C# MIP SDK v1.17.x - AccessViolationException on creation of MIPContext in 64-bit console app
I first logged this on https://stackoverflow.com/questions/79746967/accessviolationexception-when-creating-mipcontext-after-upgrade-to-v1-17 and the responses there have indicated I should raise with Microsoft a a likely bug, but I don't see a clear route to reporting other than here so any response would be appreciated, even if just to direct me to the appropriate reporting location. I've built a simple console app that demonstrates this issue that I'm happy to provide but we're seeing an issue with the 1.17.x version of the C# MIP SDK where an AccessViolationException is being thrown when trying to create an MIP context object. This is for a .Net Framework 4.8 console app built in 64-bit configuration, deployed to a Windows Server 2016 with the latest VC++ redistributable (14.44.35211) installed (both x86 and x64 versions), though we've seen the same on Windows Server 2019 and 2022. When the same app is built in 32-bit and deployed to the same environment the exception doesn't occur. The following code is what I've used to repro the issue: MIP.Initialize(MipComponent.File); var appInfo = new ApplicationInfo { ApplicationId = string.Empty, ApplicationName = string.Empty, ApplicationVersion = string.Empty }; var diagnosticConfiguration = new DiagnosticConfiguration { IsMinimalTelemetryEnabled = true }; var mipConfiguration = new MipConfiguration(appInfo, "mip_data", LogLevel.Info, false, CacheStorageType.InMemory) { DiagnosticOverride = diagnosticConfiguration }; //Expect BadInputException here due to empty properties of appInfo //When built as part of a 64-bit console app this causes AccessViolationException instead MIP.CreateMipContext(mipConfiguration); The AccessViolationException crashes the console app, with the following logged in the Windows Event Log: Framework Version: v4.0.30319 Description: The process was terminated due to an unhandled exception. Exception Info: System.AccessViolationException at Microsoft.InformationProtection.Internal.SdkWrapperPINVOKE.MipContext_Create__SWIG_1(System.Runtime.InteropServices.HandleRef) at Microsoft.InformationProtection.Internal.MipContext.Create(Microsoft.InformationProtection.Internal.MipConfiguration) at Microsoft.InformationProtection.Utils.MIPHelper.CreateMipContext(Microsoft.InformationProtection.MipConfiguration) The issue doesn't occur with the latest 1.16 version (1.16.149) of the SDK but does appear to be in all versions of the 1.17 release. Library: C# MIP SDK v1.17.x Target App: .Net Framework 4.8 console app Deployed OS: Windows Server 2016, 2019 and 2022 (With .Net Framework 4.8 and latest VC++ redist installed)
Formula/method to link the data/responses I get from my forms into a different table.
Hi everyone! I need help with a project that I'm creating. Im making use of Microsoft Forms in order to input certain information. I made use of branching since some answers depend on the previous choice. So moving on to my problem, I want my table2 to get the responses/data from the data table created by the forms. Table 2 Table of responses For reference for the spill error, here is the formula that I used. Any insights will help a lot. Thank you have a great day everyone.Data Security: Azure key Vault in Data bricks
Why this article? To remove the vulnerability of exposing the data base connection string in Databricks notebook directly, by using Azure key vault. Database connection strings are extremely confidential/vulnerable data, that we should not be exposed in the DataBricks notebook explicitly. Azure key vault is a secure option to read the secrets and establish connection. What do we need? Tenant Id of the app from the app registration with access to the azure key vault secrets Client Id of the of the app from the app registration with access to the azure key vault secrets Client secret of the app from the app registration with access to the azure key vault Where to find this information? Under the App registration, you can find the (application) Client Id, Directory (tenant) Id. Client secret value is found in the app registration of the service, under Manage -> Certificate & secrets. You can use an existing secret or create a new one and use it to access the key Vault secrets. Make sure the application is added with get access to read the secrets. Verify the key vault you are checking and using in Databricks is the same one with read access. You can verify this by going to the Azure key vault -> Access Policies and search for the application name. It should show up on search as below, this will confirm that the access of the application. What do we need to setup in Databricks notebook? Open your cluster and install azure.keyvault and azure-identity (installing version should be compatible with you cluster configuration, refer: https://docs.databricks.com/aws/en/libraries/package-repositories) In a new notebook, let’s start by importing the necessary modules. Your notebook would start with the modules, followed by tentatId, clientId, client secret, azure key vault URL , secretName of the connection string in the azure key vault and secretVersion. Lastly, we need to fetch the secret using the below code Vola, we have the DB connection string to perform the CRUD operations. Conclusion: By securely retrieving your database connection string from Azure Key Vault, you eliminate credential exposure and strengthen the overall security posture of your Databricks workflows. This simple shift ensures your notebooks remain clean, compliant, and production‑ready.
