问题描述
使用华为Avago SAS3508型号RAID控制器或RAID控制器芯片是 Broadcom / LSI MegaRAID 的云平台,在运行过程中,系统messages/dmesg日志提示 tag#20 BRCM Debug mfi stat 0x2d, data len requested/completed 0x20000/0x0 和 I/O error 报错。
[ 37.432714] print_req_error: I/O error, dev sdf, sector 0 flags 0
[ 37.443770] sd 4:0:7:0: [sdg] tag#14 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 37.457104] sd 4:0:7:0: [sdg] tag#14 Sense Key : Aborted Command [current]
[ 37.469015] sd 4:0:7:0: [sdg] tag#14 Add. Sense: No additional sense information
[ 37.481288] sd 4:0:7:0: [sdg] tag#14 CDB: Read(10) 28 00 00 00 00 00 00 01 00 00
[ 37.493537] print_req_error: I/O error, dev sdg, sector 0 flags 0
[ 37.504511] sd 4:0:10:0: [sdh] tag#15 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 37.517905] sd 4:0:10:0: [sdh] tag#15 Sense Key : Aborted Command [current]
[ 37.529836] sd 4:0:10:0: [sdh] tag#15 Add. Sense: No additional sense information
[ 37.542237] sd 4:0:10:0: [sdh] tag#15 CDB: Read(16) 88 00 00 00 00 00 00 00 00 00 00 00 01 00 00 00
[ 37.561009] print_req_error: I/O error, dev sdh, sector 0 flags 0
[ 37.676043] sd 4:0:0:0: [sda] tag#20 BRCM Debug mfi stat 0x2d, data len requested/completed 0x20000/0x0
[ 37.695313] sd 4:0:1:0: [sdb] tag#21 BRCM Debug mfi stat 0x2d, data len requested/completed 0x20000/0x0
[ 37.714839] sd 4:0:2:0: [sdc] tag#22 BRCM Debug mfi stat 0x2d, data len requested/completed 0x20000/0x0
[ 37.734640] sd 4:0:4:0: [sdd] tag#23 BRCM Debug mfi stat 0x2d, data len requested/completed 0x20000/0x0
[ 37.754594] sd 4:0:5:0: [sde] tag#24 BRCM Debug mfi stat 0x2d, data len requested/completed 0x20000/0x0
[ 37.774823] sd 4:0:6:0: [sdf] tag#25 BRCM Debug mfi stat 0x2d, data len requested/completed 0x20000/0x0
问题原因
V6.1.1及其之后版本的云平台,已升级MegaRAID驱动版本,升级后版本为 07.719.04.00 。当云平台部署环境中的RAID卡固件版本过低时,则可能导致此问题。
解决方案
当服务器使用华为Avago SAS3508型号的RAID控制器时,请将RAID卡固件升级至 FW: 5.140.00-3515 版本。
当服务器未使用华为Avago SAS3508型号的RAID控制器,但使用的RAID控制器芯片是 Broadcom / LSI MegaRAID 时,请将RAID卡固件升级至2021年8月前的最新版本。
说明:
对于上述版本RAID卡固件的获取,请通过对应官方网站获取,或直接联系硬件厂商获取。