AFF A250、AFF C250 或 FAS500f panic：看门狗 NMI，因为 IPMI 接口拥塞

最后更新
另存为PDF

Views:: 31

Visibility:: Public

Votes:: 0

Category:: aff-series

Specialty:: hw

Last Updated:

适用于

AFF A250，AFF C250
ASA A250、ASA C250
FAS500f
BMC 15.11 或更早版本

问题描述

节点死机重启：

PANIC: watchdog nmi because IPMI interface congested. in process idle: cpu9

BMC 事件日志或 SP-LATEST-SYSTEM-EVENT-LOG 指示看门狗中断，后跟多个总线可更正错误：

292 | 12/02/2022 | 17:14:14 | Watchdog 2 #0x0f | Timer interrupt | Asserted

293 | 12/02/2022 | 17:14:16 | Watchdog 2 #0x0f | Hard reset | Asserted

294 | 12/02/2022 | 17:14:17 | Unknown #0x51 | State Asserted

2a2 | 12/02/2022 | 17:15:00 | Critical Interrupt #0x31 | Bus Correctable error | Asserted

2a3 | 12/02/2022 | 17:15:00 | Critical Interrupt #0x31 | Bus Correctable error | Asserted

2a4 | 12/02/2022 | 17:15:00 | Critical Interrupt #0x31 | Bus Correctable error | Asserted

SSRAM 日志报告NMIsource(WdogBMCFail)
紧急重新启动后，节点可能会在 Cluster::>event log show 报告温度错误

[monitor.temp.unreadable:error]: The controller temperature (HIC2 Temp0) is not readable. [monitor.temp.unreadable:error]: The controller temperature (HIC2 Temp1) is not readable. [callhome.chassis.hitemp:error]: Call home for CHASSIS OVER TEMPERATURE