I'm having trouble defining an error I've found on an Intel S5000VSA motherboard based system. The last two nights it's hung although and the BMC has logged this:
---SNIP--- 57 7/12/12-12:04:41 AM Processor Processor 1 Stat (#0x90) CRITICAL event: Processor 1 Stat reports CATERR has occurred. BMC - LUN#0 (Channel#0 58 7/12/12-12:05:57 AM OEM Reserved SMI Timeout (#0x85) CRITICAL event: SMI Timeout reports it has been asserted. BMC - LUN#0 (Channel#0 59 7/12/12-8:11:01 AM OEM Reserved SMI Timeout (#0x85) Informational event: SMI Timeout reports it has been deasserted. BMC - LUN#0 (Channel#0 60 7/12/12-8:11:01 AM Power Unit Power Unit (#0x01) Informational event: Power Unit reports the power unit is powered off or being powered down. BMC - LUN#0 (Channel#0 ---SNIP---
The 8:11:01 AM events are obviously the client rebooting the server - just included to show that the system was hung completely and required a reboot.
The first two errors just after Midnight are the ones I am not sure about. CATERR == Catastrophic Error, but is that the processor itself claiming catastrophic error (processor is dead), or that it encountered data it couldn't recover from (bad checksum of some sort). The SMI Timeout appears to be the physical processor failing to return control of the processor to the OS..?
The room this unit is in gets pretty hot over night, but because the unit isn't throwing any thermal errors, I hesitate to act as though this a simple thermal shutdown error.
Thanks for any constructive input and "have you considered this?" that anyone might be able to offer.