Dell OpenManage Array Manager User's Guide Version 3.5
Managing the Controller Battery
Array Manager and FAST Compatibility Issues
This chapter describes how to configure and manage arrays attached to PERC subsystem 2 controllers. You can display your system's PERC subsystem 2 controllers in the Array Manager tree view by doing the following:
The following controllers are supported by the PERC subsystem 2:
The PERC 2 RAID controller contains four channels. PERC 2/Si and PERC 3/Si are single integrated channel controllers. The PERC 3/Di is a dual integrated channel controller.
The PERC 2, PERC 2/Si, PERC 3/Si, and PERC 3/Di controllers support the following RAID levels:
For a description of the RAID levels and concatenation, see:
You can create a maximum of 24 virtual disks on a PERC 2, PERC 2/Si, PERC 3/Si, or PERC 3/Di controller.
Array Manager allows you to create virtual disks, which are logical disks that can be created from one or more physical disks (known as array disks). The logical disks can have RAID layouts, which let you to manage your storage more efficiently. Once a virtual disk is created and a rescan is done, the Windows® operating system sees the virtual disk as if it were a physical disk. The topics in this section include:
See also the section in this chapter on array disks, "Array Disk Commands."
Creating virtual disks is the first step in configuring your RAID storage subsystem. For more information about creating RAID solutions, see "What Is RAID?" and "Choosing RAID Levels and Concatenation" in the "Understanding Storage Concepts" chapter.
A virtual disk is channel-redundant when it maintains redundant data on more than one channel. If one of the channels fails, data will not be lost because redundant data resides on another channel. For more information, see "Channel Redundancy."
The Create Virtual Disk Express Mode
The Create Virtual Disk Advanced Mode
When using Express Mode, the wizard selects disks for you depending on the RAID level you selected and on the controller. When using Advanced Mode, the wizard displays the disks that you selected from the Available Disks section of Advanced Mode. Use the Blink and Unblink buttons to identify which disks are selected by blinking their LED display.
When using Advanced Mode, you can select the disks to be used from the Available Disks tabs. These tabs indicate the disks attached to each controller channel that are available depending on the RAID level you selected and the controller.
If the newly created virtual disk does not appear under Disks in the tree view, then perform a rescan by selecting Rescan from the View pull-down menu. Rescan forces the Array Manager software to locate and query all the objects in the system. This can take several minutes if there are numerous disks on the system.
When initializing is complete, the virtual disk appears in the tree view of the Array Manager console. Each virtual disk is automatically mapped to a Windows disk. For example, if Virtual disk 0 is mapped to Windows Disk 0, the Array Manager console will display Virtual disk 0 as "Virtual Disk 0 (Disk 0)," and the Windows disk 0 as "Disk 0 (Virtual Disk 0)."
If a virtual disk is not mapped to a Windows disk, reboot the system.
When a virtual disk is created with PERC 2, 2/Si, 3/Si, and 3/Di controllers, a write signature is not automatically written to the disk. Windows requires a write signature on virtual disks. For instructions on writing a disk signature, see the section "Write a Disk Signature" in the "Disk Management" chapter.
Deleting a virtual disk permanently deletes all information contained within all volumes residing on that virtual disk.
|
Caution All data (including file systems, volumes, Windows disks) on the virtual disk is lost when the disk is deleted. |
Use this procedure to delete a virtual disk on a PERC 2, PERC 2/Si, PERC 3/Si, or PERC 3/Di controller:
Use the Virtual Disk Reconfiguration wizard to change attributes of a virtual disk on a PERC 2, PERC 2/Si, PERC 3/Si, or PERC 3/Di controller. You can:
Use this procedure to reconfigure a virtual disk on a PERC 2, PERC 2/Si, PERC 3/Si, or PERC 3/Di controller:
This section describes the following commands that are used to manage virtual disks:
Use this function to change the read and write cache policies of a virtual disk.
The options are to enable or disable the write cache and to enable or disable the read cache.
Use this command to rename a virtual disk. Renaming a virtual disk enables you to change the virtual disk name displayed in the Array Manager tree view. The name you specify is also displayed in the BIOS.
If you do not specify a name for a virtual disk (either when you create the virtual disk or by using the Rename command) then the name for the virtual disk in the BIOS will be "Virtual Disk".