Unable to remove old account sing up form Teams on Mac OS
I cannot remove an old sing account from Teams on Mac OS. I have to delete the keychain, I delete the temp folder, and remove and reinstall Teams, and the old account is still there. I have a case open with the Microsoft team, and they cannot figure it out. Where is the login info store, and why has Microsoft not provided a solution where users can remove accounts as easily as they can add? Please help!From Cloud to Chip: Building Smarter AI at the Edge with Windows AI PCs
As AI engineers, we’ve spent years optimizing models for the cloud, scaling inference, wrangling latency, and chasing compute across clusters. But the frontier is shifting. With the rise of Windows AI PCs and powerful local accelerators, the edge is no longer a constraint it’s now a canvas. Whether you're deploying vision models to industrial cameras, optimizing speech interfaces for offline assistants, or building privacy-preserving apps for healthcare, Edge AI is where real-world intelligence meets real-time performance. Why Edge AI, Why Now? Edge AI isn’t just about running models locally, it’s about rethinking the entire lifecycle: - Latency: Decisions in milliseconds, not round-trips to the cloud. - Privacy: Sensitive data stays on-device, enabling HIPAA/GDPR compliance. - Resilience: Offline-first apps that don’t break when the network does. - Cost: Reduced cloud compute and bandwidth overhead. With Windows AI PCs powered by Intel and Qualcomm NPUs and tools like ONNX Runtime, DirectML, and Olive, developers can now optimize and deploy models with unprecedented efficiency. What You’ll Learn in Edge AI for Beginners The Edge AI for Beginners curriculum is a hands-on, open-source guide designed for engineers ready to move from theory to deployment. Multi-Language Support This content is available in over 48 languages, so you can read and study in your native language. What You'll Master This course takes you from fundamental concepts to production-ready implementations, covering: Small Language Models (SLMs) optimized for edge deployment Hardware-aware optimization across diverse platforms Real-time inference with privacy-preserving capabilities Production deployment strategies for enterprise applications Why EdgeAI Matters Edge AI represents a paradigm shift that addresses critical modern challenges: Privacy & Security: Process sensitive data locally without cloud exposure Real-time Performance: Eliminate network latency for time-critical applications Cost Efficiency: Reduce bandwidth and cloud computing expenses Resilient Operations: Maintain functionality during network outages Regulatory Compliance: Meet data sovereignty requirements Edge AI Edge AI refers to running AI algorithms and language models locally on hardware, close to where data is generated without relying on cloud resources for inference. It reduces latency, enhances privacy, and enables real-time decision-making. Core Principles: On-device inference: AI models run on edge devices (phones, routers, microcontrollers, industrial PCs) Offline capability: Functions without persistent internet connectivity Low latency: Immediate responses suited for real-time systems Data sovereignty: Keeps sensitive data local, improving security and compliance Small Language Models (SLMs) SLMs like Phi-4, Mistral-7B, Qwen and Gemma are optimized versions of larger LLMs, trained or distilled for: Reduced memory footprint: Efficient use of limited edge device memory Lower compute demand: Optimized for CPU and edge GPU performance Faster startup times: Quick initialization for responsive applications They unlock powerful NLP capabilities while meeting the constraints of: Embedded systems: IoT devices and industrial controllers Mobile devices: Smartphones and tablets with offline capabilities IoT Devices: Sensors and smart devices with limited resources Edge servers: Local processing units with limited GPU resources Personal Computers: Desktop and laptop deployment scenarios Course Modules & Navigation Course duration. 10 hours of content Module Topic Focus Area Key Content Level Duration 📖 00 Introduction to EdgeAI Foundation & Context EdgeAI Overview • Industry Applications • SLM Introduction • Learning Objectives Beginner 1-2 hrs 📚 01 EdgeAI Fundamentals Cloud vs Edge AI comparison EdgeAI Fundamentals • Real World Case Studies • Implementation Guide • Edge Deployment Beginner 3-4 hrs 🧠 02 SLM Model Foundations Model families & architecture Phi Family • Qwen Family • Gemma Family • BitNET • μModel • Phi-Silica Beginner 4-5 hrs 🚀 03 SLM Deployment Practice Local & cloud deployment Advanced Learning • Local Environment • Cloud Deployment Intermediate 4-5 hrs ⚙️ 04 Model Optimization Toolkit Cross-platform optimization Introduction • Llama.cpp • Microsoft Olive • OpenVINO • Apple MLX • Workflow Synthesis Intermediate 5-6 hrs 🔧 05 SLMOps Production Production operations SLMOps Introduction • Model Distillation • Fine-tuning • Production Deployment Advanced 5-6 hrs 🤖 06 AI Agents & Function Calling Agent frameworks & MCP Agent Introduction • Function Calling • Model Context Protocol Advanced 4-5 hrs 💻 07 