Ibm hs22 memory scrub failed relationship

dW:Eserver:IBM BladeCenter Forum:HS22 memory scrub failed - IBM BladeCenter Forum Forum

I have two hs22 blades that are giving 0xfc memory scrub errors for certain dimm slots. This happens regardless of the module that I. Remove all storage drives from the blade server (see Removing a hot-swap .. Error, Memory device X (DIMM X Status) memory scrub failed [Note: X=]. Overview The IBM® BladeCenter HS22 offers great performance balanced with Light Path Diagnostic and Predictive Failure Analysis® enables quick serviceability and . Memory hardware scrubbing, designed to correct many soft memory errors of cause-and-effect relationships from different log sources in the system.

Blades make it easier than ever to try new ideas and technologies to improve your business. Examples of good solutions for BladeCenter S Company A has an older application that is central to their business model. It requires an older version of Linux, is not virtualized, and the cost to move away from the system is expensive. In the meantime, they want to test customer analytic software that requires Windows Server Their storage requirements are low, but the new software has robust system requirements.

Company A can benefit greatly from the BladeCenter S product line. They can use an entry level HS23E blade to support older applications, and add a more powerful x86 blade to support the new software.

Using the drive and storage modules in the chassis provides all of their business requirements in one package. They would have four remaining slots for growth, increasing their speed to market and lowering their overall support costs.

Company B is a large pharmacy retailer with several hundred locations throughout the United States. Each location has nearly a hundred clients that depend on a back-office server for the point of sale system. If the server crashes, the entire location cannot service their customers. They have no onsite IT personnel and no real data center environment in their stores. The business is losing money as their rack mounted servers age in the environment. Company B 4 Implementing the IBM BladeCenter S Chassis 23 must remain flexible because retail trends change quickly and they do not want to fall behind the competition.

Company B is also a good example of an environment where the BladeCenter S can be a great benefit. The solution provides a flexible store in a box solution that can run in the manager s office or any secured location on the existing power infrastructure. Store employees can be used for support by working with remote personnel off-site.

Figure shows an illustration of the internal components of BladeCenter S. Chassis The term chassis typically refers to the physical frame of the BladeCenter, including the interconnects that are contained within the frame: The midplane, power supply, and blower modules. Although the chassis does not possess any inherent intelligence, it does contain a significant amount of circuitry.

IMM error messages

This circuitry is often referred to as paths within the chassis. These paths comprise the backbone of the BladeCenter and allow it to share the servers, switches, and power. Blade servers Any serve that is designed in the standard BladeCenter form factor is considered to be a blade server, or blade.

Blades are universal in size and shape. They can be inserted in any of the existing BladeCenter portfolio. However, processor power demands might prevent their usage in all chassis. Blades consist of a system board, processors, memory, expansion ports, and two redundant midplane connections. Blades rely on the chassis to provide all necessary power, cooling, connectivity, and management.

The midplane is set up so that there is a series of upper connection points and an identical set of lower connection points. These provide power and redundancy. On the opposite side of the midplane are other numerous connection points, which allow for power supply, blower, and expansion module bay connectivity.

The power supply and blower connections on the midplane function independently of their upper or lower status. However, the expansion bays are uniquely pathed to upper or lower connections points as they relate to blades. The midplane does not possess any intelligence. It is strictly a pass-through mechanism for blade, module bay, and power connectivity. BladeCenter S also has a serial pass-through bay and two storage module bays.

These bays are all directly connected to the midplane. However, they do not necessarily correlate to the upper or lower connections of the blade servers. Power supply and blower modules are independent of blade connectivity. The management module bay can communicate to a blade through the upper or lower midplane connection.

IBM BladeCenter HS22 is a versatile, easy-to |

The expansion module bays have committed communication paths that are mapped directly to upper or lower connections for the blade slots in the chassis. Advanced management module The advanced management module AMM is the center point for the BladeCenter infrastructure intelligence. The AMM is the primary means of management for the chassis, and controls all aspects of power, connectivity, and communication. It uses an upgradeable firmware and web user interface to run all routine hardware-based management tasks for blades, expansion modules, and configuration of storage modules.

The AMM also acts as a proxy for expansion modules. This configuration allows access by using direct that is, IP address or indirect that is, internal chassis methods for specific module management.

The module has six external RJ45 console access ports on it, each of which is physically pathed directly to a blade slot. The module was designed for Chapter 1. Introduction to the IBM BladeCenter S chassis 7 26 serial console access only, and it is not intended for use by modems or other serial port devices. Storage module The concept of a chassis with a self-contained disk subsystem or storage module is unique to the BladeCenter S.

