Incident Management

From detection to resolution, automatically

Mihari connects your monitors, telemetry, error tracking, and Git repositories to detect, diagnose, and resolve incidents automatically. Generate postmortems in one click.

Incident INC-3012 - Automated Resolution
DetectMonitor
CorrelateTelemetry + Errors
PatchGit MR
ResolveAuto-close
Telemetry

Error rate spike: +340%

Correlated with deployment v2.8.1

Error Tracking

NullReferenceException

UserService.cs:142 — first seen 3m ago

Auto Patch

MR #847 created on GitLab

Null check added, CI passing

Automatic Resolution

Resolve incidents before your team wakes up

When a monitor or heartbeat is linked to telemetry and error tracking sources, Mihari correlates signals to identify the root cause automatically. If the error clears or the deployment rolls back, the incident is resolved without human intervention.

  • Link monitors and heartbeats to telemetry sources (logs, metrics, traces)
  • Connect error tracking to automatically identify the faulty code path
  • Correlate deployment events with error spikes to pinpoint the root cause
  • Auto-resolve incidents when the root cause is remediated or rolled back
  • Reduce mean time to resolution (MTTR) by eliminating manual triage
Monitor: api-gateway-health
DOWN
Since 14:32 UTC

Linked Sources

Telemetry - Logs & Metrics

api-gateway / production

Error spike

Error Tracking

api-gateway / Sentry

New errors

GitHub Repository

acme/api-gateway

Connected
Root Cause Identified

Error spike correlates with deployment v2.8.1 (deployed 14:30 UTC). NullReferenceException in UserService.cs:142 matches the HTTP 503 pattern from telemetry logs.

Repository Connection

acme/api-gateway

main

Connected

acme/frontend-app

main

Connected
Merge Request Created Automatically
Open!847 fix: add null check in UserService

// UserService.cs:142

- var user = _repo.GetById(id);

- return user.Profile.Name;

+ var user = _repo.GetById(id);

+ if (user?.Profile == null) return null;

+ return user.Profile.Name;

CI passing
1 approval
Created 2m ago
Git Integration

Auto-patch incidents with commits and merge requests

Connect your GitHub, GitLab, or Bitbucket account, then link monitors and sources to a repository. When an incident is detected and the root cause identified, Mihari can automatically create a commit and a merge request to patch the issue.

  • Connect GitHub, GitLab, or Bitbucket accounts in one click
  • Link monitors and telemetry sources to specific repositories
  • Auto-generate fix commits based on error tracking root cause analysis
  • Create merge requests with CI pipeline validation before merge
  • Full audit trail linking incidents to commits, MRs, and deployments
GitHub
GitLab
Bitbucket
Postmortem

Learn from every incident with structured postmortems

After an incident is resolved, Mihari generates a comprehensive, blameless postmortem pre-filled with the timeline, root cause analysis, severity matrix, and action items. Export it or use our template to run your own post-incident review.

  • Blameless postmortems auto-generated from the incident timeline and telemetry data
  • Severity matrix with customer impact, time to identification, and time to resolution
  • Root cause analysis pre-filled using the 5 Whys technique
  • Short-term and long-term action items with assignees and deadlines
  • Export as Markdown, PDF, or share via Slack and Confluence
Postmortem - INC-3012
Auto-generated

Summary

On March 18, 2026 at 14:32 UTC, the API Gateway monitor detected HTTP 503 errors across US East and EU West regions. The incident was automatically correlated with a NullReferenceException in UserService.cs introduced in deployment v2.8.1.

Severity Assessment

Customer Impact

High

Severity

Critical

Time to Identification

2 min

Time to Resolution

23 min

Timeline

14:32

Monitor api-gateway-health detected HTTP 503 from US East, EU West

14:32

Incident INC-3012 created. Escalation policy "Production Critical" triggered.

14:33

Telemetry correlation: error rate +340% since deployment v2.8.1

14:34

Error tracking: NullReferenceException in UserService.cs:142

14:35

Auto-patch: MR !847 created on acme/api-gateway (null check fix)

14:48

MR !847 merged after CI validation and team approval

14:55

All regions reporting HTTP 200. Incident auto-resolved.

Root Cause Analysis

Deployment v2.8.1 introduced a code path in UserService.GetProfile() that did not handle null user profiles. The 5 Whys analysis traces the root cause to missing integration tests for the profile lookup flow.

Action Items

Add integration tests for UserService.GetProfile() null path
Enable mandatory null-safety checks in CI pipeline
Add deployment canary step for API gateway releases
Update runbook with UserService failure patterns
End-to-End Workflow

The complete incident lifecycle, automated

From the first alert to the final postmortem, every step is connected and intelligent.

1

Detection

Monitors and heartbeats detect anomalies in real-time across all your services and endpoints.

2

Correlation

Telemetry logs, metrics, and error tracking data are correlated to identify the root cause automatically.

3

Resolution

A fix commit and merge request are created on your linked Git repository. CI validates the patch before merge.

4

Postmortem

A blameless postmortem is generated with the complete timeline, root cause analysis, and actionable improvements.

Incidents resolved faster, insights preserved

Automated incident management reduces MTTR and ensures every outage leads to concrete improvements.

80%

Incidents auto-resolved

<5m

Avg. time to identify

100%

Postmortems documented

3x

Fewer recurring incidents

Frequently asked questions

Stop firefighting, start auto-resolving

Connect your monitors, telemetry, and Git repos. Let Mihari detect, patch, and document incidents for you.