Hi, I have a quad processor Opteron machine with SLES9 installed and that reboots about once every hour after cron.hourly is run. Looking in corn.hourly I see that "mcelog" is the only thing run here. Looking in /var/log/mcelog I get messages like the following: MCE 0 CPU 0 BANK 4 TSC 3cea13080d266 ADDR a98b0538 STATUS 9474c00100000813 MCGSTATUS 0 MCE 0 CPU 2 BANK 4 TSC 70a3e89bae5c6 ADDR 523748358 STATUS 946cc00100000813 MCGSTATUS 0 MCE 1 CPU 2 BANK 4 TSC 70ae68df9e742 ADDR 523748358 STATUS 946cc00100000813 MCGSTATUS 0 MCE 2 CPU 2 BANK 4 TSC 70b8e925da516 ADDR 523748358 STATUS 946cc00100000813 MCGSTATUS 0 MCE 3 CPU 2 BANK 4 TSC 71026b0a9814b ADDR 523748358 STATUS 946cc00100000813 MCGSTATUS 0 Parsing the MCE messages gives me the following ./parsemce -e 62680aa4f40e6 -a 523748340 -b 0 -s 946cc00000000833 Status: (62680aa4f40e6) Machine Check in progress. Error IP valid Restart IP invalid. parsebank(0): 946cc00000000833 @ 523748340 External tag parity error Correctable ECC error Address in addr register valid Error enabled in control register Bus and interconnect error Participation: Local processor originated request Timeout: Request did not timeout Request: Generic error Transaction type : Instruction Memory/IO : Other The machine is based on a Tyan 4882 MB with BIOS 1.03. It has 4 850 CPUs and 32 GB of Crucial DDR400 memory (2GB modules). What do you think is the problem? Memory? CPU? BIOS? /peter