The module is fundamentally a collection of disk drives that are made accessible through a SAS module and a SAS daughter card. The SAS module is responsible for both the provisioning of physical disk drives through zoning and for failover redundancy when two SAS modules are present.

The expansion card can address only those disks that are assigned to it from the SAS module, and can create arrays only from those disks.

There are various hard disk drive types available, so it is important to understand the features and benefits of each type. Certain considerations must be taken into account because there might be specific hardware requirements to support certain types. Serial-Attached SCSI disks are designed for high performance requirements where maximum throughput and transaction capability is needed. Serial Advanced Technology Attachment disks are designed to provide lower-cost mass storage capacity, when compared to SAS disk.

They are designed for systems where performance is not a requirement. They do not fit well into server environments. It is intended to overcome performance and scalability limitations of bus topology while providing enterprise-class reliability and software stack compatibility.

Because of its high performance, reliability, and scalability features, the SAS interface is widely used in the IBM System x server systems for both internal and external storage connectivity for wide range of applications and usage patterns.

In general, three types of devices form the SAS topology: The targets are the end-point devices, such as disk or tape drives. The SAS expander is a switch device that connects more target devices to the initiator than the number of ports the initiator has. This dramatically increases SAS fabric scalability without sacrificing reliability and performance. In addition, expanders support wide SAS links or aggregated links that consist Chapter 1.

Introduction to the IBM BladeCenter S chassis 9 28 of several narrow SAS links for expander-expander or expander-initiator connections to increase overall performance of the fabric Near line disk drives NL disks Data that are stored on an NL disk drive are intended to be accessed infrequently, but require 24x7 availability. From the application perspective, these drives are commonly used for storing archives, document images, multimedia libraries, and backups. There are two types of NL drives: Because of these advantages, IBM System x servers always use NL SATA drives as they offer almost twice the reliability, better support of multi-drive RAID array deployments, and 24x7 hours of operation 24 hours per day, 7 days per week in multi-user workload environments.

They do so without significant cost disadvantage when compared to desktop SATA drives. A RAID array is a group of physical disks that uses certain common method to distribute data across the disks. The data are distributed by stripe units. A stripe unit is the portion of data that are written to one disk drive immediately before the write operation continues on next drive. When the last drive in array is reached, the write operation continues on the first drive in the block that is adjacent to the previous stripe unit written to this drive, and so on.

The process of distributing data in this way is called striping. A stripe unit is a minimal element that can be read from or written to the RAID array.

Stripe units can contain data or recovery information. The particular striping method that is used for data distribution is also known as the RAID level. The RAID level reduces availability, performance, and available storage capacity because achieving redundancy always lessens disk space by the amount reserved for storing recovery information. Spanned RAID arrays combine two or more basic RAID arrays to provide higher performance, capacity, and availability by overcoming limitation of the maximum number of drives per array supported by a particular RAID controller.

This flexibility allows you to run all of your diverse workloads inside a single architecture. The slim, hot-swappable blade servers fit in a single chassis similar to books in a bookshelf. Each blade server is an independent server with its own processors, memory, storage, network controllers, operating system, and applications.

The blade server simply slides into a bay in the chassis and plugs into a midplane or backplane, sharing power, fans, diskette drives, switches, and ports with other blade servers.

The benefits of the blade approach are obvious to anyone tasked with running down hundreds of cables strung through racks just to add and remove servers. With shared switches and power units, precious space is freed up, and blade servers enable higher density with far greater ease.

This book covers only a subset of the blade servers that are supported by the IBM BladeCenter S chassis, specifically the latest models. The following servers are covered in this section: These servers can support up to two multi-core Intel Xeon microprocessors. They are ideally suited for medium and large businesses for virtualization, hosted client, SAP, and enterprise applications.

Figure shows the HS23 server with its cover removed. Two QPI links up to 8. Up to MHz memory speed. Up to 20 MB L3 cache. Support for up to MHz memory speed, depending on the processor. Four memory channels per processor 2 DIMMs per channel. ECC, Chipkill, memory mirroring, and memory rank sparing.

The IBM BladeCenter HS22V high-performance blade server features new processor options

Two extra PCIe 2. One internal USB port for embedded hypervisor. Maximum resolution is x at 75 Hz with 16 M colors 32 bits per pixel.

Designed specifically for use in business critical and entry virtualization applications, the blade offers higher performance with MHz memory and optimal processor performance in a standard 30 mm form factor. Figure shows a blade server HS23E with its top cover removed.

