outages.live
Live
NebiusIncident · 2026-10-03 02:30 UTC

InfiniBand connectivity issues for GPU clusters in eu-west2

ResolvedOfficial update

This incident has been resolved.

The latest update from Nebius.

State

Resolved

Started

30 d ago

2026-10-03 02:30 UTC

Lasted

6 h 28 min

until 2026-10-03 08:59 UTC

Worst impact

Degraded

Timeline

4 updates
02:3004:0705:4407:2208:59
DegradedUpdate postedTimes in UTC

Investigating

2 h 15 min 35%

Identified

2 h 59 min 46%

Monitoring

1 h 15 min 19%

Updates

Times in UTC

As posted by Nebius

Resolved2026-10-03 08:59 UTC

This incident has been resolved.

Monitoring2026-10-03 07:44 UTC

A fix has been implemented and we are monitoring the results.

Identified2026-10-03 04:45 UTC

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.

Investigating2026-10-03 02:30 UTC

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.

Affected components

2
Compute CloudEU-WEST2 - Compute Cloud

Nebius in the last 90 days

Full history
90 days agoToday

Incidents in 90 days

27

Days with an incident

27

Median duration

1 h 18 min

This incident

5.0× the median

How long each incident lasted, on the day it started. The outlined bar is this incident.

24 h 33 min
median 1 h 18 min
90 days agoToday

Other Nebius incidents

All incidents

Nebius·Token Factory: some dedicated endpoints partially unavailable

This incident has been resolved.

Resolvedresolved 2 d ago

Nebius·Compute Cloud: disk images and snapshots partial unavailability

The incident has been resolved

Resolvedresolved 4 d ago

Nebius·Token Factory 503 errors in eu-west2

The issue has been resolved.

Resolvedresolved 8 d ago

Nebius·Partial IAM and console unavailability

The console and IAM were unavailable for approximately 10 minutes. The incident has been mitigated, and service availability has been restored.

Resolvedresolved 14 d ago

Nebius·CLI login issues

This incident has been resolved.

Resolvedresolved 16 d ago

Follow Nebius

Your coding agent knows when Nebius is down, and tells you when a failure is theirs, not yours. You can also get Nebius incidents in a feed reader or in Slack.

Prompt
Install the outages.live MCP server, so that when something breaks you can tell if the cause is a provider we depend on and not my code. It reads the official status pages of cloud, AI, payment and developer providers and gives you their incidents as MCP tools.

Add it as a remote MCP server named "outages" at https://outages.cc/mcp (Streamable HTTP, no authentication), with your CLI command or your MCP config file.

Then:
1. Make sure it connects and shows the tools get_provider, list_feed and send_feedback.
2. Call get_provider with "nebius" and tell me if Nebius has a problem right now.
3. From now on, when a call to a third-party service fails or becomes slow, call get_provider for that service before you change my code.

Paste this into any coding agent. It adds the server with its own MCP setup.

RSS
https://outages.cc/feed.xml?providers=nebius
Slack
/feed subscribe https://outages.cc/feed.xml?providers=nebius

FAQ

Go from tracking outages to fixing production issues with Polylane.

Try Polylane