Starting Dell ManagerDell™ PowerEdge™ Expandable RAID Controller 4/Di User's Guide
Using Dell Manager in Red Hat Linux GUI Mode
Configuring Arrays and Logical Drives
Performing a Consistency Check
Exiting the Dell Manager Utility
Dell™ Manager is a character-based, non-GUI utility that changes policies, and parameters, and monitors RAID systems. Dell Manager runs under Red Hat Linux 8.x or later, Advanced Server 2.1 or later, and Novell NetWare 6.0 or later.
To start Dell Manager from any directory, make sure the program file is in the usr/sbin directory, and type:
dellmgr
On a Red Hat Linux system, for Dell Manager to work correctly in a terminal in GUI Mode, you must set the terminal type to linux and keyboard mappings.
Perform the procedure below if you use konsole, gnome terminal, or xterm.
The linux console mode, which you select from the terminal with the File —> Linux Console command, works correctly by default. The text mode console (non-GUI) also works correctly by default.
To prepare the system to use Dell Manager, perform the following steps:
TERM=linux
Export TERM
 |
NOTE: On a Red Hat Linux 8.x system, when you run Dell Manager (v. x.xx) from a Gnome-terminal in XWindows, the <F10> key cannot be used to create a logical drive. Instead, use the alternate keys <Shift><0>. (This is not an issue if Xterm is used to call dellmgr). The following is a list of alternate keys you can use in case of problems with keys <F1> through <F6>, and <F10>: |
See "Designating Drives as Hot Spares" for more information.
See "Creating Arrays" for more information.
Hot spares are physical drives that are powered up along with the RAID drives and usually stay in a standby state. If a hard drive used in a RAID logical drive fails, a hot spare automatically takes its place and the data on the failed drive is reconstructed on the hot spare. Hot spares can be used for RAID levels 1, 5, 10, and 50. The RAID controller supports up to eight hot spares.
The two methods for designating physical drives as hot spares are:
When you select any configuration option in the Configure menu, all physical devices connected to the current controller appear. Perform the following steps to designate a drive as a hot spare:
The list of hard drives displays.
The drive displays as HOTSP.
A physical drive selection screen displays.
- Rebuild
- Force Online
- Force Offline/Remove HSP (hot spare)
- Spin Up Drive
- Spin Off Drive
- Prepare to Remove
- Make Hotspare
- View Drive Information
- View Rebuild Progress
The selected drive displays as HOTSP.
Create arrays using Automatic, Easy or New Configuration. See "Using Auto Configuration", "Using Easy Configuration", and "Using New Configuration" for procedures.
In Automatic Configuration, the RAID controller examines the physical drives connected to it and automatically configures them into arrays and logical drives.
If logical drives have already been configured when you select Automatic Configuration, the configuration information is not disturbed. The RAID controller uses the following configuration guidelines in the following sequence:
Table 5-1 displays the logical drive parameters and default settings.
|
Parameter |
Setting |
|---|---|
Stripe size | 64 KB |
Write policy | Write-through |
Read policy | Adaptive |
Cache policy | Direct I/O |
Spanning | Enabled |
The write policy, read policy, and cache policy can be changed after configuration is complete.
The logical drives that result from Automatic Configuration display on the screen with a save prompt.
See "Initializing Logical Drives" for additional information.
In Easy Configuration, each physical array you create is associated with exactly one logical drive, and you can modify the following parameters:
If logical drives have already been configured when you select Easy Configuration, the configuration information is not disturbed.
The hot key information appears at the bottom of the screen.
The selected drive changes from MASTER to ONLIN A[array number]-[drive number]. For example, ONLIN A2-3 means hard drive 3 in array 2.
Use drives of the same capacity in a specific array. If you use drives with different capacities in an array, all the drives in the array is treated as though they have the capacity of the smallest drive in the array.
The number of physical drives in a specific array determines the RAID levels that can be implemented with the array.
The logical drive configuration screen appears. The window at the top of the screen shows the logical drive that is currently being configured as well as any existing logical drives.
