今天跑去機房把一台有問題的伺服器重裝了。
回到家,檢查一下所有的機器,結果發現一台伺服器不停的列印錯誤資訊:
Message from syslogd@ at Thu Nov 13 21:51:21 2008 ...
www2 kernel: EDAC MC0: UE row 7, channel-a= 0 channel-b= 1 labels "-": (Branch=0 DRAM-Bank=2 RDWR=Read RAS=5074 CAS=0 FATAL Err=0x4)
Message from syslogd@ at Thu Nov 13 21:51:22 2008 ...
www2 kernel: EDAC MC0: UE row 0, channel-a= 0 channel-b= 1 labels "-": (Branch=0 DRAM-Bank=6 RDWR=Read RAS=4255 CAS=0 FATAL Err=0x4)
Message from syslogd@ at Thu Nov 13 21:51:23 2008 ...
www2 kernel: EDAC MC0: UE row 4, channel-a= 0 channel-b= 1 labels "-": (Branch=0 DRAM-Bank=4 RDWR=Read RAS=9592 CAS=0 FATAL Err=0x4)
Message from syslogd@ at Thu Nov 13 21:51:24 2008 ...
www2 kernel: EDAC MC0: UE row 6, channel-a= 0 channel-b= 1 labels "-": (Branch=0 DRAM-Bank=7 RDWR=Read RAS=15886 CAS=0 FATAL Err=0x4)
Message from syslogd@ at Thu Nov 13 21:51:25 2008 ...
www2 kernel: EDAC MC0: UE row 0, channel-a= 0 channel-b= 1 labels "-": (Branch=0 DRAM-Bank=5 RDWR=Read RAS=12555 CAS=0 FATAL Err=0x4)
Message from syslogd@ at Thu Nov 13 21:51:26 2008 ...
www2 kernel: EDAC MC0: UE row 6, channel-a= 0 channel-b= 1 labels "-": (Branch=0 DRAM-Bank=1 RDWR=Read RAS=14076 CAS=0 FATAL Err=0x4)
Message from syslogd@ at Thu Nov 13 21:51:27 2008 ...
www2 kernel: EDAC MC0: UE row 6, channel-a= 0 channel-b= 1 labels "-": (Branch=0 DRAM-Bank=5 RDWR=Write RAS=12783 CAS=0 FATAL Err=0x4)
以為記憶體出問題了,嚇了一跳。上網一查,原來可能是核心的問題
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=16528
關注一下