Support

Incidents

Active and recent incidents across the platform. All SEV-1/2 events include post-mortems published within 5 business days.

Active incidents
1
MTTA (30d)
3m 12s
MTTR (30d)
42m
Post-mortems posted
8/8
Recent incidents
IDTitleSeverityStartedDurationImpactedIncident leadState
INC-2026-0428Frankfurt inference latency p99 elevatedSEV-210:42 UTC1h 32mEU tenants · 12James ParkMitigated
INC-2026-0427PACS sync delay · St Mary's (HL7 backlog)SEV-308:14 UTC3h 08m1 tenantMaya SinghResolved
INC-2026-0426BrainSeg-MR v3.0.0 rollback (precision regression)SEV-2May 22 18:2242mAll MR tenantsDr. R. KapoorResolved
INC-2026-0425Login outage · Okta IdP error 503SEV-1May 18 14:0818mAll tenantsOn-callResolved
INC-2026-0424DICOM router OOM · sg-router-02SEV-3May 14 02:111h 04mAPACDiego AlvarezResolved
Active incident · INC-2026-0428
Timeline
  1. 10:42 UTC
    Alert · p99 inference latency > 1200ms (Frankfurt)
  2. 10:44 UTC
    Acked by on-call J. Park · paging SRE
  3. 10:51 UTC
    Identified · GPU node a100-fra-07 thermal throttle
  4. 11:02 UTC
    Mitigation · drained node, rerouted traffic
  5. 11:24 UTC
    Latency restored to baseline · monitoring
  6. 12:14 UTC
    Status: mitigated · root cause in progress
StatusMitigated
Customers notified12 / 12
Status pageUpdated
Slack war room#inc-0428
Post-mortem dueMay 28