dlm-diagnostics

作者: microsoft

诊断 Microsoft Purview 数据生命周期管理 (DLM) 在 Exchange Online 中的问题。当用户报告保留策略未应用于…时,使用此技能。

npx skills add https://github.com/microsoft/purview-dlm-mcp --skill dlm-diagnostics

DLM Diagnostics

Investigate and resolve Microsoft Purview Data Lifecycle Management issues in Exchange Online. Follow the decision tree below to identify the correct troubleshooting guide, then load the matching reference file for step-by-step diagnostic instructions.

Prerequisites

Before running any diagnostic commands, ensure both PowerShell sessions are connected:

Connect-ExchangeOnline -UserPrincipalName '<admin@contoso.com>' -ShowBanner:$false
Connect-IPPSSession -UserPrincipalName '<admin@contoso.com>' -ShowBanner:$false

Safety Rules

Only execute read-only commands. Allowed cmdlet verbs: Get-*, Test-*, Export-*.

Never execute mutating commands — Set-*, New-*, Remove-*, Enable-*, Start-*, Invoke-* cmdlets must only be returned as text recommendations for the admin to review and run manually.

Decision Tree

Identify the reported symptom and load the matching reference:

SymptomReference
Policy shows Success but content is not retained or deleted on target workloadsretention-policy-not-applying.md
Policy status shows Error, PolicySyncTimeout, or PendingDeletionpolicy-stuck-error.md
Archive mailbox exists but items stay in the primary mailboxitems-not-moving-to-archive.md
Archive is near 100 GB but no auxiliary archive is being createdauto-expanding-archive.md
User was deleted but mailbox was purged instead of becoming inactiveinactive-mailbox.md
Recoverable Items folder growing uncontrollably or SubstrateHolds is largesubstrateholds-quota.md
Teams retention policy exists but messages remain visible past retention periodteams-messages-not-deleting.md
Both MRM and Purview retention on a mailbox causing unexpected behaviormrm-purview-conflict.md
Adaptive scope includes wrong members, shows no members, or scope query not targeting correct users/sitesadaptive-scope.md
Auto-apply retention label policy not labeling content, stuck in simulation, or shows "Off (Error)"auto-apply-labels.md
SharePoint site cannot be deleted because a retention policy or hold is blocking itsharepoint-site-deletion-blocked.md

If the issue does not match a specific symptom above, or for ad-hoc investigations, see diagnostic-commands.md for a quick-reference of common diagnostic commands. For adaptive scope validation commands, see adaptive-scope.md.

Workflow

  1. Identify the symptom from the user's description using the decision tree above.
  2. Load the matching reference file — read it fully before starting the investigation.
  3. Execute diagnostic commands step by step using the run_powershell MCP tool, following the reference guide's sequence.
  4. Interpret results at each step — the reference guide explains what to look for and how each finding maps to a root cause.
  5. Cross-reference — some guides link to sibling references (e.g., policy distribution failures link to policy-stuck-error.md). Load those when directed.
  6. Report findings — summarize the root cause, present the root-cause confirmation table, and provide the recommended remediation actions as text for the admin to review.
  7. Review the execution log — use the get_execution_log MCP tool to include a full audit trail of all commands run during the investigation.

Output Format

Present findings to the user in this structure:

## Investigation Summary

**Symptom:** <what was reported>
**Mailbox/Policy:** <target>

## Diagnostic Steps

✅/❌ Step N — <step name>
   Finding: <what was found>
   Command: <command that was run>

## Root Cause

<identified root cause from the reference guide's root-cause table>

## Recommended Actions

⚠️ These commands are NOT executed automatically. Review and run manually:
- <remediation command 1>
- <remediation command 2>

来自 microsoft 的更多技能

oss-growth
microsoft
OSS增长黑客角色
agent-framework-azure-ai-py
microsoft
使用Microsoft Agent Framework Python SDK(agent-framework-azure-ai)构建Azure AI Foundry代理。在创建使用AzureAIAgentsProvider的持久化代理、使用托管工具(代码解释器、文件搜索、网络搜索)、集成MCP服务器、管理对话线程或实现流式响应时使用。涵盖函数工具、结构化输出和多工具代理。
development
airunway-aks-setup
microsoft
在AKS上设置AI Runway——从裸集群到运行模型。涵盖集群验证、控制器安装、GPU评估、提供商设置和首次部署。适用场景:“设置AI Runway”、“接入AKS集群”、“安装AI Runway”、“airunway设置”、“将模型部署到AKS”、“在AKS上进行GPU推理”、“在AKS上配置KAITO”、“在AKS上运行LLM”、“在AKS上使用vLLM”、“在AKS上设置模型服务”、“AI Runway控制器”。
devops
appinsights-instrumentation
microsoft
使用Azure Application Insights对Web应用进行插桩的指南。提供遥测模式、SDK设置和配置参考。适用场景:如何对应用进行插桩、App Insights SDK、遥测模式、什么是App Insights、Application Insights指南、插桩示例、APM最佳实践。
devops
applicationinsights-web-ts
microsoft
使用Application Insights JavaScript SDK(@microsoft/applicationinsights-web)为浏览器/Web应用添加检测。用于真实用户监控(RUM)——页面视图、点击、AJAX/fetch依赖项、异常、自定义事件,以及与后端OpenTelemetry追踪关联的浏览器端GenAI代理追踪。涵盖SDK加载器脚本和npm设置、框架扩展(React、React Native、Angular)、点击分析、遥测初始化器,以及从浏览器发出的代理/工具/模型跨度所遵循的OTel GenAI语义约定。
devops
azure-ai-anomalydetector-java
microsoft
使用适用于 Java 的 Azure AI 异常检测器 SDK 构建异常检测应用程序。在实现单变量/多变量异常检测、时间序列分析或 AI 驱动的监控时使用。
development
azure-ai-language-conversations-py
microsoft
使用azure-ai-language-conversations Python SDK实现对话语言理解(CLU)。当使用ConversationAnalysisClient分析对话意图和实体、构建NLP功能或将语言理解集成到应用程序中时使用。
development
azure-ai-ml-py
microsoft
Azure Machine Learning SDK v2 for Python。用于机器学习工作区、作业、模型、数据集、计算资源和管道。 触发词:“azure-ai-ml”、“MLClient”、“工作区”、“模型注册表”、“训练作业”、“数据集”。
development