IPMI/KVM console unreachable for part of ZRH

时间线

1. 已解决2024年8月21日 09:12 UTC

   Out-of-band access in rows B–C has been stable since 09:12 UTC. Duration: 32 minutes. Nothing on the production side was affected; this only mattered to customers who were trying to reach a console during that window.

2. 观察中2024年8月21日 09:04 UTC

   All consoles respond. Watching the switch for a further reset.

3. 已确定原因2024年8月21日 08:49 UTC

   The management switch for those rows rebooted. It is back and its ports are coming up; consoles will return as BMCs re-register.

4. 调查中2024年8月21日 08:45 UTC

   The out-of-band consoles for machines in rows B–C of ZRH are unreachable since 08:40 UTC. The machines themselves are up and their production network is unaffected.

事故复盘会在解决后 5 天内发布于此。