New B200 spot capacity is live in US East from $1.69 per GPU-hour. See availability
Partial outage

Stop and restart operations delayed in US West (Oregon)

Stop and restart operations were delayed by up to 53 minutes in US West (Oregon).

Reference 6a17e721dbb1 · US West (Oregon) · Updated just now

Status
ResolvedResolved — last update 25 Oct 2025, 08:01 UTC
Started
25 Oct, 07:08Saturday 25 October 2025 UTC
Duration
53 minEnded 25 Oct 2025, 08:01 UTC
Components
1Instance launch and lifecycle — US West (Oregon)
Timeline

What we posted, as we posted it.

Resolved4 updates · UTC
  1. ResolvedThis incident has been resolved. Minutes spent waiting for a stop were refunded.25 Oct, 08:01
  2. MonitoringOperations have been applying within seconds since 07:42 UTC. We are monitoring.25 Oct, 07:42
  3. IdentifiedThe lifecycle worker in US West (Oregon) was stuck on a node that stopped responding to shutdown commands. The node has been isolated and the queue is draining.25 Oct, 07:31
  4. InvestigatingStop and restart requests in US West (Oregon) are queued and taking several minutes to apply. Launches and running instances are not affected.25 Oct, 07:08

Affected components

Details

Type
Partial outage
Started
25 Oct 2025, 07:08 UTC
Ended
25 Oct 2025, 08:01 UTC
Region
US West (Oregon)
Impact
Stop and restart operations were delayed by up to 53 minutes in US West (Oregon).
Reference
6a17e721dbb1

Follow the Atom feed