Intertrace
System statusJSONRSSCalendarBadge

Intertrace

Partial degradation

Updated Nov 13, 12:44 AM (UTC) · 68.3% uptime since Aug 24, 2026 · 539ms probe

Auto-refreshes every 30 seconds. This page stays up independent of sign-in.

Active incidents

  • MonitoringDetected

    HIRO accuracy, Sign-in, Dashboard degraded

    Ongoing degradation on HIRO accuracy, Sign-in, Dashboard.

    Nov 12, 3:30 AM (UTC) · HIRO accuracy, Sign-in, Dashboard

Components

Control plane

  • DashboardOperational · 529ms · 97.6%

    platform.intertrace.ai — investigation, settings, and org workspace. Checked by loading the dashboard signed out.

  • Sign-inOperational · 333ms · 97.6%

    Clerk-backed sign-in on platform.intertrace.ai.

  • Dashboard dataOperational · 267ms · 95.1%

    Org-scoped telemetry store used by the dashboard.

  • Background jobsOperational · 298ms · 40.9%

    Scheduled work: Guardian reviews, session close, activity ledger, SOC SLA escalations, runtime findings, assessments and the autonomous security agents. Degraded when one has not completed within about twice its interval.

Data plane

  • Telemetry ingestOperational · 316ms · 95.1%

    Gateway events written into the investigation spine.

  • AI runtime gatewayOperational · 256ms · 98.8%

    intertrace.fly.dev — LLM, MCP, and authorize.

Detection

  • HIRO accuracyDegraded · 301ms · 66.7%

    Each night a sample of the day's HIRO answers is checked again against the sessions, incidents, findings and agents they cite, each as it was when HIRO answered. Degraded when more than 1% of its statements don't hold up or the last check is more than two days old; a night with nothing to check says "no recent sample".

  • Reporting accuracyOperational · 301ms · 66.7%

    Each day every organization's stored records are checked against facts that must hold for the dashboard's numbers to be true: a session's cost is the sum of its requests', the month's spend on the budget meter is billable spend, an allowed request never carries a block label, every decided approval names its decider, no action is both run and blocked, a held call is never counted as blocked, every cost is the price list's or marked estimated, and each MCP server key appears once. Degraded while one does not hold anywhere or the last check is more than two days old.

Each bar is one UTC day · left is 90 days ago · right is today · click a day for detail

Get incident updates

Confirm your email. We send mail only when Mission Control posts an incident or maintenance window — not on routine probe flaps.

Past incidents

  • ResolvedDetected

    Sign-in, Dashboard degraded

    Resolved after impact on Sign-in, Dashboard.

    Nov 11, 5:00 PM (UTC) → Nov 11, 5:15 PM (UTC) · Sign-in, Dashboard

  • ResolvedDetected

    Background jobs degraded

    Resolved after impact on Background jobs.

    Nov 10, 2:30 PM (UTC) → Nov 11, 2:15 AM (UTC) · Background jobs

  • ResolvedDetected

    Background jobs, Reporting accuracy degraded

    Resolved after impact on Background jobs, Reporting accuracy.

    Nov 5, 8:45 AM (UTC) → Nov 7, 1:45 AM (UTC) · Background jobs, Reporting accuracy

  • ResolvedDetected

    Background jobs, Dashboard data, Telemetry ingest, HIRO accuracy unavailable

    Resolved after impact on Background jobs, Dashboard data, Telemetry ingest, HIRO accuracy.

    Nov 2, 9:45 PM (UTC) → Nov 3, 10:45 PM (UTC) · Background jobs, Dashboard data, Telemetry ingest, HIRO accuracy

  • ResolvedDetected

    Background jobs degraded

    Resolved after impact on Background jobs.

    Nov 1, 4:45 AM (UTC) → Nov 2, 2:45 PM (UTC) · Background jobs

  • ResolvedDetected

    Background jobs degraded

    Resolved after impact on Background jobs.

    Oct 31, 12:00 PM (UTC) → Nov 1, 3:15 AM (UTC) · Background jobs

  • ResolvedDetected

    Background jobs degraded

    Resolved after impact on Background jobs.

    Oct 30, 4:15 PM (UTC) → Oct 30, 4:45 PM (UTC) · Background jobs

  • ResolvedDetected

    Background jobs degraded

    Resolved after impact on Background jobs.

    Oct 29, 3:45 PM (UTC) → Oct 30, 4:45 AM (UTC) · Background jobs

  • ResolvedDetected

    Dashboard data, Telemetry ingest, Background jobs, HIRO accuracy unavailable

    Resolved after impact on Dashboard data, Telemetry ingest, Background jobs, HIRO accuracy.

    Oct 28, 11:15 PM (UTC) → Oct 28, 11:30 PM (UTC) · Dashboard data, Telemetry ingest, Background jobs, HIRO accuracy

  • ResolvedDetected

    Background jobs, Telemetry ingest, HIRO accuracy, Dashboard data unavailable

    Resolved after impact on Background jobs, Telemetry ingest, HIRO accuracy, Dashboard data.

    Oct 23, 5:15 PM (UTC) → Oct 26, 3:30 AM (UTC) · Background jobs, Telemetry ingest, HIRO accuracy, Dashboard data

  • ResolvedDetected

    Dashboard data unavailable

    Resolved after impact on Dashboard data.

    Oct 12, 5:00 PM (UTC) → Oct 12, 5:15 PM (UTC) · Dashboard data

  • ResolvedDetected

    AI runtime gateway degraded

    Resolved after impact on AI runtime gateway.

    Oct 1, 2:00 AM (UTC) → Oct 1, 2:15 AM (UTC) · AI runtime gateway

What this page measures

  • Dashboard, sign-in page, and control-plane data store.
  • Telemetry ingest (gateway events).
  • AI runtime gateway at intertrace.fly.dev/healthz.

Internal schema checks and optional guardrail sidecars stay off this page. Email, Stripe, and customer connectors are not probed; those land as posted incidents. For a production issue, write support@intertrace.ai with your org ID and timestamps.