IBM Support

VMWare fails to load or install after adding a PCI-X card - IBM eServer xSeries 440

Troubleshooting


Problem

Depending on a customer's PCI-X configuration (i.e. types and number of adapters), VMware will load and install without any difficulties. However, after adding a single adapter to the existing configuration, VMware will sometimes fail to load or installwith a variety of error messages.

Resolving The Problem

Source

RETAIN tip: H177332

Symptom

Depending on a customer's PCI-X configuration (i.e. types and number of adapters), VMware will load and install without any difficulties. However, after adding a single adapter to the existing configuration, VMware will sometimes fail to load or install with a variety of error messages.

Some of the error messages are "SCSI: device set offline-not ready or command retry failed after host reset; host 2 channel 1 id 11 LUN 0" and "bad eip value". While other variations of the errors may appear, no hardware failures are indicated and no messages appear in the service processor log. In addition, no POST or beep codes appear during the boot process and system diagnostics passes all tests. The problem can be tested by removing the last adapter installed to see if VMware functions normally.

Affected configurations

The system may be any of the following IBM eServer xSeries servers:

  • xSeries 440, Type 8687, any model
The system is configured with one or more of the following:
  • IBM RXE-100 Remote Expansion Enclosure, type 8684, model 1RZ

The system is configured with the following option(s):

  • Netfinity ServeRAID-4H Ultra160 SCSI Controller, option part number 37L6889, FRU part number 37L6892 (controller), FRU part number 37L6902 (cache daughter card), FRU part number 37L6903 (cache battery)
  • Netfinity ServeRAID-4L Ultra160 SCSI Controller, option part number 37L6091, FRU part number 09N9540
  • Netfinity ServeRAID-4M Ultra160 SCSI Controller, option part number 37L6080, FRU part number 37L7258
  • Netfinity ServeRAID-4M Ultra160 SCSI Controller (Japan), option part number 19K0565, FRU part number 00N9543 (Japan)
  • ServeRAID-4Lx Ultra160 SCSI Controller, option part number 06P5740, FRU part number 06P5741
  • ServeRAID-4Mx Ultra160 SCSI Controller, option part number 06P5736, FRU part number 06P5737
The following operating systems are affected:
  • VMware ESX Server, version 1.5x
Workaround

Disabling the onboard SCSI controller will avoid the problem. If the onboard SCSI controller is needed, open Advanced setup/PCI configuration and change the ServeRAID IRQ to 11. The ServeRAID and onboard SCSI controller share the same IRQ by default.

Additional information

Interrupt resources under VMware 1.5x may be limited. Depending on the adapter configuration, the available PCI-X interrupt resource allocation may be exceeded. VMware has been made aware of this issue and is investigating. The current workaround is to limit the amount of interrupt resource being used. One way is to disable any PCI-X devices or controllers that are not being used. Another option is to move the ServeRAID adapter to an IRQ which will not conflict with the onboard SCSI controller. Sharing the IRQ between the onboard SCSI controller and the ServeRAID adapter causes an excess amount of resource to be allocated.
 
Visit the links below for proper VMware installation instructions:
Installing VMware ESX Server version 1.5
Installing VMware ESX Server v1.5.2

 

Document Location

Worldwide

Operating System

Older System x:VMware ESX Server

[{"Type":"HW","Business Unit":{"code":"BU016","label":"Multiple Vendor Support"},"Product":{"code":"HW191","label":"Older System x->xSeries 440"},"Platform":[{"code":"PF032","label":"VM"}],"Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
29 January 2019

UID

ibm1MIGR-50459

Page Feedback