Cooling issue in US-EAST-15A — potential impact to compute, storage, and networking
CoreWeave reports a cooling issue in the US-EAST-15A availability zone that affected compute, storage, and networking. CoreWeave reports the incident resolved after 17 hours and 39 minutes.
State
Resolved
Started
Lasted
19 h 39 min
Worst impact
Partial outage
Timeline
8 updatesInvestigating
19 h 39 min 100%
Updates
Times in UTCAs posted by CoreWeave
When we believe the issue has been mitigated and we’re still actively assessing the full impact and ensuring it doesn’t reoccur. All affected services, compute, storage, and object storage are back online, and the large majority of nodes have been returned to production. Customers in US-EAST-15A can resume normal workloads. Our teams are still completing final verification of any remaining nodes and are continuing to monitor environmental stability. We apologise for any inconvenience this issue may have caused
Cooling has been fully restored at the US-EAST-15A data center and data-hall temperatures have returned to normal operating levels. All affected services, compute, storage, and object storage are back online, and the large majority of nodes have been returned to production. Customers in US-EAST-15A can resume normal workloads. Our teams are completing final verification of the remaining nodes and continuing to monitor environmental stability. We will post a final update once full recovery is confirmed.
Cooling at the US-EAST-15A facility has been restored and confirmed stable by the facility provider. Network connectivity and storage services have recovered, and core services are operating normally. We are now restoring the remaining compute capacity and monitoring the environment to confirm sustained stability. Some customers may continue to see individual nodes returning to service; we recommend validating workloads as capacity comes back online. We will provide a further update as recovery completes. For questions, contact support@coreweave.com
Cooling at the US-EAST-15A facility has been restored, and network connectivity has largely recovered. Storage services are back online. Our teams are now working on bringing the compute nodes back online. As the situation evolves, we will continue to provide our customers with updates on the matter. If you have any questions, we encourage you to reach out to support by email: support@coreweave.com
Cooling conditions in US-EAST-15A are improving, and our teams are actively working to restore affected infrastructure. We do not yet have a confirmed restoration ETA and will provide further updates as recovery progresses. As the situation evolves, we will continue to provide our customers with updates on the matter. If you have any questions, we encourage you to reach out to support by email: support@coreweave.com
We have identified the cause as a cooling system failure at the facility serving part of the US-EAST-15A availability zone. Customers may continue to experience storage disruption, and some compute capacity remains powered down to reduce heat. As a result, additional compute, storage, and networking resources may experience interruptions. We are working with the facility provider on restoration. As the situation evolves, we will continue to provide our customers with updates on the matter. If you have any questions, we encourage you to reach out to support by email: support@coreweave.com
We are investigating a cooling system issue affecting part of the US-EAST-15A availability zone. We have received reports of storage disruption and have powered down some compute capacity to reduce heat. Additional compute, storage, and networking resources may experience interruptions. Our teams are working with the facility provider to restore cooling and determine the full scope of customer impact. As the situation evolves, we will continue to provide our customers with updates on the matter. If you have any questions, we encourage you to reach out to support by email: support@coreweave.com
Affected components
10CoreWeave in the last 90 days
Full historyIncidents in 90 days
6
Days with an incident
26
Median duration
4 days
This incident
80% shorter than the median
Other CoreWeave incidents
All incidents
CoreWeave·Serverless Inference unavailable globally
CoreWeave Serverless Inference is unavailable globally, including the RNO2 region. CoreWeave reports that the incident is resolved. The outage lasted 1 hour and 40 minutes.

CoreWeave·Storage Incident
CoreWeave Object Storage in US-EAST-04 was degraded. Some customers saw HTTP 500 errors or timeouts on object storage requests, especially uploads. CoreWeave reports that the incident is resolved. It lasted about 5 hours and 54 minutes.

CoreWeave·Storage Service Degradation
CoreWeave Premium Shared Filesystem Volumes in US-EAST-13 do not fully work. CoreWeave reports that the incident is resolved. The outage lasted from 5 November 2026 to 12 November 2026.

CoreWeave·Object Storage Service Degradation in US-EAST-04
CoreWeave Object Storage in US-EAST-04 had a partial outage. Some customers saw intermittent request failures. CoreWeave reports that the issue is resolved, and it lasted from November 2 to November 12, 2026.

CoreWeave·Facility power incident impacting services in US-EAST-15A
A facility power incident affected CoreWeave Infrastructure, Object Storage, Services, and Inference in the US-EAST-15 region. CoreWeave reports that the incident is resolved. It lasted from 22 October 2026 to 3 November 2026.
Follow CoreWeave
Your coding agent knows when CoreWeave is down, and tells you when a failure is theirs, not yours. You can also get CoreWeave incidents in a feed reader or in Slack.
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.live/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 "coreweave" and tell me if CoreWeave 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.
https://outages.live/feed.xml?providers=coreweave/feed subscribe https://outages.live/feed.xml?providers=coreweaveFAQ
Go from tracking outages to fixing production issues with Polylane.
Try Polylane