互联网说这是一个真正的内核错误，但没有人知道是什么触发？
运行内核2.6.27.24的CentOS x86_64的服务器
这是我的崩溃输出：
[root@u15345757 httpdocs]# Message from syslogd@ at Thu Aug 6 01:42:22 2009 ... u15345757 kernel: [1145736.506380] Eeek! page_mapcount(page) went negative! (-1) Message from syslogd@ at Thu Aug 6 01:42:22 2009 ... u15345757 kernel: [1145736.517515] page pfn = d0a3 Message from syslogd@ at Thu Aug 6 01:42:22 2009 ... u15345757 kernel: [1145736.523814] page->flags = 10000000000083c Message from syslogd@ at Thu Aug 6 01:42:22 2009 ... u15345757 kernel: [1145736.532489] page->count = 2 Message from syslogd@ at Thu Aug 6 01:42:22 2009 ... u15345757 kernel: [1145736.538741] page->mapping = ffff88001f01a110 Message from syslogd@ at Thu Aug 6 01:42:22 2009 ... u15345757 kernel: [1145736.547924] vma->vm_ops = 0x0 Message from syslogd@ at Thu Aug 6 01:42:22 2009 ... u15345757 kernel: [1145736.554543] [ cut here ] Message from syslogd@ at Thu Aug 6 01:42:23 2009 ... u15345757 kernel: [1145736.564528] invalid opcode: 0000 [1] SMP Message from syslogd@ at Thu Aug 6 01:42:23 2009 ... u15345757 kernel: [1145736.564528] Code: 80 e8 22 51 fd ff 48 8b 85 90 00 00 00 48 85 c0 74 19 48 8b 40 20 48 85 c0 74 10 48 8b 70 58 48 c7 c7 10 7f 7d 80 e8 fd 50 fd ff <0f> 0b eb fe 8b 77 18 41 58 5b 5d 83 e6 01 f7 de 83 c6 04 e9 df Broadcast message from root (pts/3) (Thu Aug 6 01:49:29 2009): The system is going down for reboot NOW! 这不是CentOS的标准内核，这意味着它可以是任何数量的问题。 如果你真的需要为CentOS运行这个特殊的非标准内核，我build议按照这个指南
首先要做的就是运行memcheck来放弃失败的内存。 否则，它可能是一个宇宙位翻转，或一个真正的内核错误。
那么这取决于你当时的工作stream程（是CPU饿了吗？FS饿了吗？等等）。一些工作stream程更可能触发错误的RAM。