No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search


To have a better experience, please upgrade your IE browser.


How to Troubleshoot the Fault That an IBM System X3650-M2 Server Fails to be Started

Publication Date:  2012-07-18 Views:  141 Downloads:  0

Issue Description

An IBM System X3650 M PC server running the Windows Enterprise 2003 SP2 OS can be properly started. After the server is mounted to a Huawei Symantec storage system, the server cannot be started from hard disks and is restarted repeatedly. Symptom 1: After the server is powered on, partitions are searched according to the BIOS startup sequence, but errors occur, such as CD ROM timeout, Hard Disk 0 timeout, and DHCP PXE. The server fails to be started. Symptom 2: The local disk and the mounted disk displayed in SCSI Disk Device properties are Bus Number 1, Target ID 0, LUN 0 and Bus Number 0,Taeget ID 0 ,LUN 0.

Alarm Information


Handling Process

Solution: Set the BIOS parameters and improve the working mode of the SATA controller to Legacy.

 Step 1 Set the BIOS parameters. Restart the OS and press F1. Select Boot Manager and press Enter. Select Add Boot Option and press Enter.

 Step 2 Improve the working mode of the SATA controller to Legacy. After Step 1 , the Legacy Only option is added in the startup sequence management. Press Esc to return to Boot Manager and press Enter. Select Change Boot Order and press Enter. Select Legacy Only to improve Legacy Only ahead of Hard Disk 0 (using + or - to adjust the location). Select Commit Changes and press Enter.

 Step 3 Restart the system.

Root Cause

After the symptoms mentioned above occur, first check whether the BIOS startup sequence is properly set and alarms are generated on the storage system, and then check the system logs. The troubleshooting procedure is as follows:

 Step 1 Change the startup sequence. Restart the OS and enter the BIOS interface. Set Hard Disk 0 behind CD-ROM. Restart the OS.

 Step 2 View the alarms generated on the storage system. No alarm is generated on the disk array. Remove the optical fiber from the server and restart the server to enter the OS.

 Step 3 View the system logs of the server. No alarm or error exists. The result indicates that the disk array and server are normal and unrelated to the fault. As the OS on the server fails to be started from hard disks and has being searched and restarted, it indicates that fault lies in hard disks because hard disks are classified into local disks and disks mapped from the disk array. According to symptom 2, the X3650 M2 PC server chooses the disk whose bus number is 0 as the startup disk by default. The BIOS startup sequence of the X3650 M2 PC server has special settings. After consulting IBM 400 hotline, know that the fault may be related to the working mode (Legacy working mode) of the ATA controller.


The fault is related to the working mode of the ATA controller. The IBM server has four 146 GB local SAS disks. Three disks form a RAID 5 group and another one disk is used as the hot spare disk. The server is connected to a S5300 by using multiple FC links. The working modes of the ATA controller include legacy mode and native mode. Different modes support different numbers of ATA devices. The legacy mode supports a maximum of four ATA devices and allocates only terminals 14 and 15 to the controller (the legacy mode is supported only by the DOS, Windows 98/ME, and Windows 2000 OSs). The legacy mode is classified into combine mode and non-combine mode. The combine mode supports up to two P-ATA devices and two S-ATA devices. The non-combine mode supports up to four P-ATA devices (P-ATA devices only) or four S-ATA devices (S-ATA devices only). The native mode can allocate all free terminals to the controller and is supported only by the Windows XP and Windows NET Server OSs. The native mode supports up to six ATA devices (four P-ATA and two S-ATA).