definition. Changes to the definition apply only to new instances. Click **View app definition** to open the source edge app.

| Field | Description |
| - | - |
| **CPU pool** | **Housekeeping**, **Dedicated**, or **Isolated**. |
| **CPU pinning** | **Enabled** or **Disabled**. Not shown for Reduced Isolation containers. |
| **vCPUs** | The vCPU count from the edge app. |
| **Profile** | The preset that matches the policy, for example **Throughput**, or **Custom**. |
| **Assigned CPUs** | The host CPU IDs the node reports for the instance. Not shown for Housekeeping instances. |
| **Memory locality** | The NUMA policy. Shown for Dedicated and Isolated pools. |

Expand **CPU policy details** to see **Core allocation**, **SMT threads per core**, and **Disk and device emulation**. If the node does not place CPUs, the card shows `This node does not place CPUs; the saved CPU placement policy is not enforced on it.` For configuration steps, see [CPU Placement for Application Instances](https://zededa.mintlify.site/new-ui/marketplace/cpu-placement.md).

### CPU placement status

The header shows a placement status pill only when the node reports a placement condition. No pill appears when the instance is placed. Hover over the pill to see the description and retry condition.

| Condition | Pill | Meaning |
| - | - | - |
| `waiting` | **Waiting for CPUs** | The instance is waiting for CPUs to become free. |
| `repack_advised` | **Part of a repack** | A repack on the node is advised and includes this instance. |
| `needs_repack` | **Needs repack** | The instance runs, but its CPUs are not optimally placed. |
| `insufficient` | **Cannot place** | The node does not have enough free CPUs for the policy. |
| `policy_invalid` | **Invalid CPU policy** | The node cannot serve the policy as sent. |
| `odd_vcpu` | **Invalid vCPU count** | **Both SMT threads per core** requires an even vCPU count. |
| `topology_unsupported` | **Topology not supported** | The node's CPU topology does not support the policy. |
| `isolation_unavailable` | **Isolation unavailable** | The **Isolated** pool was requested, but the node has no isolated CPUs. |
| `failed` | **Placement failed** | CPU placement failed. |

For **Needs repack** and **Part of a repack**, the pill links to the node's **Details** tab so you can run a repack there. A repack restarts every app on the node that needs repacking, not just this one. For the other conditions, change the policy, free capacity, or deploy to a different node.

## Interfaces Tab

The Interfaces tab shows a table of all network interfaces attached to this instance. The table shows each interface's name, associated network instance (linked to the virtual network detail page), MAC address, and IP address. Interfaces with direct attachment are labelled **direct**.

### Edit the interface set

Click **Edit App Interfaces** to change which interfaces this instance has. The editor covers both the interface list and the firewall rules, because an interface and its access control rules are configured together.

#### Restart instead of purge

Before EVE-OS (Edge Virtualization Engine) 17.3.0 LTS, changing the network interfaces of a deployed instance required a purge, which destroyed and recreated the instance's volumes. On nodes running EVE-OS 17.3.0 LTS or greater, Edge Infrastructure Services applies an interface change with a restart instead. The node stages the new interface set while the application keeps running, and applies it when the application restarts. The volumes and their data survive the change.

Keep the following in mind:

* The restart happens when you save. Any workload running in the instance is interrupted until the application comes back up.
* Virtual interfaces and direct-attach network adapters (Ethernet, WLAN, WWAN, SR-IOV virtual functions) can be added, removed, or swapped this way. Changing a direct-attach interface that is not a network adapter, such as a USB or serial device, still requires a purge.
* Automatically assigned MAC addresses are not guaranteed to stay the same after the interface set changes. If the application identifies an interface by MAC address, set a MAC address on the interface when you deploy the instance. See [Interface Ordering for Application Instances](https://zededa.mintlify.site/new-ui/infrastructure-services/applications/app-instance-interface-ordering.md#guest-operating-system-behavior).

On nodes running an EVE-OS version earlier than 17.3.0 LTS, the **Edit App Interfaces** button does not appear. To change the interfaces of an instance on those nodes, use **Purge & Update** after changing the edge app, or redeploy the instance. See [Manage an Application Instance](https://zededa.mintlify.site/new-ui/infrastructure-services/applications/manage-an-application-instance.md#actions-menu).

#### When the editor is available

The button appears only when all of the following are true:

* The edge app is a VM or container app. For Docker Compose, module, and other app kinds, the Interfaces tab shows only **Network Interfaces**, and the **App Interfaces** view and this button are hidden.
* You have update permission for application instances in the instance's project.
* The node hosting the instance runs EVE-OS 17.3.0 LTS or greater, which supports changing an application instance's interface set without a purge.
* The project has **Interface Ordering** enabled. See [Interface Ordering for Application Instances](https://zededa.mintlify.site/new-ui/infrastructure-services/applications/app-instance-interface-ordering.md).