Support for up to MHz memory speed depending on the processor. Three memory channels per processor two DIMMs per channel. ECC, Chipkill, memory mirroring, and memory sparing. Hard disk drives and solid-state drives. It delivers innovation with enhanced scalability, reliability, and availability features to enable optimal performance for databases, enterprise applications, and virtualized environments.

Two HX5 servers can be connected together for a high-performance single image with four processors and up to 1 TB of RAM in a blade form factor. For applications that must maximize available memory but that do not need four processors, a single HX5 server can be attached to an MAX5 memory expansion blade to form a single image with two processors and up to 1.

This level of processing and memory capacity is ideal for large-scale database or virtualization requirements. Up to two processors per single wide HX5; scalable to four processors. Intel Xeon E, E, and E families, up to 10 cores. Two non-hot-swap bays per single-wide HX5 supporting solid-state drives. Integrated systems management processor.

Integrated Matrox GeV video controller. Their ability to coexist in the same chassis with other IBM BladeCenter blade servers enhances the ability to deliver the rapid return of investment ROI that is demanded by clients and businesses.

It supports two onboard disk drive bays. Infrastructure applications Virtualization General enterprise applications High-availability and serviceability features Hot-swap blades enable easy access to each blade server.

The management module interfaces with each blade server for single systems management control. The BladeCenter HS22 blade servers deliver reliability and serviceability.

High-performance ECC memory, combined with an integrated ECC memory controller, to help correct soft and hard single-bit memory errors, while reducing disruption of service to LAN clients. Chipkill memory correction for up to four bits per DIMM to help keep your blade server up and running. Memory hardware scrubbing, designed to correct many soft memory errors automatically without software intervention. ECC L2 cache processors to help improve data reliability and reduce downtime.

Support for dual Gigabit Ethernet connections: Failover, adapter fault tolerance PXE 2. Alerts are generated when certain thresholds are exceeded refer to the Limitations section for restrictions. Two SATA connectors for two 1. The HS22V server memory is contiguous and is shared by both processors when both processors are installed. Chipkill is supported in independent mode when x4-based DIMMs are installed.

One or two non-hot-swap SATA 1. Forces failed processor offline Automatically reboots server Generates alerts Continues operations with the working processor PFA on SAS HDD options, memory, and processors to help alert the system administrator of imminent component failures.

Optional add-ons available for an additional charge Active Energy Manager AEM is positioned as a key component of the energyefficient technologies and services of IBM, which are part of the IBM Project Green initiative that began May AEM measures, monitors, and manages the energy management components built into IBM servers and provides a cross-platform management solution. AEM also retrieves temperature and power information via wireless sensors and collects alerts, events, and data from certain facility providers related to power and cooling equipment.

BladeCenter Open Fabric Manager helps make blade deployment easy: ToolsCenter helps make managing your server environment less complicated, more productive, and more cost-effective. ServerGuide automates installation of Microsoft Windows server operating systems, device drivers, and other system components, with minimal user intervention.

The ServerGuide Scripting Toolkit enables you to tailor and build custom hardware deployment solutions. BladeCenter Start Now Advisor is a configuration tool that can help you quickly configure components of the BladeCenter chassis.

It automatically updates the firmware for selected chassis components, and provides you with the option of saving your configuration. The Start Now Advisor guides you through the process of connecting your computer to the chassis, either over a network or through a direct attachment to the Ethernet port on the advanced management module. Storage Configuration Manager is an open-standardsbased management tool that provides a uniform and rich user interface that is easy to use. UXSPs facilitate the downloading and installation of drivers and firmware for a given system and verify that you are working with a complete set of updates which have been tested together.

IMM error messages - IBM BladeCenter HS22

Diagnostics Dynamic System Analysis DSA collects and analyzes system information to aid in diagnosing system problems. DSA creates a merged log that helps provide easy identification of cause-and-effect relationships from different log sources in the system. This network relays vital information about individual blade servers, such as: You can shut down and restart any blade server from anywhere on the network to help save time and costs associated with travel to the actual installation. These manageability functions are provided through a self-contained web page, creating an easy and familiar way to help administrators monitor, control, and maintain high availability.

The BladeCenter HS22 and HS22V blades can require less space and power resources than traditional rack offerings because of their high-density design, reduced power requirements, and single environment systems management. This is an extremely important consideration for: Large enterprises Application service providers Scientific and technical computing businesses They are an excellent fit for applications such as: All statements regarding IBM's plans, directions, and intent are subject to change or withdrawal without notice.

Any reliance on these statements of general direction is at the relying party's sole risk and will not create liability or obligation for IBM. The following Pseudo parts numbers cannot be ordered as stand-alone parts and can only be ordered as part of a configuration. The following publications are available immediately.