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: 1-71-2001180.1
Update Date:2017-07-06
Keywords:

Solution Type  Technical Instruction Sure

Solution  2001180.1 :   How to Replace an Infiniband Card in an Oracle Database Appliance X5-2 Server node  


Related Items
  • Oracle Database Appliance X5-2
  •  
Related Categories
  • PLA-Support>Sun Systems>Sun_Other>Sun Collections>SN-OTH: x64-CAP VCAP
  •  




In this Document
Goal
Solution
 1. Shut down the server node that needs the IB replaced
 2. Remove/Replace the IB card
 3.  Poweron/boot  the server node
References


Applies to:

Oracle Database Appliance X5-2 - Version All Versions to All Versions [Release All Releases]
x86_64
The IB card in a server node is only used to cable up to the other server node for the private interconnect.
The IB card is in slot 1 (slot closest to the power supplies).

Goal

 Describe the procedure needed to replace an IB card in a server node of an Oracle Database Appliance X5-2.

Solution

DISPATCH INSTRUCTIONS

- WHAT SKILLS DOES THE FIELD ENGINEER/ADMINISTRATOR NEED?

                Training and experience with Oracle Database Appliance systems

   TIME ESTIMATE:  50 minutes

   TASK COMPLEXITY:  3

FIELD ENGINEER/ADMINISTRATOR INSTRUCTIONS:


   PROBLEM OVERVIEW: There is only one IB card in each server node of the Oracle Databaes Appliance X5-2.  This document describes how to replace one of these cards, should it go bad.

 

WHAT STATE SHOULD THE SYSTEM BE IN TO BE READY TO PERFORM THE RESOLUTION ACTIVITY?:

   The OS must be shut down completely, and then powered off.  The server node will need to be pulled out far enough to open the front lid to in order to take out the IB card.

WHAT ACTION DOES THE FIELD ENGINEER/ADMINISTRATOR NEED TO TAKE?:

 

1. Shut down the server node that needs the IB replaced

The other node can remain up and running, you will only need to shut down the node that needs the IB card replaced.

Work with the customer on shutting down this node.

2. Remove/Replace the IB card

Reference links for Service Manual:

 The field engineer can now slide out the server for maintenance. Remember to disconnect the power cords before opening the top of the server.

Locate and Remove the PCIe card.

(a) There are three external PCIe slots in the system. The external PCIe slots are numbered 1, 2, and 3 from left to right when you
view the server from the rear. The Infiniband card is always installed in PCIe slot 1.

(b) Locate the Infiniband card in PCIe slot 1 and unplug the two cables from the PCIe card making note of their locations so that they
can be re-installed in the same configuration (label if needed).

(c) lift the green-tabbed latch on the rear of the server's chassis next to the PCIe slot to release the PCIe card's rear bracket.

(d) To release the riser from the motherboard connector, lift the green-tabbed lease
lever on the PCIe riser to the open position.

(e) Slide the plastic PCIe card retainer, which is mounted on the side of the chassis,
toward the front of the server to release the card(s) installed in the riser .

(f) Grasp the riser with both hands and remove it from the server.

(g) Remove the Infiniband card from the PCIe riser. Hold the riser in one hand and use your other hand to carefully pull the PCIe
card connector out of the riser.

(h) Disconnect the rear bracket that is attached to the Infiniband card from the rear of the PCIe riser.


Replace the PCIe card.

(a) Insert the rear bracket that is attached to the Infiniband card into the PCIe riser.

(c) Hold the riser in one hand and use your other hand to carefully insert the PCIe card connector into the Riser.

(d) Install the PCIe riser with the installed PCIe cards into the server.

(e) Raise the PCIe riser release lever (marked with a green tab) to the open (up) position
Making sure to replace the riser into the same position from which it was removed (PCIe slot 3), gently press the riser into the
motherboard connector until it seats and press the green-tabbed, riser release lever to the closed (down) position.

(f) Close the green-tabbed latch on the rear of the server's chassis next to the applicable PCIe slot to secure the PCIe card's rear
bracket to the server's chassis.

(g) Reconnect the Infiniband cables to the PCIe card that were unplugged during the removal procedure making sure to connect them in the
same configuration as when they were disconnected.

 

3.  Poweron/boot  the server node

 

OBTAIN CUSTOMER ACCEPTANCE

The system administrator should verify the system is functioning correctly. Some suggested actions they can take to verify are:


1. On the host that the card was replaced run:

# ibstat
CA 'mlx4_0'
        CA type: MT4100
        Number of ports: 2
        Firmware version: 2.11.1280
        Hardware version: 0
        Node GUID: 0x0010e000016759c8
        System image GUID: 0x0010e000016759cb
        Port 1:
                State: Active
                Physical state: LinkUp
                Rate: 40
                Base lid: 1
                LMC: 0
                SM lid: 1
                Capability mask: 0x02514868
                Port GUID: 0x0010e000656759c9
                Link layer: IB
        Port 2:
                State: Active
                Physical state: LinkUp
                Rate: 40
                Base lid: 4
                LMC: 0
                SM lid: 3
                Capability mask: 0x02514868
                Port GUID: 0x0010e000656759ca
                Link layer: IB

Ensure both Port 1 & Port2:
State is "Active"
Physical state: "LinkUp"
Rate: "40"

2. Run oakcli validate to verify that the ibbond0 is running.  

ibbond0 may have a different IP address than listed in this example.

# oakcli validate -c networkcomponents
INFO: Doing oak network checks
RESULT: Detected active link for interface eth0 with link speed 1000Mb/s and cable type as TwistedPair
WARNING: No Link detected for interface eth1 with cable type as TwistedPair
WARNING: No Link detected for interface eth2 with cable type as TwistedPair
WARNING: No Link detected for interface eth3 with cable type as TwistedPair
INFO: Checking bonding interface status
RESULT: No Bond Interface Found
SUCCESS: ibbond0 is running 192.168.16.27

 

3. Ping other node over the Infiniband subnet

 

PARTS NOTE: 7092757, 7046442

References

<BUG:15789419> - SUNBT7166073 REPLACED DISKS ASSIOCIATED WITH OLD JBOD SERIAL NUMBER
<NOTE:1920586.1> - Oracle ZFS Storage Appliance: DE2 Tray Serial Number Reprogramming
https://stbeehive.oracle.com/teamcollab/wiki/AmberRoadSupport:DE2-24C+and+DE2-24P+Serial+number+re-programming+procedure.

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