Skip to main content

Infrastructure Status

Real-time operational health across every Yobitel region, platform service, and AI endpoint.

All Systems Operational

Last checked: April 8, 2026 at 14:32 UTC

GPU Cloud

US East (Virginia)
99.99% uptimeOperational
30 days agoToday
US West (Oregon)
99.98% uptimeOperational
30 days agoToday
EU (Frankfurt)
99.99% uptimeOperational
30 days agoToday
APAC (Singapore)
99.97% uptimeOperational
30 days agoToday

Platform Services

Kubernetes Control Plane
99.99% uptimeOperational
30 days agoToday
Object Storage (S3-Compatible)
99.999% uptimeOperational
30 days agoToday
API Gateway
99.99% uptimeOperational
30 days agoToday
Dashboard & Billing Portal
99.95% uptimeOperational
30 days agoToday

AI Services

Inference API
99.98% uptimeOperational
30 days agoToday
Fine-Tuning Pipeline
99.96% uptimeOperational
30 days agoToday
Marketplace
99.99% uptimeOperational
30 days agoToday
Model Registry
99.99% uptimeOperational
30 days agoToday
Operational Degraded Outage

Recent Incidents

A transparent record of every event affecting Yobitel infrastructure.

April 5, 2026ResolvedminorDuration: 22 minutes

Elevated Latency · Inference API (US East)

Between 09:12 and 09:34 UTC, the Inference API in US East experienced elevated p99 latency (avg 320ms vs normal 85ms). Root cause identified as a misconfigured autoscaler threshold during a rolling update. Autoscaler limits corrected and verified.

March 30, 2026ResolvedminorDuration: 12 minutes

Scheduled Maintenance · US West Network Upgrade

Planned 15-minute maintenance window for spine switch firmware upgrade in US West. Completed in 12 minutes with zero customer-facing impact thanks to redundant path failover.

March 18, 2026ResolvedminorDuration: 8 minutes

Elevated API Latency · EU Region

API Gateway in EU Frankfurt experienced degraded performance from 14:05 to 14:13 UTC. Root cause was a configuration rollout delay in the edge proxy layer. Traffic was automatically rerouted within 3 minutes; full resolution at 14:13 UTC.

March 2, 2026ResolvedmajorDuration: 50 minutes

Fine-Tuning Job Queue Backlog · APAC Region

Fine-tuning jobs submitted to APAC Singapore between 02:00 and 02:45 UTC experienced queuing delays of up to 15 minutes. Cause traced to an underprovisioned scheduler pod. Additional replicas deployed and queue drained by 02:50 UTC.

Upcoming Maintenance

Scheduled maintenance windows with expected impact and duration.

April 12, 2026

03:00–03:30 UTC

EU Frankfurt · Network Fabric Upgrade

Rolling firmware update on ToR switches. Redundant paths ensure zero expected downtime.

GPU Cloud (EU Frankfurt)Object Storage
Expected duration: 30 minutes

April 20, 2026

06:00–06:45 UTC

Kubernetes Control Plane · Version Upgrade

Upgrading managed Kubernetes to v1.31. Workloads continue running; brief API server failover expected.

Kubernetes Control PlaneDashboard
Expected duration: 45 minutes

Subscribe to Status Updates

Get notified about incidents, maintenance windows, and resolutions delivered straight to your inbox.

We only send status-related notifications. Unsubscribe anytime.