When changing a virtual disk name, the numbering format for the virtual disk remains unchanged. The numbering displayed in the Array Manager tree view is not displayed in the BIOS.
Use this operation to assign and configure a dedicated hot spare. This hot spare will be used to rebuild a degraded array disk, if possible.
If an array disk belonging to a virtual disk fails or is removed, the virtual disk will become degraded. A degraded virtual disk that has a properly configured hot spare will automatically rebuild onto that hot spare.
![]() |
Note This operation is highly recommended for redundant RAID types such as RAID-1, RAID-5, RAID-1 Concatenated, and RAID-10. |
![]() |
Note Hot spares for PERC 2, PERC 2/Si, PERC 3/Si, and PERC 3/Di controllers are always associated with a specific virtual disk. These controllers do not have a global hot spare option. |
This operation verifies the accuracy of the redundant data in the virtual disk. You cannot check the consistency of a non-redundant RAID type. If necessary, you can pause the consistency check once it is in progress with the Pause Consistency Check option.
![]() |
Note It is recommended that you do not access any volumes on the virtual disk during the consistency check. |
If you have initiated a consistency check using the Check Consistency option, you can suspend or pause the consistency check with the Pause Consistency Check option. Once paused, you can later resume the consistency check by using the Resume Consistency Check option. A paused consistency check will also automatically resume when the system is rebooted.
After having suspended a consistency check by using the Pause Consistency Check option, you can resume the consistency check with the Resume Consistency Check option.
Use this procedure to format a virtual disk on a PERC 2, PERC 2/Si, PERC 3/Si, or PERC 3/Di controller. This operation erases the files and removes the file systems on virtual disks but keeps the virtual disk configuration intact.
![]() |
Note If you are formatting a virtual disk on a NetWare® system, you must first go to the NetWare console and remove the corresponding volume and partition. For information on how to perform these operations, see the appropriate NetWare documentation. |
![]() |
Note If you format a virtual disk that has been upgraded to dynamic, the disk functions may become inactive or the status of the disk may change to "foreign" or "offline." You can view a disk's status by selecting the disk's General tab. When using a PERC 2, 2/Si, 3/Si, and 3/Di controller, you can resolve these problems by doing a global rescan. |
This command recovers the disk space occupied by an orphan or dead disk segment. An orphan disk segment is part of a multipartition virtual disk that cannot access another disk segment within the same virtual disk. A dead disk segment is either no longer needed by the virtual disk or is physically damaged. This operation can be performed only on failed virtual disks of redundant RAID types.
![]() |
Note You can restore only segments of a redundant RAID type. |
![]() |
Note This operation is not supported in NetWare. |
Use the Split Mirror command to permanently convert a mirror set or a multilevel virtual disk of mirror sets into two concatenated sets or two multilevel virtual disks of concatenated sets and preserve the data in each resulting virtual disk.
![]() |
Note This operation is not supported in NetWare. It is also not supported when using a dynamic virtual disk with a PERC 2, PERC 2/Si, PERC 3/Si, or PERC 3/Di controller. |
|
Caution Your virtual disk will no longer be redundant after performing a Split Mirror operation. |
Use the Unmirror command to convert the mirror set to a concatenated set or convert a multi-level virtual disk of mirror sets to a multilevel virtual disk of concatenated sets. Unmirroring preserves the data in the mirror set or multilevel virtual disk of mirror sets, but erases any data maintained by the redundant segment and returns the redundant virtual disk's segments to free space.
![]() |
Note This operation is not supported in NetWare. It is also not supported when using a dynamic virtual disk with a PERC 2, PERC 2/Si, PERC 3/Si, or PERC 3/Di controller. |
|
Caution Your virtual disk will no longer be redundant after performing an Unmirror operation. |
The Blink Virtual Disk command allows you to locate the array disks included in a virtual disk by blinking the light-emitting diodes (LEDs) on the array disks. This command will automatically cancel in 30 seconds.
![]() |
Note This procedure is only supported for hotswap array disks (disks that reside in a carrier). |
The Unblink Virtual Disk command allows you to cancel the Blink Virtual Disk command before the 30-second threshold has been reached.
