Messages in this thread | | | Date | Wed, 23 Jun 1999 13:56:37 -0500 | From | Shaw Terwilliger <> | Subject | Re: SMP Question. |
| |
John Ronan wrote: > > Just a thought... what happens the kernel if a processor fails in a smp box?
Long ago, in 2.0.twenties I had a dual Pentium II 266 sitting in an un-air-conditioned room in July in 90 F weather. I was logged into the machine remotely when everything got REALLY slow... the shell was unresponsive, things like ps took 10 seconds. The load was 0.00.
I looked through syslog and messages to find an OOPS on CPU1. CPU0, though, was still chugging along. I'm not sure if the kernel started scheduling processes only on CPU0, or if new ones just went there, but the system was still running enough to be cleanly shut down.
I drove home from work, pulled the case off the machine, and a wave of heat just poured out. I rebooted with better ventilation (no case cover) and saw the BIOS had detected the core temperature of one of the chips to be near 120 F. They usually run near 90 F.
I've since never had a problem with the machine, since I've got 9 fans in it now, but I was still impressed how a kernel which wasn't supposed to be that good at SMP handled the situation.
-- Shaw Terwilliger
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |