sys/ Status

All systems operational

Checked every 60 seconds from 9 locations · demo data, times in UTC

Scheduled maintenance: database failover drill in eu-central, . No downtime expected; ingestion may see up to 200 ms extra latency.

Current status

API base: api.forge.example

Componentus-eastVirginiaeu-centralFrankfurtap-southeastSingapore
REST APIOperationalOperationalOperational
Event ingestionOperationalOperationalOperational
Delivery workersOperationalOperationalOperational
DashboardOperationalOperationalOperational
Customer portalOperationalOperationalOperational

Uptime, last 90 days

Hover or focus a day for details. Use the arrow keys to move between days.

Uptime bars need JavaScript. Every component was above 99.9% over the last 90 days.

  • Operational
  • Degraded
  • Outage
  • Maintenance

Incident history

Every incident gets a public write-up within five business days.

  1. Degraded

    Delivery logs slow to load

    Dashboard · All regions

    1. Resolved ·

      Log queries are back to normal. A missing index after the v4.1.4 migration caused full scans on large projects.

    2. Identified ·

      Slow queries traced to the delivery log index. Rebuilding it now. Event delivery is not affected.

    3. Investigating ·

      The dashboard is slow to load delivery logs for some projects.

  2. Degraded

    Elevated delivery latency in ap-southeast

    Delivery workers, Event ingestion · ap-southeast

    1. Resolved ·

      Latency back to baseline. A worker pool failed to scale out during a traffic spike; we raised the minimum pool size.

    2. Monitoring ·

      Extra capacity added. p99 delivery latency is falling.

    3. Investigating ·

      Deliveries in ap-southeast are taking up to 40 seconds to start. No events have been lost.

  3. Outage

    API errors in us-east

    REST API, Event ingestion · us-east

    1. Resolved ·

      A bad configuration push to our load balancers returned 503s for 11 minutes. It was rolled back; 0.7% of requests failed and SDK retries recovered most of them. Postmortem published.

    2. Identified ·

      Rolling back a load balancer configuration change.

    3. Investigating ·

      Elevated 503 errors on api.forge.example in us-east.

  4. Maintenance

    Scheduled database maintenance

    Event ingestion · eu-central

    1. Completed ·

      Maintenance finished on time with no failed requests.

    2. In progress ·

      Failover to the new database primary started.

  5. Degraded

    Portal sign-in failures

    Customer portal · All regions

    1. Resolved ·

      Sign-in works again. An expired certificate on the portal’s identity service was renewed and alerting added.

    2. Investigating ·

      Some customers cannot sign in to the embedded portal.