8805RAID 卡 RAID5 做 xfs 本地文件系统，业务压力稍大时会偶现 IO 归零，有时还...

测试积点老人

现象：
业务过程中，RAID 的 IO 会偶现归零的情况
报错打印如下：

1. 2022-10-26T13:48:02.114399+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Host adapter abort request (11,0,0,0)
   
2. 2022-10-26T13:48:02.114510+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Timed out Command: 8a 00 00 00 00 00 07 71 a8 00 00 00 02 00 00 00
   
3. 2022-10-26T13:48:02.115882+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_fib_debug_print:FIB(17) = ffff9600a1d40cc0 : 35808a40 Command = 503 XferState = 830ad Wait Time = 63 Sec
   
4. 2022-10-26T13:48:02.118834+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Host adapter abort request (11,0,1,0)
   
5. 2022-10-26T13:48:02.118946+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Timed out Command: 8a 00 00 00 00 00 04 5e 9a 00 00 00 02 00 00 00
   
6. 2022-10-26T13:48:02.121773+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_fib_debug_print:FIB(260) = ffff9600a1d4c300 : 358840a0 Command = 503 XferState = 830ad Wait Time = 62 Sec
   
7. 2022-10-26T13:48:02.121884+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Host adapter abort request (11,0,1,0)
   
8. 2022-10-26T13:48:02.123269+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Timed out Command: 8a 00 00 00 00 00 04 5e 98 00 00 00 02 00 00 00
   
9. 2022-10-26T13:48:02.126204+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_fib_debug_print:FIB(259) = ffff9600a1d4c240 : 35883880 Command = 503 XferState = 830ad Wait Time = 62 Sec
   
10. 2022-10-26T13:48:02.126331+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Host adapter abort request (11,0,1,0)
    
11. 2022-10-26T13:48:02.127660+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Timed out Command: 8a 00 00 00 00 00 04 5e 96 00 00 00 02 00 00 00
    
12. 2022-10-26T13:48:02.130572+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_fib_debug_print:FIB(144) = ffff9600a1d46c00 : 35849220 Command = 503 XferState = 830ad Wait Time = 62 Sec
    
13. 2022-10-26T13:48:02.130710+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Host adapter abort request (11,0,1,0)
    
14. 2022-10-26T13:48:02.133508+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Timed out Command: 8a 00 00 00 00 00 06 6f d8 00 00 00 02 00 00 00
    
15. 2022-10-26T13:48:02.133642+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_fib_debug_print:FIB(4) = ffff9600a1d40300 : 358020a0 Command = 503 XferState = 830ad Wait Time = 62 Sec
    
16. 2022-10-26T13:48:02.136450+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Host adapter abort request (11,0,1,0)
    
17. 2022-10-26T13:48:02.136592+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_eh_abort:Timed out Command: 8a 00 00 00 00 00 1d f3 02 00 00 00 02 00 00 00
    
18. 2022-10-26T13:48:02.139355+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_fib_debug_print:FIB(139) = ffff9600a1d46840 : 35846980 Command = 503 XferState = 830ad Wait Time = 63 Sec
    
19. 2022-10-26T13:48:02.149197+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_dev_reset:Host device reset request (11,0,0,0)
    
20. 2022-10-26T13:48:02.149299+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_print_command_queue_states:outstanding cmnd: midlevel 0, lowlevel 0, error handler 0, firmware 0, kernel 0
    
21. 2022-10-26T13:48:02.149385+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_dev_reset:Host device reset request (11,0,1,0)
    
22. 2022-10-26T13:48:02.152169+08:00 node01 kernel: aacraid 0000:82:00.0: AAC1:aac_print_command_queue_states:outstanding cmnd: midlevel 0, lowlevel 0, error handler 0, firmware 0, kernel 0

复制代码

有没有见过的大佬，这是 RAID 卡坏了吗

bellas

等大神

jingzizx

测试一下读写