Azure Application Insights Formatter

Application Insights traces come out of the box in several shapes: structured JSON lines with numeric severityLevel, the traditional on-prem prefix (timestamp followed by a role), and Analytics query output. Azure Application Insights Formatter detects the shape, resolves numeric severities to names, and lines each trace up into time | severity | role | operation | message columns.

A severity histogram, roles breakdown, and per-operation volume complete the picture — enough to jump straight from paste to “The webapp started 4xxing orders at 14:00.”

• Shape autodetection — JSON, on-prem prefix, Analytics rows.
• Severity mapping — 0–4 to Verbose…Critical.
• Role/operation columns — parsed where present.
• Severity histogram — errors isolated.
• Copyable report.
• Local only — nothing uploaded.

1. Paste traces — JSON lines, stdout, or Analytics rows.
2. Format. Columns align, severities resolve.
3. Read the severity and role summaries.
4. Copy the report for the incident.

Example 1 — Incident replay. Structured traces decode numeric severities so severityLevel: 3 becomes an obvious Error spike.

Example 2 — Role attribution. Traces carrying cloud_RoleName show which service owned the failures.

Example 3 — Dependency impact. Operation names make it easy to correlate a failing endpoint with its callers.

• Severities decoded — numbers become names.
• Roles visible — service attribution.
• Operations grouped — endpoint impact.
• Shape-tolerant — JSON and stdout both.
• Private — all processing client-side.