![]() |
Note This procedure is only supported for hotswap array disks (disks that reside in a carrier). |
The Properties command brings up a screen that displays the properties of the virtual disk.
This section describes the commands available on the Array Group context menu. These commands are:
For more information about this command, see the section "Creating Virtual Disks" in this chapter.
Use Reset Configuration to erase all information on the controller, so that you can perform a fresh configuration.
This operation will destroy any existing virtual disks on the controller. A warning is displayed before the operation is performed. You can cancel the execution at this step.
|
Caution Resetting a configuration permanently destroys all data on these virtual disks. If the system or boot partition resides on these virtual disks, it will be destroyed. |
When a controller has array disks, the array disks attached to the controller can be displayed in the right pane of the Array Manager console. You can also specify that the array disks be hidden from view in the right pane.
To show or hide the display of array disks in the right pane:
When you right-click an array group object in the tree view, the pop-up menu displays either the Show Array Disks or Hide Array Disks option. Both options are not displayed on this menu at the same time. If the controller does not have array disks attached to it, then the Show Array Disks and Hide Array Disks options are not displayed.
This command brings up a screen with the properties of the array group.
This section describes Array Manager's commands for working with array disks. These commands are:
![]() |
Note A right-click any one of the array disks displays a menu of supported operations. These menus are dynamic. Not all menu options are available at all times. |
Use this operation to verify that a disk is free from defects. This assures that the disk is safe to use as a member of a virtual disk.
The Initialize command prepares an array disk for use as a member of a virtual disk. For details, see the "Initialize Array Disks" topic under the "General Controller Commands" section later in this chapter.
|
Caution This is a data-destructive command. Any virtual disks residing on this array disk will be removed. |
The Format command performs a low-level format on an array disk.
|
Caution This is a data-destructive command. Any virtual disks residing on this array disk will be removed. |
This command removes the disk space occupied by an orphan or dead disk segment.
A segment may be designated as dead if it is logically bad, is physically damaged, or resides on a disk that has been disconnected from and then reconnected to the controller.
The Blink command allows you to find an array disk within an enclosure by blinking one of the light-emitting diodes (LEDs) on the disk. This command will automatically cancel in 30 seconds. You may wish to use this command to locate an array disk that has failed.
![]() |
Note This procedure is only supported for hotswap array disks (disks that reside in a carrier). |
The Unblink command allows you to cancel the Blink command before the 30-second threshold has been reached.
![]() |
Note This procedure is only supported for hotswap array disks (disks that reside in a carrier). |
Use this procedure to prepare for removing an array disk from a PERC 2, 2/Si, 3/Si, and 3/Di controller.
Before you remove any physical disk from an enclosure, it is recommended that you perform this operation to prevent data loss.
The disk that you have prepared to remove will not be displayed in the Array Manager tree view. After you have replaced this disk and performed a rescan, it is displayed once again in the tree view.
![]() |
Note This procedure is only supported for hotswap array disks (disks that reside in a carrier). |
Use this command to display the array disk properties.
![]() |
Note Array disks for the PERC 2, PERC 2/Si, PERC 3/Si, and PERC 3/Di controllers display Ready status regardless of whether they are a part of a virtual disk or not. Ready status means the disk is operational. |
The PERC subsystem 2 has the following commands on its context menu:
This command allows you to view a report on the PERC subsystem 2.
This command allows you to save the controller configuration.
Use this command to view the properties of the PERC subsystem 2.
By default, the operating system and server shut down when the PV220S and PV221S enclosures reach a critical temperature of 0 or 50 degrees celsius. If you have implemented Channel Redundancy on the PV220S and PV221S enclosures, however, you can specify that only the enclosure and not the operating system and server be shut down when the enclosure reaches a critical temperature of 0 or 50 degrees celsius.
![]() |
Note You must perform this procedure on each of the server's subsystems in the Array Manager tree view in order to implement smart thermal shutdown. |
To specify that only the enclosure shut down at critical temperatures:
For more information, see:
If you have previously selected the Enable Smart Thermal Shutdown for PV22xS option, you can deselect this option and return the system to its default setting by selecting the Disable Smart Thermal Shutdown for PV22xS option. The Disable Smart Thermal Shutdown for PV22xS setting indicates that the operating system and the server will shut down when the PV220S and PV221S enclosures reach a critical temperature of 0 or 50 degrees celsius.