The column headings are:
LD - The logical drive number
RAID - The RAID level
Size - The logical drive size
#Stripes - The number of stripes in the associated physical array
StrpSz - The stripe size
DriveState The state of the logical drive
The available RAID levels for the current logical drive display.
Stripe Size specifies the size of the segments written to each drive in a RAID 1, 5 or 10 logical drive. You can set the stripe size to 2 KB, 4 KB, 8 KB, 16 KB, 32 KB, 64 KB, or 128 KB. The default stripe size is 64 MB.
A larger stripe size provides better read performance, especially if your computer does mostly sequential reads. However, if you are sure that your computer does random read requests more often, select a small stripe size.
Write Policy specifies the cache write policy. You can set the write policy to Write-back or Write-through. The default is Write-through.
Write-through caching has a data security advantage over write-back caching, while write-back caching has a performance advantage over write-through caching.
 |
NOTICE: You should not use write-back for any logical drive that is to be used as a Novell NetWare volume. |
Read Policy enables the IDE read-ahead feature for the logical drive. Read-ahead specifies that the controller uses read-ahead for the current logical drive.
Cache Policy applies to reads on a specific logical drive. It does not affect the Read-ahead cache. The default setting is Direct I/O.
The array selection screen appears if any unconfigured hard drives remain.
A list of the currently configured logical drives appears. You are prompted to save the configuration.
The Configure menu displays.
See "Initializing Logical Drives" for more information.
In New Configuration, you can modify the following array parameters:
If you select New Configuration, the existing configuration information on the selected controller is destroyed when the new configuration is saved.
Hot key information appears at the bottom of the screen. <F10>.
The selected drive changes from MASTER to ONLIN A[array number]-[drive number]. For example, ONLIN A2-3 means hard drive 3 in array 2.
The available RAID levels for the current logical drive are listed.
The choices are:
To span two arrays, they must have the same stripe width (they must contain the same number of physical drives) and the arrays must be consecutively numbered.
For example, assuming Array 2 contains four hard drives, it can be spanned only with Array 1 and/or Array 3, and only if Arrays 1 and 3 also contain four hard drives. If the two criteria for spanning are met, the RAID controller allows spanning. If the criteria are not met, the Span setting makes no difference for the current logical drive.
The RAID 1 logical drives must have the same stripe size and use the whole capacity of the hard drives.
By default, the logical drive size is set to all available space in the array(s) being associated with the current logical drive, accounting for the Span setting and for partially used array space. For example, if the previous logical drive used only a part of the space in an array, the current logical drive size is set to the remaining space by default.
Stripe Size specifies the size of the segment written to each drive in a RAID 1, 5 or 10 logical drive. You can set the stripe size to 2 KB, 4 KB, 8 KB, 16 KB, 32 KB, 64 KB, or 128 KB. The default stripe size is 64 MB.
A larger stripe size produces better read performance, especially if your computer does mostly sequential reads. If you are sure that your computer does random reads more often, select a small stripe size.
Write Policy specifies the cache write policy. You can set the write policy to Write-back or Write-through. The default is Write-through.
Write-through caching has a data security advantage over write-back caching. Write-back caching has a performance advantage over write-through caching.
|
NOTICE: You should not use write-back for any logical drive that is to be used as a Novell NetWare volume. |
Read Policy specifies that the IDE read-ahead feature is used for the logical drive. You can set this parameter to Read-ahead, No Read-ahead or Adaptive. The default is Adaptive.
Cache Policy applies to reads on a specific logical drive. It does not affect the Read-ahead cache. The default is Direct I/O.
If space remains in the arrays, the next logical drive to be configured appears.
If all array space has been used, a list of the existing logical drives appears.
See "Initializing Logical Drives" for more information.
View/Add Configuration allows you to control the same parameters as New Configuration without disturbing the existing configuration information.
Hot key information appears at the bottom of the screen. <F10>.
The selected drive changes from MASTER to ONLIN A[array number]-[drive number]. For example, ONLIN A2-3 means hard drive 3 in array 2.
