Scheduled Network Maintenance: Switch Firmware Update (Multiple Availability Zones)
CoreWeave · 最近保存的提供方状态：scheduled
官方来源文字以原始语言显示。
最近保存的事件状态不是当前服务状态。来源列表可能不完整，事件从来源消失不确认恢复。请打开当前报告或官方来源查看较新证据。
已保存事件详情
- 提供方状态
- scheduled
- 提供方影响程度
- Provider status 200
- 提供方记录创建时间
- 13 Oct 2026, 20:55:38 UTC
- 提供方记录更新时间
- 13 Oct 2026, 20:55:38 UTC
- 计划窗口
- 19 Oct 2026, 00:00:00 UTC → 23 Oct 2026, 00:00:00 UTC
提供方报告的受影响组件
- Coreweave Network
6938d01eab511c0601da0119
这些关联描述此事件的报告范围。不能确认当前组件可用性或已核实依赖关系。
记录创建时间未必是故障开始时间。未报告的时间保持不可用。我们不根据采集时间计算故障时长。
CoreWeave 公布的平台和地点组件、事件及当前或未来维护。提供方地点名称不能确认 GlobalStatus 的地区探测覆盖或客户部署的健康状态。
已保存修订中的提供方更新
最新优先。最多显示最近 20 份已保存内容修订中的 100 条不同更新。同一提供方时间的文字修订单独保留。
- Provider state 100
Reported components: Coreweave Network. Reported locations: US-CENTRAL-08A, US-CENTRAL-10A, US-EAST-04, US-EAST-12A, US-CENTRAL-09A. Details: CoreWeave will be performing network maintenance from Monday, October 19, 2026 at 12:00 UTC to Friday, October 23, 2026 at 00:00 UTC. Work runs during daytime hours each day in the window, 12:00 UTC to 00:00 UTC. We are applying a vendor-recommended firmware update to the boot drives of network switches in the following availability zones: CA-EAST-01A, EU-NORTH-04A, EU-NORTH-05A, EU-SOUTH-03B, EU-SOUTH-04A, US-CENTRAL-01A, US-CENTRAL-06A, US-CENTRAL-07A, US-CENTRAL-08A, US-CENTRAL-08B, US-CENTRAL-09A, US-CENTRAL-10A, US-CENTRAL-11A, US-EAST-04A, US-EAST-04B, US-EAST-08A, US-EAST-11A, US-EAST-12A, US-EAST-13A, US-EAST-14A, US-EAST-15A, US-WEST-01A, US-WEST-04A, US-WEST-10A This update resolves a known issue that can cause a switch to become unresponsive, and it improves overall network reliability. Impact: No impact is expected. The update is applied live, without switch reboots or node drains, and running workloads can continue without interruption. In rare cases, a switch may need a brief restart after the update. Where this affects backend (GPU-to-GPU) network switches, a small number of nodes may temporarily lose part of their backend network bandwidth until the restart completes, which may affect multi-node jobs running on those nodes. Engineers and on-site technicians will be monitoring throughout the window, and affected customers will be contacted directly. No customer action is required.
在 13 Oct 2026, 20:57:31 UTC 的已保存修订中观察到