To specify that the operating system and server shut down at critical temperatures:
For more information, see:
This section describes the following controller commands and information on how to manage controller batteries:
The Rescan operation scans the channels of the controller to verify the currently connected devices or to recognize new devices that have been added to the channels.
After the operation is finished, the console is refreshed and any configuration changes appear under the Array Group object and under the channel object.
The removal of an array disk within a virtual disk is reflected immediately in the console, whereas an addition is reflected only by performing the Rescan operation. Array disks that are not part of virtual disks will not be polled for status changes.
This command pauses I/O on all channels that are polled for status changes on the controller. Use the Pause I/O command when you need to hot-swap disk drives without having to reboot.
If the disks are contained in an enclosure that supports hot-swapping, Pause I/O is not required unless the disk is part of a virtual disk.
|
Caution This can be a potentially dangerous command, and sufficient steps should be taken to warn others who may be using any virtual disks on the selected controller that I/O will be paused. |
The controller stays in this paused state for 150 seconds and then automatically resumes.
Use this function to initialize one or more array disks simultaneously. Array disks must be initialized before they can be used by Array Manager.
This command extracts a log from the controller(s) and places it in the Array Manager installation directory. When you export the log file from the controller, the controller's event information is cleared. The log gives detailed information on the controller activities and can be useful for troubleshooting. The log's file name is the word "log" and the month and day, such as "log0921" or "log_0921" for a log created on September 21.
Use this function to display controller attributes.
Properties for each of PERC 2, PERC 2/Si, PERC 3/Si, and PERC 3/Di controllers appear differently. When a controller is opened in read-only mode, the console displays Read only in the name that appears in the left pane. If the controller is already in use, it can be opened only in read-only mode.
![]() |
Note A controller will be opened in Read Only mode if it is already in use. |
The battery of the PERC 2 controller preserves the contents of its nonvolatile cache memory in the event of power loss. The battery provides a minimum of 72 hours protection for DIMMs up to 64 MB and 48 hours protection for 128-MB DIMMs. However, the duration of protection, or holdover time, depends on the battery's charge level. This section has two topics:
![]() |
Note The PERC 2/Si and PERC 3/Si RAID controllers do not contain a battery and therefore do not support battery management. |
The battery is fully discharged when you first install the controller. The controller detects this condition and automatically begins charging the battery. This charging process takes approximately three hours. The controller can be used during this time; however, the battery is unable to meet the specified holdover time until it is fully charged. The battery is still able to handle brief power losses during the initial charge cycle.
After you have installed the controller, it notifies you every six months that a battery reconditioning cycle is required. This reconditioning cycle requires a full discharge and recharge of the battery. It ensures that the battery's capacity is being measured correctly and that the battery's full holdover time is maintained.
The reconditioning cycle takes approximately 8 to 10 hours. During this time, the battery's holdover time is reduced to zero (during discharge) and then restored (during full charge). To ensure that no data is lost from nonvolatile cache memory, the controller automatically disables cache (both read and write) during a reconditioning cycle. When the cache is disabled, performance is degraded for the reconditioning period. Once the reconditioning cycle is complete, the cache is restored to its original settings.
It is highly recommended that battery reconditioning be done during off-peak times.
This section describes the following general SCSI channel commands:
This operation can be used to rescan the channel to verify the connected devices or to check for devices that have been added.
This operation displays channel attributes.
The channel properties include:
Array Manager and FAST cannot both be installed on the same machine. If the Array Manager installation process detects that FAST is already installed, the Array Manager installation will not complete.
After some planning and preparation, it is possible to migrate array disks, virtual disks, and array volumes between one controller and another, between one enclosure and another, or between one server and another. You can perform array and volume migration on PERC 2, PERC 2/Si, PERC 3/Si, and PERC 3/Di controllers as long as the following conditions are met:
To migrate virtual disks and volumes from one system to another, do the following.
The migration is complete. The virtual disk is now manageable via Array Manager.
The terminology used in the Array Manager console differs from the terminology used in the PERC 2 BIOS Container Configuration Utility (CCU). The following table shows a summary of these differences.