| Greg Lopp on 30 Aug 2006 17:27:41 -0000 |
|
In the late 90s, I worked for a company that included as its regular mainenance the instructions to restart/reboot the equipment every week. As quality of the embedded control SW improved, a new bug was discovered: a system crash every 49 days. It was eventually traced to the system's millisecond timer, which rolled over from 429467296 (0xFFFFFFFF) to 0 on that 49th day. That is not your problem. Linux is using a 64bit counter under the hood.....wait, that's for the jiffies counter (which shouldn't get far past 0x48190800 in 14 days), but many kernels are replacing the jiffies thing with direct accesses to the hardware clock. Anybody know if FC4 is one of those kernels? Point me at the source? ....and there are no reports of similar problems by other FC4 users? Does "hwclock --show" suggest anything odd? On 8/30/06, Sean C. Sheridan
<scs@campusclients.com> wrote: Have you ever heard of anything so weird as this? ___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug
|
|