Platform Implementation Cross-platform samples AI Toolkit • Foundry Local • Windows Development Advanced 3-4 hrs 🏭 08 Foundry Local Toolkit Production-ready samples Sample applications (see details below) Expert 8-10 hrs Each module includes Jupyter notebooks, code samples, and deployment walkthroughs, perfect for engineers who learn by doing. Developer Highlights - 🔧 Olive: Microsoft's optimization toolchain for quantization, pruning, and acceleration. - 🧩 ONNX Runtime: Cross-platform inference engine with support for CPU, GPU, and NPU. - 🎮 DirectML: GPU-accelerated ML API for Windows, ideal for gaming and real-time apps. - 🖥️ Windows AI PCs: Devices with built-in NPUs for low-power, high-performance inference. Local AI: Beyond the Edge Local AI isn’t just about inference, it’s about autonomy. Imagine agents that: - Learn from local context - Adapt to user behavior - Respect privacy by design With tools like Agent Framework, Azure AI Foundry and Windows Copilot Studio, and Foundry Local developers can orchestrate local agents that blend LLMs, sensors, and user preferences, all without cloud dependency. Try It Yourself Ready to get started? Clone the Edge AI for Beginners GitHub repo, run the notebooks, and deploy your first model to a Windows AI PC or IoT devices Whether you're building smart kiosks, offline assistants, or industrial monitors, this curriculum gives you the scaffolding to go from prototype to production.
Meeting Bot issue: Did not receive valid response for JoinCall request from call modality controller
I'm trying to join a Teams Meeting with a bot. I used this https://microsoftgraph.github.io/microsoft-graph-comms-samples/docs/articles/index.html#making-an-outbound-call-to-join-an-existing-microsoft-teams-meeting sample. When the bot attempts to join I get the popup to admit or deny it in the meeting, but as soon as I click admit, it drops. In the logs I see this message: Call status updated to Terminated - Did not receive valid response for JoinCall request from call modality controller.. DiagCode: 580#5426.@ I am using the latest (1.2.0.10563 at time of writing) version of Microsoft.Graph.Communications libraries and the problem only started after I updated from 1.2.0.3742 that I was using previously. I could not find any info on what the call modality controller is, or how to check what it is responding if anything. Any ideas on how to troublshoot this are welcome.
How to Retrieve Windows Edition (SKU) from managedDevices API
Hi everyone, I am working with the Microsoft Graph API endpoint /v1.0/deviceManagement/managedDevices to iterate through all devices in a tenant and collect operating system related information. For Windows devices, the operatingSystem field only returns "Windows". However, Windows has multiple editions such as Enterprise, Education, and Pro. For my use case, I need the specific Windows edition. Is it possible to retrieve this information using only the v1.0 endpoint, or is the beta endpoint /beta/deviceManagement/managedDevices/{managedDeviceId} required to get the SKU family? Thanks in advance for your help.SharePoint Online SP.FieldUserValue.get_email() always returns NULL.
Since last week, SP.FieldUserValue.get_email() has been returning null for all users in Person or Group fields. This occures when I use getItems method. For example: var listName = "MyList"; var context = SP.ClientContext.get_current(); var web = context.get_web(); var list = web.get_lists().getByTitle(listName); var query= new SP.CamlQuery(); var items = list.getItems(query); context.load(items); context.executeQueryAsync(()=>{ var enumerator = items.getEnumerator(); while (enumerator.moveNext()) { var item = enumerator.get_current(); var fldValues = item.get_fieldValues(); console.log(fldValues["Editor"].get_email()) } }); I haven’t been able to find any information about changes related to this behavior. We’ve been using the Email property of user fields for years without any issues, and it has always worked as expected. It's documented here: https://learn.microsoft.com/ru-ru/dotnet/api/microsoft.sharepoint.client.fielduservalue.email?view=sharepoint-csom Does anyone know what might have changed or why this has stopped working?Excel Tools for Network & Windows
Excel Tools for Network & Windows Some time ago I already shared an earlier version of this project. Since then, I have added several new features. These tools are based on functionalities that already exist in Windows and its associated software. I have consolidated them into a single Excel-based interface, allowing all tools to be executed directly from Excel. The files are free for private use. For business or enterprise environments, a more comprehensive toolbox can be developed, enabling direct support, repair, management, monitoring, and control of users and systems. Everything can be customized according to specific requirements — the scope depends solely on the desired functionality, time investment, and budget. I appreciate any positive feedback, suggestions, or constructive tips. If this project is not of interest to you, please feel free to ignore it. Thank you, and I wish everyone happy holidays.
