Nebius: InfiniBand connectivity issues for GPU clusters in eu-west2
Affected components: Compute CloudEU-WEST2 - Compute Cloud
Timeline of updates
- investigating
We are investigating InfiniBand connectivity issues affecting some GPU clusters in Compute Cloud in eu-west2. Distributed training jobs on affected clusters may experience degraded interconnect performance or failures. We will provide an update as soon as we have more information.
- identified
We have identified the cause of the InfiniBand connectivity issues. Our on-site team is working to resolve it. We will provide an update as soon as we have more information.
- monitoring
A fix has been implemented and we are monitoring the results.
- resolved
This incident has been resolved.
Archive proof
Rewindex commits each version of an incident it captures to a daily checkpoint, chains it to the day before, and has the checkpoint timestamped by FreeTSA and DigiCert. That proves this record existed by then and has not been changed since, by anyone, Rewindex included.
How the archive is checked, and what it cannot prove.
First version captured 17:03 UTC
In the 2026-10-19 checkpoint, timestamped by FreeTSA and DigiCert on Oct 20, 00:41 UTC. Record hash
b19a4ca7d1b9779a…
Impact, status and times are the vendor’s own, as last archived. View on stspg.io (opens in new tab).