Troubleshooting Scenarios

There are four basic scenarios that you can expect to encounter during life of a server:

Scenario 1 (Typical first time server installations: No server options need to be installed)—If the server was shipped to you with all the server options you ordered already installed, run the diagnostic tests before placing the server into service. This way you can verify whether the server is in good working order or not (the server might have been damaged in shipment). If the diagnostic tests indicate that the server has a faulty component, notify the seller (Oracle or the Oracle reseller) that the server failed initial testing and should be repaired or replaced.
Scenario 2 (First time server installations: Server options shipped separately and need to be installed)—If the server was shipped to you with options that you need to install, run the diagnostic tests before installing the options. This will ensure that the server arrived in good working order. Only after determining that the server is in good working order should you install the options. After installing the options, you should run the diagnostic tests again to ensure that the server is in good working order with the options installed. Then you can place the server into service. If the diagnostic tests reveal that an installed option is faulty, remove the option and return it for replacement.
Scenario 3 (Upgrading a server: You have placed the server into service and then you receive an upgrade or option that needs to be installed)—In this case, take the server out of service, run the recommended diagnostic tests, install the upgrade, and then run the diagnostic test again before placing the server back into service. If the diagnostic tests reveal that the installed upgrade is faulty, remove the upgrade and return the upgrade for replacement.
Scenario 4 (An in-service server fails: You placed the server into service and it has been operating properly for a period of time but then fails)—In this case, take the server out of service, do a visual inspection of the server’s internal components to see if you can detect any obvious problems (such as a burnt component or dust accumulation that is preventing cooling of server components). If necessary, run the recommended diagnostic tests to identity the faulty component, remove and replace the faulty component, run the diagnostic tests again to ensure that the server has been repaired, and then place the server into service.

Note - For scenario 4, if you have a service contract with Oracle that includes the Automated Service Request (ASR) service, Oracle Service diagnoses and repairs server failures. Whether you have the ASR service or not depends on availability of this support to your site.

The following table provides step-by-step troubleshooting procedures.

Troubleshooting Scenarios	Do this...
Scenario 1: Typical first time server installations (no server options that need to be installed)	Before you place a server into service, that is, configure the server’s operating system and use it to process data, run the following diagnostic tests to confirm that the server arrived in good operating condition: U-Boot Pc-Check Tests failed: If the previous tests identify a server failure, assume that the server was damaged in shipment. Terminate the installation process and notify Oracle Service of the problem. These steps ensure that the server is covered under warranty. Tests passed: Place the server into service. Note - If a network connectivity problem occurs when placing a server into service for the first time, ensure that the network access point for the server is activated.
Scenario 2: First time server installations (server options need to be installed)	Before you install the server options that shipped separately and place a server into service, that is, configure the server’s operating system and use it to process data, run the following diagnostic tests to confirm that the server arrived in good operating condition: U-Boot Pc-Check Tests failed: If the previous tests identify a server failure, assume that the server was damaged in shipment. Terminate the installation process and notify Oracle Service of the problem. These steps ensure that the server is covered under warranty. Tests passed: If the previous tests show the server to be in good working order, do the following before placing the server into service: Install the server options that were shipped separately. Run U-Boot. Run Pc-Check. Note - If a network connectivity problem occurs when placing a server into service for the first time, ensure the network access point for the server is activated.
Scenario 3: Upgrading a server that has been in service	Before you install a server upgrade (memory, hard disk drives, I/O cards, or power supply), take the server out of service and run the diagnostic tests to confirm that the server is in good operating condition. Follow these steps: Take the server out of service. Run the following diagnostic tests: U-Boot Pc-Check Install the server upgrade. Run the diagnostic tests again. Tests failed: If the diagnostic tests identify a server component failure, one of the installed options was faulty or the server was damaged when you installed the options. In either case, remove and replace the faulty component, run the above diagnostic tests again to confirm that the problem has been corrected, and place the server into service. Note - If the failed component is a non-replaceable component on the server’s motherboard, the motherboard can be returned to Oracle for repair or a replacement motherboard can be ordered and replaced in the field by authorized service personnel. Tests passed: Place the server into service.
Scenario 4: The server has been operating problem free for a long time (for example, months or years) and the Service Action Required LED on the server’s front panel illuminates.	Do the following: Before opening the server’s cover, inspect the server and SP log files for obvious fault messages and any fault LED indicators. Inspect the server’s internal components for obvious problems, such as: A burnt component A smell caused by a burnt component Anything that would inhibit proper cooling of server components, such as the accumulation of dust on the CPU heatsinks If the visual inspection does not identify a failed component, run the following diagnostic tests to identify the failed component: U-Boot Pc-Check If the previous tests do not identify the failed component, run SunVTS. If the failed component is a customer-replaceable unit (CRU), replace it. CRUs are defined in the service manual for the particular server model. If the failed component is a field-replaceable unit (FRU), initiate a service request with Oracle Service. FRUs are defined in the service manual for the particular server model. Note - If the failed component is a non-replaceable component on the server’s motherboard, the motherboard can be returned to Oracle for repair or a replacement motherboard can be ordered and replaced in the field by authorized service personnel. Note - For scenario 4, if you have a service contract with Oracle that includes the ASR service, then Oracle Service diagnoses and repairs server failures. Whether you have the ASR service or not depends on availability of this support to your site.

Caution - If you do not have a replacement component for a failed component and you want to place the server into service, before doing so, replace the faulty component with a filler panel. Placing the server into service without a filler panel in a blank component space could cause the server to overheat and shut down.

Skip Navigation Links
Exit Print View
	Oracle x86 Server Diagnostics Guide