Oracle Database Appliance
 

PLA-Support>Eng Systems>Exadata/ODA/SSC>Oracle Database Appliance>DB: ODA_EST
 

In this Document

Purpose

Details

System Controller (SC) Failure

Storage Failures

Local Disk Failure

Shared Disk Failure

NIC Card Failures

Applies to:

Purpose

 This document covers various recovery scenarios on Oracle Database Appliance for hardware components failures.

Details

System Controller (SC) Failure

Storage Failures

NIC Card Failures

Memory Module Failures

Power Supply Unit Failures

System Controller (SC) Failure

System Controller is a Field Replaceable Unit (FRU). It must be replaced by Oracle-qualified service technician.
Once the SC has been replaced on a node, following steps need to be performed.

1. Configure the ILOM (via BIOS or Serial Port).
   
   
2. The eth0, eth1, eth2, eth3 interfaces reside on the SC itself. Replacing the SC will cause these interfaces to have new MAC addresses.  Launch the 'Remote Console' and edit the following files in the /etc/sysconfig/network-scripts directory to change the MAC address to the new MAC address.
   
   ifcfg-eth0
   ifcfg-eth1
   ifcfg-eth2
   ifcfg-eth3
   
   NOTE: See the "NIC Card Failures" section for an example.
   
   
3. Run the 'service network restart' command as root user.
   
   
4. If core key count was configured, run the 'oakcli apply core_config_key' command.

Storage Failures

Disks in an Oracle Database Appliance are hot-swappable and can be replaced by customers themselves if required.

Local Disk Failure

Oracle Database Appliance has two mirrored local disks that house the operating system and other installed software. If one of the local disks fails, replace the disk and follow the procedure outlined in the following document

Note 1382300.1 - ODA (Oracle Database Appliance) : How to replace FAILED SYSTEM BOOT DISK

If both local disks fail on a node, we need to do re-imaging (bare metal restore) and follow the add node procedures to bring the node back to the cluster. Refer to the following documents.

Note 1373599.1 - Oracle Database Appliance Bare Metal Restore Procedure.

Note 1469963.1 - Oracle Database Appliance : How to perform Bare Metal Restore of a Single node on ODA

Shared Disk Failure

Once the failed disk is replaced, the Oracle Appliance Manager automatically adds the new replacement disk to appropriate ASM diskgroup. No user action is required.

See the following known issues:

Note 1401471.1 - ODA :- After replacing the disk ,the disk has not shown and not added to ASM

NIC Card Failures

     Oracle Database Appliance has two PCIe Ethernet NIC cards:

     PCIe slot 1 – Quad-port 1-Gb Ethernet NIC
     PCIe slot 0 – Dual-port 10-Gb Ethernet NIC

     If any of these cards are replaced, the MAC addresses associated with the interfaces on the card will change. Accordingly, you will need to modify the following files to reflect the new MAC addresses:

          /etc/sysconfig/network-scripts/ifcfg-<interface_name>

     Here is an example file:

      Notes:

• •  You need to change the MAC addresses of all the interfaces on the new replacement card.
  • Network service needs to be restarted after making the changes. (service network restart).
  • Since network service needs to be restarted, it is better to log into the box using ILOM interface when making these changes.

This solution has no attachment