When a CX-Series array tries to boot, SP-A first looks for a good image on drive 0-0, its primary OS drive. When monitoring the boot process from HyperTerminal, you will see messages like the following when a serial cable is inserted into the SP:
DDBS: MDDE (Rev 2) on disk 0 DDBS: MDDE (Rev 2) on disk 2DDBS: MDB read from both disks. DDBS: Chassis and disk WWN seeds match. DDBS: First (primary) disk needs rebuild. <=== Indicates the image is bad. DDBS: Second disk is valid for boot. <=== Indicates the image is good. NT FLARE image (0x00400007) located at sector LBA 0x0002284B
When the array fails to find a good image on drive 0-0, it fails the boot process. If you were to pull drive 0-0, the array will look to secondary drive 0-2 for a good image. If it finds a good image on that drive, it will boot up. In this example, the image on the secondary drive is valid for the array boots. If a valid image is not found on either drive, the boot-up fails and the drives must be re-imaged.
SP- A uses 0-0 and 0-2 as primary and secondary boot drives in that order.
SP-B uses 0-1 and 0-3 as primary and secondary boot drives in that order.
Roger_Wu
4K 消息
0
2017年1月4日 19:00
Bootlog中可以看到如下提示：
DDBS: First (primary) disk needs rebuild.
DDBS: Second disk is valid for boot.
根据KB: Interpreting DDBS messages when a CLARiiON CX-Series array boots https://support.emc.com/kb/338122，解释如下：
DDBS: MDDE (Rev 2) on disk 0
DDBS: MDDE (Rev 2) on disk 2DDBS: MDB read from both disks.
DDBS: Chassis and disk WWN seeds match.
DDBS: First (primary) disk needs rebuild. <=== Indicates the image is bad.
DDBS: Second disk is valid for boot. <=== Indicates the image is good.
NT FLARE image (0x00400007) located at sector LBA 0x0002284B
When the array fails to find a good image on drive 0-0, it fails the boot process. If you were to pull drive 0-0, the array will look to secondary drive 0-2 for a good image. If it finds a good image on that drive, it will boot up. In this example, the image on the secondary drive is valid for the array boots. If a valid image is not found on either drive, the boot-up fails and the drives must be re-imaged.
SP- A uses 0-0 and 0-2 as primary and secondary boot drives in that order.
SP-B uses 0-1 and 0-3 as primary and secondary boot drives in that order.
“蓝灯慢闪”是1/4 Hz频率的话，症状也符合KB: VNX How can we troubleshoot a faulted SP ? https://support.emc.com/kb/464621 中的描述：
另外是日志收集不完整吗？我跑完TRiiAGE后，TRiiAGE_SPlogs.txt是空的，查不到之前为何换SPB的原因，如果有其他时间收集的日志也可以传上来看看。
whchenlong
30 消息
0
2017年1月4日 04:00
Bootlog和sp collect上传到百度云了，
http://pan.baidu.com/s/1pKKM69P
whchenlong
30 消息
0
2017年1月4日 19:00
您好，我这边Second disk 是有效的，应该不需要re-image吧
whchenlong
30 消息
0
2017年1月4日 19:00
我也是怎么想的 ，之前碰到过类似问题，disabled spa write cache，然后单独启动spb，正常后插入spA，准备试试
Severity : Warning
System : FCN00114300207
Domain : Local
Created : Jan 5, 2017 6:37:42 PM
Message : Valid SP-Cache is for Storage Processor SP B.
Full Description : Valid SP-Cache is for Storage Processor SP B.
Recommended Action : Valid SP-Cache is for Storage Processor SP B.
Event Code : 0x7241
zhangjia
73 消息
0
2017年1月4日 19:00
停业务，把SP A 拔出来，单独启动SP B,起来后把SP A插进去
Roger_Wu
4K 消息
0
2017年1月4日 21:00
照着KB做，拔出primary disk，能启动的话，应该还能凑合用。
whchenlong
30 消息
0
2017年1月5日 17:00
好的，下次过去试试拔出primary disk
Roger_Wu
4K 消息
0
2017年1月5日 18:00
关键数据备份下，毕竟是系统盘，真要re-image挺花时间的。