The available RAID levels for the current logical drive display.
The choices are:
To span two arrays, they must have the same stripe width (they must contain the same number of physical drives) and must be consecutively numbered. For example, assuming array 2 contains four hard drives, it can be spanned only with array 1 and/or array 3, and only if arrays 1 and 3 also contain four hard drives. If the two criteria for spanning are not met, the Span setting ignored.
The RAID 1 logical drives must have the same stripe size and use the whole capacity of the hard drives.
By default, the logical drive size is set to all available space in the array(s) being associated with the current logical drive, accounting for the Span setting and for partially used array space. For example, if the previous logical drive used only a part of the space in an array, the current logical drive size is set to the remaining space by default.
Stripe size sets the size of the segment written to each drive in a RAID 0, 1, or 5 logical drive. You can set the stripe size to 2 KB, 4 KB, 8 KB, 16 KB, 32 KB, 64 KB, or 128 KB. The default stripe size is 64 MB.
A larger stripe size produces better read performance, especially if your computer does mostly sequential reads. However, if you are sure that your computer does random reads more often, select a small stripe size.
Write Policy sets the cache write policy. You can set the write policy to Write-back or Write-through. The default is Write-through.
Write-through caching has a data security advantage over write-back caching, whereas write-back caching has a performance advantage over write-through caching.
|
NOTICE: You should not use write-back for any logical drive to be used as a Novell NetWare volume. |
Read Policy enables the IDE read-ahead feature for the logical drive. Set this parameter to Read-ahead, No Read-ahead, or Adaptive. The default is Adaptive.
Cache Policy applies to reads on a specific logical drive. It does not affect the Read ahead cache. The default setting is Direct I/O.
If space remains in the arrays, the next logical drive to be configured appears.
If the array space is used, a list of the existing logical drives appears.
You are prompted to save the configuration.
See "Initializing Logical Drives" for more information.
You should initialize each new logical drive you configure. You can initialize the logical drives individually or in batches (up to 40).
A list of the current logical drives appears.
The progress of the initialization for each drive appears in bar graph format.
The logical drive action menu displays.
The progress of the initialization appears as a graph on the screen.
If a hard drive fails in an array configured as a RAID 1 or 5 logical drive, you can recover the lost data by rebuilding the drive. If a rebuilding spare fails, a new rebuild is started using a second spare, if available. The capacity of the second spare must be equal to or greater than that of the failed drive.
Table 5-2 describes the automatic and manual rebuilds.
The list displays of the actions for the physical drives.
Rebuilding can take some time, depending on the drive capacity.
A device selection window displays the devices connected to the current controller. The failed drives display as FAIL.
The selected drives change to REBLD. Rebuilding can take some time, depending on the number of drives you have selected and their capacities.
Your RAID controller supports the ability to delete any unwanted logical drives and use that space for a new logical drive. You can have an array with multiple logical drives and delete a logical drive without deleting the whole array.
After you delete a logical drive, you can create a new one using the configuration utilities. You can create the next logical drive from the non-contiguous free space (`holes') and from the newly created arrays. The configuration utility provides a list of configurable arrays where there is a space to configure.
|
NOTICE: The deletion of the logical drive can fail under certain conditions: during a rebuild, initialization, or check consistency of a logical drive, if that drive has a higher logical drive number than the drive you want to delete. |
Perform the following steps to delete a logical drive:
The logical drives display.
This deletes the logical drive and makes the space it occupied available for you to make another logical drive.
Select Check Consistency to verify the redundancy data in logical drives that use RAID levels 1, 5, 10, or 50.
When you select Check Consistency, the parameters of the existing logical drives on the current controller and a selection menu listing the logical drives by number appear. If a discrepancy is found, it is automatically corrected, assuming always that the data is correct. However, if the failure is a read error on a data drive, the bad data block is reassigned with the generated data.
Perform the following steps to run the Check Consistency.
A progress graph displays for each logical drive selected.
A message appears if uninitialized logical drives remain in the system. Initialize any newly created logical drives.
Please read all restrictions and disclaimers.