InfiniBand connectivity issues for GPU clusters in eu-west2

Event timeline

How this incident unfolded, from the official status page.

1. Investigating

   20 days ago

   Nebius

   We are currently investigating this issue.

2. Investigating

   20 days ago

   Nebius

   We have identified the root cause and taken the necessary corrective actions to address the issue. We are monitoring the service to confirm recovery.

3. Monitoring

   20 days ago

   Nebius

   We have taken corrective actions to stabilize the affected infrastructure and are working to address the root cause. Recovery is still in progress, and we are closely monitoring service health. We will confirm mitigation once recovery and stability checks are complete.

4. Resolved

   20 days ago

   Nebius

   This incident has been resolved.