Kontron S5500 SEL Troubleshooting - Manual (Page 92)

System Event Log Troubleshooting Guide for Intel ® S5500/S3420 series Server Boards Miscellaneous events
Revision 1.0 Intel order number G74211-001 83
Table 77: IPMI Watchdog Sensor Event Trigger Offset – Next Steps
Event Trigger Offset
Description Next Steps
Hex Description
00h
timer expired,
status only
Our server systems support a BMC watchdog timer, which can
check to see if the OS is still responsive. The timer is disabled by
default, and would have to be enabled manually. It then requires
an IPMI-aware utility in the operating system that will reset the
timer before it expires. If the timer does expire, the BMC can
take action if it is configured to do so: (reset, power down, power
cycle, or generate a critical interrupt) If this event is being logged it is because the BMC has been configured to check the
watchdog timer.
1. Make sure you have support for this in your OS (typically using a third party
IPMI-aware utility like ipmitool or ipmiutil along with the openipmi driver).
2. If this is the case, then it is likely your OS has hung, and you should investigate
OS event logs to determine what may have caused this.
01h hard reset
02h power down
03h power cycle
08h timer interrupt
13.2 SMI Timeout
SMI stands for system management interrupt and is an interrupt that gets generated so the processor can service server management events
(typically memory or PCI errors, or other forms of critical interrupts), in order to log them to the SEL. If this interrupt times out, the system is
frozen.
Table 78: SMI Timeout Sensor Typical Characteristics
Byte Field Description
11 Sensor Type F3h = SMI Timeout
12 Sensor Number 06h
13
Event Direction and
Event Type
[7] Event direction
0b = Assertion Event
1b = Deassertion Event
[6:0] Event Type = 03h ( ‘digital’ Discrete)
14 Event Data 1
[7:6] – 00b = Unspecified Event Data 2
[5:4] – 00b = Unspecified Event Data 3
[3:0] – Event Trigger Offset = 1 = State Asserted
15 Event Data 2 Not used.