Sun Microsystems, Inc.  Oracle System Handbook - ISO 7.0 May 2018 Internal/Partner Edition
   Home | Current Systems | Former STK Products | EOL Systems | Components | General Info | Search | Feedback

Asset ID: 1-72-2382098.1
Update Date:2018-04-09
Keywords:

Solution Type  Problem Resolution Sure

Solution  2382098.1 :   SPARC T8-1, T7-1: Faults on NVMe-SFF drives due to poorly seated cables  


Related Items
  • SPARC T7-1
  •  
  • SPARC T8-1
  •  
Related Categories
  • PLA-Support>Sun Systems>SPARC>CMT>SN-SPARC: T8
  •  




In this Document
Symptoms
Cause
Solution


Applies to:

SPARC T7-1 - Version All Versions and later
SPARC T8-1 - Version All Versions and later
Information in this document applies to any platform.

Symptoms

SPARC T8-1 and T7-1 with NVMe-SFF drives, may experience faults if cables not seated correctly.

This procedure is to make sure the T8-1 and T7-1 internal NVMe cable is fully connected / seated at the T8-1 and T7-1 Disk Backplane.  If it is not fully connected there may be link errors resulting in faults on the NVMe-SFF drives.  This procedure applies to two general cases:

 

1.  After servicing any item in the Motherboard Assembly this procedure should be followed as the final step before reinstalling the top cover. This applies to Customer and Oracle service actions that may cause the internal NVMe cable to be touched, including replacing or upgrading memory, replacing fans, changing a battery, adding an NVMe-switch card and NVMe-cable or replacing the Disk Backplane.

2.  This procedure should also be followed if a fault has occurred on the link between a NVMe SFF drive and the NVME switch card in slot 3.  This procedure should be done and system results observed before replacing the NVMe switch card.  Here is a list of typical faults that have been observed in a T8-1 that may be corrected by this procedure:

 

Fault fault.chassis.voltage.failing on FRU /SYS/DBP at component /SYS/DBP

Fault fault.io.pciex.bus-linkerr on FRU /SYS/DBP/NVME1 at component /SYS/DBP/NVME1
Fault fault.io.pciex.bus-linkerr on FRU /SYS/DBP/NVME0 at component /SYS/DBP/NVME0

Fault fault.io.pciex.bus-linkerr-deg on FRU /SYS/DBP/NVME1 at component /SYS/DBP/NVME1
Fault fault.io.pciex.device-interr-deg on FRU /SYS/DBP/NVME0 at path
             hc:///chassis=0/motherboard=0/nvme-bay=0/disk=0

 

 

 

 

 

Cause

NVMe cables not seated correctly.

Solution

Overall View of NVMe Cable:

 

 pic1
Picture 1 – information only: The clear plastic air baffle has been removed to show the NVME switch card
with the NVMe cable plugged into the connectors on the NVMe switch card.

 

pic2
Picture 2 – information only: The clear plastic air baffle has been removed and the fan cover lifted to show the
NVMe cable sitting in the plastic cable trough, going through the fan wall opening, passing between the two
middle fans, and splitting left to one Disk Backplane connector and right to a second Disk Backplane connector.

 

 

Procedure to make sure the T8-1 and T7-1 internal NVMe cable is fully
connected / seated at the T8-1 and T7-1 Disk Backplane:

 

pic3
Step 1 – After all other service items are completed, reinstall the clear plastic air baffle, or in the case of
attempting to correct a NVMe fault, remove top cover.

 

 pic4
Step 2 – Lift the fan cover to get access to the NVMe cable connectors on the disk backplane.

 

pic5
Step 3 – Push the left and right connectors of the NVMe cable into the disk backplane to make sure they are fully connected.

 

Step 4 – Lower / latch the fan cover.
Step 5 – Reinstall the top cover. 

 

 

 

 

 


Attachments
This solution has no attachment
  Copyright © 2018 Oracle, Inc.  All rights reserved.
 Feedback