community postPlatform, DevOps & Performancecommunity - not Microsoftreviewed - checked by Opus

BC Telemetry Buddy  – When Your 12-Year-Old Accidentally Helps You Find a Problem

BC Telemetry Buddy is an MCP-based VS Code tool that lets you question Business Central telemetry in natural language instead of writing KQL. A casual demo question exposed a customer whose background queue codeunit ran slow full table scans on an unindexed queue table that lacked a retention policy. Diagnosis took about 15 minutes, and enabling retention plus adding an index removed the waste.

post waldo's blog

waldo (Eric Wauters) · 2025-12-24 · 991 words · about 5 min

"One question. One answer. Immediate diagnosis with lots of details!"

Key points

  • BC Telemetry Buddy uses AI over MCP so telemetry can be explored conversationally instead of through hand-written KQL.
  • It flagged a customer where the codeunit ISC Process Update Queue Meth caused most of the wasted processing time through full table scans on an unindexed, unpurged queue table.
  • Before the fix, about 19,946 slow queries ran each morning at 3.4 s on average; a month later only 18 ran in 7 days at 0.8 s, and Job Queue Dispatcher dropped below 25 minutes.
  • Investigations can be saved as markdown files in the workspace, which makes it easy to check later that a fix worked.

Quotes

  • “One question. One answer. Immediate diagnosis with lots of details!” (Demonstrates the core value proposition: BC Telemetry Buddy transforms hours of manual investigation into a single conversational query, dramatically reducing time investment for performance diagnostics.)

AL objects mentioned

As named in the post. A name that matches one object page by exact type and name links to it; the others stay as named.

  • codeunit “ISC Process Update Queue Meth”

Not found in BC28-30: codeunit “ISC Process Update Queue Meth”.

Context

  • Features: Natural language telemetry queries, Automated performance analysis, Investigation documentation, Table indexing recommendations, Retention policy management, CPU time analysis

Source: waldo’s blog, community blog. Summary, key points and quotes are derived (CONTENT-NOTICE.md); read the original for the full text.