- 论坛徽章:
- 0
|
原帖由 hunterdong 于 2006-7-6 22:04 发表
哈哈,用钥匙毁尸灭迹了(想试试能否启动)、
最后发现是系统板问题,把系统板抽暇来一块就好了。
然后更换。。。(据说如果是内存问题,工程师要16条一条一条试。。。开16次机箱)
Sun不把客户的时间当时 ...
不会一条一条去试的,内存可以通过messages或POST所报的代码来确定的
不要把sun工程师都想的那么傻。
下面举个例子:
SF4800 failed POST with ECC errors.
Now the system is up.
From the provided data we can't break the failure down to a single DIMM:
A.SC: /N0/SB4 reported ECC error
Jun 08 16:26:08 4800a2 Domain-A.SC:
Status: 0xc1038103
Syndrome from DX2: 0x00a180a1
Syndrome from DX3: 0x00a180a1
Read transaction for CPU A: DTrans: 0x103 (AID=16, SEQ#=3)
ECC Syndrome: 0x0a1
MTag Syndrome: 0x0
Read transaction from CPU C: DTrans: 0x103 (AID=16, SEQ#=3)
ECC Syndrome: 0x0a1
MTag Syndrome: 0x0
Serial Jun 08 16:26:16 4800a2 Platform.SC: /N0/SB4 reported first ECC error
Jun 08 16:26:16 4800a2 Platform.SC: Bad data read from a DIMM or cache controlled by /N0/SB4/P2
Syndrome=0x0A1
D23 located on Dram id 7
Dram id 7 is located on SB4/P2/B0 or B1/D0 (J15300 or J15301)
替换掉J15300和J15301,问题解决 |
|