C
Outage recap · Crusoe

Crusoe outage — May 18, 2026

On May 18, 2026, Crusoe had a reported incident on Crusoe's official status page. Below is what was reported — impact, timing, and the AI services the same upstream outage could affect.

VM Creation and Start Issues in us-southcentral1-a and eu-iceland1-a

minor
Started
May 18, 4:24 AM EDT
Resolved
May 18, 9:19 PM EDT
Duration
16h 55m

Crusoe's updates (4)

  1. identified · May 18, 4:24 AM EDT

    We are investigating an issue affecting the ability to create and start Virtual Machines in the us-southcentral1-a and eu-iceland1-a regions. Existing running workloads and resources remain fully operational and are not impacted. Our Engineering teams are actively working on recovery. We will provide an update when more information is available.

  2. identified · May 18, 7:27 AM EDT

    We have identified the underlying issue affecting the ability to create and start Virtual Machines in the us-southcentral1-a and eu-iceland1-a regions. Our Engineering teams are actively working on recovery. Some customers may still be experiencing impact while this work continues. Existing running workloads and resources remain fully operational and are not impacted. We will provide a further update as soon as more information becomes available.

  3. monitoring · May 18, 10:34 AM EDT

    The issue affecting the ability to create and start Virtual Machines in the us-southcentral1-a and eu-iceland1-a regions has been mitigated. VM creation and start operations are functioning normally. No customer impact has been observed as of 14:00 UTC. Our Engineering teams continue to monitor both regions closely and are working on a permanent fix to prevent recurrence. We apologize for any inconvenience this may have caused.

  4. resolved · May 18, 9:19 PM EDT

    VM creation and start operations in the us-southcentral1-a and eu-iceland1-a regions have been functioning normally for several hours. Our Engineering teams have addressed the underlying issue and a permanent fix is being prepared for deployment to prevent recurrence. Existing running workloads and resources remained fully operational throughout the incident. We apologize for any inconvenience this may have caused.

Updates are Crusoe's own words from its status page (status: resolved); we never add an inferred cause. The vendor's incident page ↗

AIDownCheck is not affiliated with Crusoe. This recap summarizes incidents published on Crusoe's official status page.