> ## 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. # Release 17.6 > Edge Infrastructure Services GUI 17.6.6 and API 17.6.7, published November 4, 2025. **Published:** 11/4/2025 GUI: 17.6.6 | API: 17.6.7 ## New GA Features ### ZEDEDA Edge Kubernetes Service and ZEDEDA Edge Kubernetes App Flows The ZEDEDA Edge Kubernetes Solution extends the power of VM and container orchestration to resource-constrained edge devices. * This platform-native implementation enables the deployment, scaling, and management of applications across diverse edge locations. * It leverages open-source Kubernetes' extensible architecture and adapts it for edge environments, providing a robust, scalable, and user-friendly platform that addresses the unique challenges of edge computing. * A new top-level menu in the GUI, **Edge Kubernetes**, is introduced to support the new functionality. * This feature requires EVE-k v16.0.0 or greater. * Note that ZEDEDA Edge Kubernetes Service and App Flows is a commercial capability. Contact your ZEDEDA sales representative for more details. * Terraform Provider support is a work in progress. ### ZEDEDA Flash Support Introducing [ZEDEDA Flash](https://github.com/zededa/zflash) for Mac, Windows, and Linux. This new utility simplifies the process of flashing EVE-OS (Edge Virtualization Engine) images to a USB drive or an SD card during the process of onboarding nodes. The ZEDEDA Flash UI streamlines the initial steps by providing a user interface that reduces configuration errors to ensure consistency. ZEDEDA Flash includes safety features, such as system drive protection, to prevent accidentally flashing your laptop's main drive. Other key features include: * Easily choose EVE-OS LTS images directly from the official GitHub repository. * Support for AMD64, (generic) ARM64, and NVIDIA Jetson (JP5/JP6) images. * Local caching of downloaded images. * Support for common file formats such as .raw, .img, .wic, .zip, .gz, .xz, .iso, and .zst for generic flashing needs. ### Edge Sync UI The existing Edge Sync provides a solution for air-gapped environments allowing control over a fleet of EVE devices through an intermediary system, which is the Edge Sync server. You can now use the new Edge Sync UI to interact with your Edge Sync instance (in addition to the existing API interactions). You can use the Edge Sync UI to observe status information for troubleshooting, and to apply a new EVE-OS config for upgrading the application stack. Note that Edge Sync UI is a commercial capability. Contact your ZEDEDA sales representative for more details. ### Node Cluster three-node or two-node with tie-breaker The Node Cluster feature is now GA and also has two configuration options. You can configure the original three-node high availability (HA) or you can configure two-node HA with one extra node local tie-breaker. * See the prerequisites for two-node HA. * When you onboard your tie-breaker node, tag your node with **tie-breaker:true**. * This feature requires EVE-k v16.0.0 or greater. ## Enhanced GA Features ### Docker Compose enhancements This release adds several key improvements for Docker Compose. * Improved customization: with \$zri variable support for Docker Compose, containers can now dynamically learn the node's name and other parameters through the cloud-init mechanism for essential first-boot tasks. See [Custom Configuration for Applications](https://zededa.mintlify.site/new-ui/marketplace/custom-app-configuration.md) for examples. * Bulk deployments: deploy Docker Compose runtime and app for multiple nodes and policies: * Import the app update from **Marketplace > Global**. * Automate fleet management with Application Profile Deployment and Deployment Projects. ### Static IP — adapter specific During node onboarding, you no longer have to create specific network objects per static IP address. You now have the option to reduce the number of static network objects in your enterprise by creating a general adapter-specific network, which allows you to manually enter IP details on-the-fly rather than choosing from pre-configured options. ### Disable vTPM vTPM is enabled by default for newly-created edge apps of type VM, but now you can disable vTPM during initial VM creation if it is not required (such as for key generation, attestation, and sealed storage in environments without a physical TPM device) or for compatibility issues with legacy applications. Note that if you edit an existing **Marketplace > Edge App** to disable vTPM, and then propagate the updated config out to your edge app instances, it is a condition that will trigger a Purge and Update. It is a destructive action. For example, if you use Windows Bitlocker and update your edge app instance to disable vTPM, you will lose all your Bitlocker keys. ### Add VLAN sub-interface directly to a Switch Network Instance You can now add a VLAN sub-interface directly to a Switch Network Instance: * From the existing **Library > Network Instance > (+)**, when adding a **Switch** as the kind of Network Instance, the **Port** drop-down now displays the available **VLAN IDs**. * This allows you to configure a combination of Switch Network Instance with access VLANs and EVE-OS management VLAN sub-interfaces, plus Local Network Instance VLAN sub-interfaces, on the same physical port. * This feature requires EVE-OS version 14.5.2-LTS or greater. ### Local network reconfiguration From the existing **Nodes > NODE\_NAME > Adapters** tab, you can now enable **Allow modifications locally (LPS)** to permit local management of network port configuration with Local Profile Server so that you can: * Reconfigure IP while disconnected from Edge Infrastructure Services. * Reconfigure IP by an onsite operator that doesn't have Edge Infrastructure Services access. * This feature requires EVE-OS version 14.5.2-LTS or greater. ### Single-use installer advanced configuration As part of the node onboarding flow, you can generate a single-use installer now with EVE-OS and application storage on a single drive of your choice, or you can specify EVE-OS and application storage disks on separate drives of your choice. To use this option, you must know the exact name of your installation disk (such as sda) or storage disk (such as sdb). ### Retrieve onboarding details later In the node onboarding workflow, the **Retrieve Onboarding details later** option is now removed from the **Hardware Identity** drop-down. This GUI change reflects a backend change that had already been removed. ### Clear session The Clear Session link is added to **Node > Remote Access > Access Application** tab so that you can clear an existing VNC session. For SSH, you can close the terminal window to clear the session. ## New Early Access Features ### ServiceGraph Connector for ServiceNow Integration (Beta) This release introduces the ZEDEDA ServiceGraph Connector for ServiceNow, a new integration developed by ZEDEDA. The connector continuously synchronizes critical data (hardware details, location, status) from your ZEDEDA-managed nodes directly into the ServiceNow Configuration Management Database (CMDB), ServiceNow's system of record for IT and OT assets. This integration allows you to view and manage your nodes and OT assets directly alongside your traditional corporate servers and laptops within the ServiceNow platform. For example, you can automatically create incident tickets for node failures or leverage ServiceNow's native workflows to streamline field service operations. * Contact your sales representative to get on the early access list and provide feedback. * This feature will be generally available (GA) in H1 2026. ## Known Issues * To have the collect-info feature working correctly with Edge Sync, the datastore configured on the node must support **POST** API calls to create resources. If it does not, the collect-info upload will fail. The current EVE-OS implementation does not dynamically adjust the upload mechanism based on the server's supported capabilities. If you are using third-party HTTP servers or your own datastore setups that don't natively support POST, for Edge Sync you need to ensure that additional configurations are applied to handle POST (such as enabling WebDAV or adding a custom handler script) for the collect-info functionality to work correctly. * Docker Compose: A Compose App Instance will create a job that shows "the requested job does not have any individual tasks" for a few minutes before generating the run status. * ZEDEDA Edge Kubernetes App Flows: Downloading kubeconfig from the cluster dashboard always provides read-only access to kubectl. * ZEDEDA Edge Kubernetes App Flows: When attempting to update a resource in a cluster via the cluster **Dashboard > Import YAML** option, 500 internal server errors are encountered due to permission restrictions for shadow users in the default namespace. * ZEDEDA Edge Kubernetes App Flows: When configuring an app repo of type HTTP(S) under **Edge Kubernetes > Kubernetes Marketplace**, the icon for the HTTP(S) repo is the GitHub icon. ## Other News ### EVE-k EVE-Kubevirt, introduced as early-access in 14.5.0, is being renamed to **EVE-k**. For example, image names will change from `*-kubevirt-amd64` to `*-k-amd64`. EVE-k is a k3s-based EVE-OS variant that enables edge-node clustering and ZEDEDA Edge Kubernetes Service. The standard EVE-OS variant is unchanged and remains the recommended option for deployments not using these capabilities. * For installation, we recommend all the nodes be of the same hardware type and model with a minimum of 12-core CPUs, 16 GB RAM, and NVMe SSDs for EVE persistent storage for optimal performance. * For runtime, EVE-k consumes approximately 4 GB of memory and 2 vCPUs if used for clustering. * It is recommended, though not strictly required, to separate management traffic from cluster traffic.