【昇腾故障案例-安装部署】Atlas 800-3010 SUSE15 fio压测过程中进行 power cycle 操作，系统启动过程挂死

问题现象描述

硬件配置：

CH121 V5 V100R001C00SPC170：

BMC 3.30

BIOS 1.10

CPLD：014

Raid FW 2139

Driver 706

OS：SLES15

CH225 V5 V100R001C00SPC170

BMC 3.30

BIOS 1.10

CPLD：014

Raid FW 2139

Driver 706

OS：SLES15

问题现象：

fio性能测试非系统分区，进行 power cycle 时，SLES 15系统启动过程挂死。

串口日志如下：

[ 0.164265] BUG: unable to handle kernel paging request at ffffffffffffffff

[ 0.168000] IP: 0x661fbb7a

[ 0.168000] PGD 19feda067 P4D 19feda067 PUD 65010067 PMD 0

[ 0.168000] Oops: 0002 [#1] SMP PTI

[ 0.168000] Modules linked in:

[ 0.168000] Supported: Yes

[ 0.168000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.12.14-23-default #1 SLE15

[ 0.168000] Hardware name: Huawei CH121 V5/IT11SPCA0, BIOS 1.10 02/16/2019

[ 0.168000] task: ffffffff920124c0 task.stack: ffffffff92000000

[ 0.168000] RIP: 0010:0x661fbb7a

[ 0.168000] RSP: 0000:ffffffff92003d98 EFLAGS: 00010086

[ 0.168000] RAX: 0000000060000202 RBX: 000000006f8f0420 RCX: 000000006f8f0398

[ 0.168000] RDX: 0000000000000000 RSI: 0000000000000330 RDI: 00000000676ee110

[ 0.168000] RBP: 0000000060000202 R08: 0000000000000001 R09: 0000000000000000

[ 0.168000] R10: 0000000067724120 R11: 0000000000000001 R12: 0000000000000030

[ 0.168000] R13: 0000000000000206 R14: 000000006500a001 R15: 0000000000000001

[ 0.168000] FS: 0000000000000000(0000) GS:ffff9dc0f4a00000(0000) knlGS:0000000000000000

[ 0.168000] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033

[ 0.168000] CR2: ffffffffffffffff CR3: 000000019fee8000 CR4: 00000000000606b0

[ 0.168000] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000

[ 0.168000] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400

[ 0.168000] Call Trace:

[ 0.168000] ? efi_call+0x58/0x90

[ 0.168000] ? efi_enter_virtual_mode+0x43e/0x4ee

[ 0.168000] ? start_kernel+0x3c0/0x446

[ 0.168000] ? early_idt_handler_array+0x120/0x120

[ 0.168000] ? x86_64_start_kernel+0x12c/0x13b

[ 0.168000] ? secondary_startup_64+0xa5/0xb0

[ 0.168000] Code: Bad RIP value.

[ 0.168000] RIP: 0x661fbb7a RSP: ffffffff92003d98

[ 0.168000] CR2: ffffffffffffffff

[ 0.168000] ---[ end trace 63d848956a3727ef ]---

[ 0.168000] Kernel panic - not syncing: Attempted to kill the idle task!

[ 0.168000] ---[ end Kernel panic - not syncing: Att

关键过程、根本原因分析

根本原因分析：

内核 efi 模块已知问题，4.12.14-150.14.2.x86_64 内核版本已修复。

结论、解决方案及效果

解决方案：

下载内核patch：kernel-default-4.12.14-150.14.2.x86_64.rpm，升级解决问题。

经验总结、预防措施和规范建议

无

备注

内核patch下载链接：https://download.microfocus.com/Download?buildid=WJ6yA5l6uxk~