一、组网图

不涉及

二、问题描述

存储版本为：V1.1.21T01P02；检查存储odsp.log日志发现存储端所有EP1均有过异常下上线的情况。

2019-03-18:17:11:09:0xb00001:SP1:DISK:Info:DSU-1:2:2's EP is offline.
2019-03-18:17:11:40:0xb00001:SP1:DISK:Info:DSU-1:2:1's EP is offline.
2019-03-18:17:11:41:0xb00001:SP1:DISK:Info:DSU-1:2:1's EP1 is online.
2019-03-18:17:11:54:0xb00001:SP1:DISK:Info:DSU-1:2:2's EP1 is online.
2019-03-18:17:12:07:0xb00001:SP1:DISK:Info:DSU-1:2:3's EP1 is online.

三、过程分析

检查存储messages日志确认现场链路出现过异常：

Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6334:phy[6] LINK_ERR_INVALID_DWORD
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 6, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:sas: phy6 not in a port.
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: 6346:phy[6] LINK_ERR_DISPARITY_ERROR
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 6, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:sas: phy6 not in a port.
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: 6359:phy[6] LINK_ERR_CODE_VIOLATION
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 6, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: phy6 not in a port.
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6334:phy[7] LINK_ERR_INVALID_DWORD
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 7, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6346:phy[7] LINK_ERR_DISPARITY_ERROR
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 7, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6359:phy[7] LINK_ERR_CODE_VIOLATION
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 7, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6248:PHY[7] DOWN
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_phy_event:received: 0, phy id: 7, event: 0.
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6248:PHY[5] DOWN
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_phy_event:received: 0, phy id: 5, event: 0.
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6248:PHY[6] DOWN
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_phy_event:received: 0, phy id: 6, event: 0.
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6334:phy[4] LINK_ERR_INVALID_DWORD
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 4, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6346:phy[4] LINK_ERR_DISPARITY_ERROR
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 4, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6359:phy[4] LINK_ERR_CODE_VIOLATION
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_port_event:broadcast received: 0, phy id: 4, event: 2
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:6248:PHY[4] DOWN
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: sas: notify_phy_event:received: 0, phy id: 4, event: 0.
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:5799: Last phy[4] Down and port[1] invalid
Mar 18 17:09:43 00:B3:42:0F:E1:3B kernel: pm80xx[1]:3727:task ffff81001875dcc0,tag 46,devid 650065,status 39,param 0,ib 0,ob 0

四、解决方法

1、如现场已恢复正常则持续观察。

2、如现场EP未正常上线，则需要交叉定位确认问题原因（SAS口或SAS线）。

五、风险提示

无

六、关键字

SAS，EP，异常下上线，LINK_ERR_DISPARITY_ERROR，LINK_ERR_INVALID_DWORD，LINK_ERR_CODE_VIOLATION