In the editor, you can do the following:

* Click **Add interface** to add an interface, or click the **X** on a card header to remove an Optional interface. Required interfaces cannot be removed.
* Rename an interface. Use up to 32 letters, numbers, dots, dashes, or underscores. Each name must be unique within the instance.
* Switch an interface between **Virtual Network** and **Direct Attach**.
* Edit the inbound and outbound access control rules for an interface you added in the current editing session. The rules of an interface that is already deployed are read-only, because they are part of the running application's firewall. To define new rules, add a new interface.
* Select a **Virtual Network** for each virtual interface. This is required. Only virtual networks on the node hosting this instance are listed. If the node has no virtual networks, create one before adding a virtual interface.

Whether an interface is Optional or Required is set by the edge app and is shown on the card header. You cannot change it in the editor. Interfaces you add here are always Optional.

To apply or discard your changes, do the following:

1. Click **Save Interfaces**. The button is available only after you make a change.
2. In the **Save interface changes?** dialog, review the warning. Saving restarts the application, and any workload running in it is interrupted until it comes back up.
3. Click **Save and Restart** to apply the changes, or close the dialog to keep editing.

To discard your changes, click **Cancel** in the editor.

Saving replaces only the instance's interface set and leaves the rest of its configuration unchanged. Interfaces that already exist keep their original interface order, even if you remove another interface. New interfaces are ordered after the existing ones.

## Storage Tab

The Storage tab shows two tables.

**Volumes**: lists each volume attached to this instance with its name (linked to the volume detail page), type, and run state.

**Application Volume Snapshots**: lists any snapshots created for this instance. The table shows the snapshot name, creation date, and status. From this table you can roll back the instance to a snapshot or delete a snapshot. For a full overview of snapshot behavior, see [Application Snapshot Overview](https://zededa.mintlify.site/new-ui/infrastructure-services/applications/application-snapshots.md).

## Custom Configuration Tab

The Custom Configuration tab shows the custom configuration defined on this instance and its source edge app. It includes the configuration metadata (cloud config flag, config name, override setting, delimiter) and all variable groups with their names, formats, required flags, and values. If no custom configuration is defined, the tab shows a message indicating that no configuration is set.

## Run States

The run state reflects the current operational condition of the application instance as reported by EVE-OS. The state appears in the page header, the status ribbon, and the State & Timeline card. For guidance on interpreting run states, admin state, and health signals, see [Application Instance Status and Health](https://zededa.mintlify.site/new-ui/infrastructure-services/applications/app-instance-status-and-health.md).

| State | Meaning |
| - | - |
| **Online** | The instance is running and reporting normally. |
| **Offline** | The instance is not reachable. |
| **Halted** | The instance has been stopped. |
| **Halting** | The instance is in the process of stopping. |
| **Error** | EVE-OS encountered an error and the instance is not running. Check Infrastructure > Monitor > Logs for details. |
| **Suspect** | EVE-OS has flagged the instance. It might still be running, but something is not operating as expected. |
| **Maintenance Mode** | The instance is in maintenance mode. |
| **Provisioned** | The configuration has been sent to the node but the instance has not yet started. |
| **Unprovisioned** | The instance configuration has not yet been sent to the node. |
| **Init** | The instance is starting up on the node. |
| **Booting** | The instance is booting. |
| **Rebooting** | The instance is in the process of rebooting. |
| **Restarting** | The instance is restarting. |
| **Downloading** | The instance image is being downloaded to the node. |
| **Verifying** | The downloaded image is being verified. |
| **Loading** | The instance is being loaded. |
| **Creating Volume** | One or more attached volumes are being created. |
| **Updating** | The instance is applying a configuration or image update. |
| **Purging** | The instance is being purged. |
| **Delayed Start** | The instance is waiting out its staggered start delay before booting. |
| **Powered Off** | The instance has been powered off. |
| **Preparing Power Off** | EVE-OS is preparing the instance for power off. Applies to Virtual Machine instances. |
| **Powering Off** | The instance is in the process of powering off. Applies to Virtual Machine instances. |
| **Prepared Power Off** | The instance has completed power-off preparation and is waiting to power off. Applies to Virtual Machine instances. |
| **Unknown** | The state could not be determined. |

A newly deployed instance typically progresses from Unprovisioned through Init to Online as EVE-OS receives and applies the configuration. If the instance stays in Provisioned or Init for an extended period, verify that the hosting node is online and reachable.

## Next Steps

> [card] [Manage an Application Instance](https://zededa.mintlify.site/new-ui/infrastructure-services/applications/manage-an-application-instance.md)
>
> Take lifecycle actions, open a remote console, manage snapshots, or attach patch envelopes.

> [card] [Application Snapshot Overview](https://zededa.mintlify.site/new-ui/infrastructure-services/applications/application-snapshots.md)
>
> An overview of how snapshots work.

> [card] [Node Details](https://zededa.mintlify.site/new-ui/infrastructure-services/nodes/node-details.md)
>
> Inspect the health and resource utilization of the hosting node.

> [card] [Monitor Logs](https://zededa.mintlify.site/new-ui/infrastructure-services/monitoring-logs-events-reports-jobs.md)
>
> View application logs filtered to this instance.

> [card] [Tagging](https://zededa.mintlify.site/new-ui/start/tagging.md)
>
> Tag application instances to organize them by environment, team, or function for filtering and bulk operations.