locked
Instability with C-state enabled RRS feed

  • Question

  • If I had to guess, I'd assume this is a hardware issue, but thought I'd post here to see if anyone else has this issue, or has any ideas. I've built a new PC for WHS, and I've had problems with it spontaneously rebooting. It would only do it when idle, or very low loads. After playing around, it appears that it only happens if I have C-states enabled in the bios. Obviously, that could be the solution, but I'd prefer to use c-states (the PC at idle uses 57 watts without, 44 watts with c-states enabled, and that's before another 12w savings when the drives spin down). This is built on a Supermicro C2SEA on the latest bios with a e8400 (no under/over clocking or volting, the bios doesn't support it). I THOUGHT I had it fixed after changing the bios from ACPI v. 2.0 to v. 3.0 ... it seemed stable, I left it sitting idle all day with no problem. But after rebooting it, it was back to the same issue.

    Server 2003 supports c-states, and the reboots aren't completely predictable (if I keep it busy it won't reboot, or if I do things then let it go idle, when it reboots seems less predictable. But if I just boot it, and log in, then do nothing else, it's fairly likely that it reboot approximately five minutes later.

    Any ideas?

    Thanks!
    Wednesday, September 16, 2009 3:19 PM

All replies

  • If I had to guess, I'd assume this is a hardware issue, but thought I'd post here to see if anyone else has this issue, or has any ideas. I've built a new PC for WHS, and I've had problems with it spontaneously rebooting. It would only do it when idle, or very low loads. After playing around, it appears that it only happens if I have C-states enabled in the bios. Obviously, that could be the solution, but I'd prefer to use c-states (the PC at idle uses 57 watts without, 44 watts with c-states enabled, and that's before another 12w savings when the drives spin down). This is built on a Supermicro C2SEA on the latest bios with a e8400 (no under/over clocking or volting, the bios doesn't support it). I THOUGHT I had it fixed after changing the bios from ACPI v. 2.0 to v. 3.0 ... it seemed stable, I left it sitting idle all day with no problem. But after rebooting it, it was back to the same issue.

    Server 2003 supports c-states, and the reboots aren't completely predictable (if I keep it busy it won't reboot, or if I do things then let it go idle, when it reboots seems less predictable. But if I just boot it, and log in, then do nothing else, it's fairly likely that it reboot approximately five minutes later.

    Any ideas?

    Thanks!
    My suggestion would be to disaable C-states for an extended period of time (at least a week) and see if it occurs.  If not, you have your answer.  :)
    Thursday, September 17, 2009 2:57 AM
    Moderator