Sonnet Technologies RAID User Manual

Conﬁguration Tool and Utilities Operation Manual

for Fusion RAID Storage Systems

Contents

1.0 ATTO Configuration Tool Overview............................................................................................. 1

About the Configuration Tool

Configuration Tool Launch

Configuration Tool Navigation

Select the Local Host

Select the Controller

Select a Channel

Select a Device

1.1 RAID Controller NVRAM Settings............................................................................................... 7

SAS Address Boot Driver Heartbeat NCQ Device Wait Time Device Wait Count Spinup Delay Multiplexing PHY Speed

1.2 RAID Group Setup and Management.......................................................................................... 9

Preliminary Configuration Steps

Custom RAID Group Setup Steps

Hot Spares Setup and Usage

Mac OS Drive Formatting

Windows 7/Server 2008/Vista Drive Formatting

Windows XP/Server 2003 Drive Formatting

RAID Group Management Overview

RAID Group Capacity Expansion

RAID Group RAID Level Migration

RAID Group Deletion

RAID Group Rebuilding

RAID Group Properties Modification

1.3 Drive and RAID Group Monitoring............................................................................................ 17

Basic Drive Information

Detailed Drive Information

RAID Group Information

Drive Identification

S.M.A.R.T. Data

S.M.A.R.T. Monitoring Enabling and Disabling

S.M.A.R.T. Status Checking

S.M.A.R.T. Attribute Filtering

S.M.A.R.T. Notifications

1.4 Enclosure (SES) Health Monitoring.......................................................................................... 23

SES Status Checking

Contents

1.5 RAID Event Notifications........................................................................................................ 25

Basic Alerts

Logging

Email Alerts

1.6 RAID Group Media Maintenance............................................................................................... 27

Media Scan Options

Start a Manual Media Scan

Schedule a Media Scan

Cancel, Pause, or Resume a Media Scan

Manage Scheduled Media Scans

Automatic Media Scan Cancellation

View a Scan Report

1.7 Identify and Replace a Faulted Drive.......................................................................................33

Automatic Faulted Drive Identification

Manual Faulted Drive Identification

Faulted Drive Replacement with Auto Rebuild

Manual Faulted Drive Replacement

1.8 Recover Data from Offline RAID Groups................................................................................... 35

Definitions

RAID Group Failure Scenarios

Drive Replacement on a Failure Condition Replace RAID Group Members as Soon as They Fail A Warning About Drive Replacement Identifying Failed Drives Recovery Mode

Recovery from a Failed Rebuild

Recovery from a Failed Rebuild with a Second Unrecoverable Drive Failure

Recovery from Faults on a Critical Number of Drives Basic Recovery Mode Extreme Recovery Mode

Recovery from Replacement of the Wrong Drive

File System Repair Tools Mac OS X Users: Disk Utility Windows Users: CHKDSK Unix Users: FSCK

Using the CLI to Identify a Failed Drive

Using the CLI to Enable/Disable Recovery Mode

Using the CLI to to Enable Recovery with Writes

Contents

1.9 SNMP Configuration................................................................................................................ 45

Definitions

Details Enabled Mode SubAgent Mode Disabled Mode Basic Setup

Configuration Options Agent Port Communities Send Authentication Trap Enable Traps Trap Destination Table

Control Buttons Commit Restore Default Test Save MIBs

Troubleshooting Requests from an MIB browser time out: Traps are not received when testing RAID OIDs are skipped when walking or the tables are empty

1.10 Configuration Tool Troubleshooting........................................................................................ 51

Messages from NVRAM Tab Actions An error occurred loading NVRAM data. Warning: NVRAM could not be read, defaults returned. An error occurred updating the NVRAM. Feature bounds checking. Execution Throttle is greater than the maximum allowable value of 255.

Messages from Flash Tab Actions This is not a flash file, or it is corrupt. This HBA is not compatible with the selected flash file. A valid file was not selected. An error occurred reading from the flash file, the file may be corrupt. An error occurred updating the flash. The card has been prepared for firmware updating…

2.0 Drive Performance Testing...................................................................................................... 53

2.1 Windows Only - ATTO Disk Benchmark..................................................................................... 55

Benchmark Fields

Radio Button Group

Multiple Benchmark Testing

2.2 ATTO Disk Benchmark Troubleshooting..................................................................................... 57

Contents

Appendix A - CLI ASCII-Based Interface........................................................................................... i

CLI Error Messages

CLI Summary

CLI Command Explanations

Appendix B - Drive Reformat Instructions..................................................................................... xiv

Mac OS Users’ Instructions

Windows 7/Server 2008/Vista Users’ Instructions

Windows XP/Server 2003 Users’ Instructions

1.0 ATTO Conﬁguration Tool Overview

About the Configuration Tool

The ATTO Conﬁguration Tool is the utility program that displays information about installed controllers, drivers and drives in your Fusion storage system, and provides the means to conﬁgure and manage them.

This program executes under:

• Mac OS

• Windows

• Linux

Note: Java version 1.5 or later must be installed.

The ATTO Configuration Tool displays:

• The name of the Sonnet RAID controller (listed as an ExpressSAS

• Information about the drivers controlling the Sonnet RAID

• Information about Fusion drive enclosures attached to the

• Information about drives within attached Fusion drive enclosures

You may use the Configuration Tool to:

• Manage RAID groups

• Conﬁgure RAID Event notiﬁcations

• Modify the RAID controller’s NVRAM settings

• Revert to default factory settings

• Update the RAID controller’s ﬂash image

• Obtain drive health information

• Obtain drive enclosure health information

The factory settings on your Sonnet RAID controller provide excellent performance for a wide range of applications. However, some applications may beneﬁt from modiﬁcation of the controller’s NVRAM settings that tune the controller for a speciﬁc performance range.

X 10.4 or later

7/Server 2008/Vista®/Server 2003/XP

2.4 and 2.6 kernels, x86 and x64

Rxxx adapter)

controller, including version information for both the currently executing driver and the ﬂash image

Sonnet RAID controller (except D400RAID, D800RAID, and R800RAID) such as fan and power supply status

Configuration Tool Launch

1. Locate the application icon in the folder created during installation.

2. Double-click the ATTO Conﬁguration icon to start the application.

WARNING: Back up system data when installing or

changing hardware configurations.

Note: The Sonnet RAID controller is designed to operate properly using

factory settings. Entering invalid or incorrect NVRAM settings may cause your Sonnet RAID controller to function incorrectly.

The main screen has three panes: Device Listing, Conﬁguration Options and Status.

Configuration Tool Navigation

The Device Listing pane at the left of the window lists all compatible devices (controllers and enclosures) currently connected to the system.

Expand the device tree to reveal additional details on connected devices.

Support Note: In the ATTO Configuration Tool’s Device

Listing pane, the Sonnet RAID controller is identified as an

ExpressSAS Rxxx.

The Conﬁguration Options pane provides information and options for a device highlighted in the device listing.

If you highlight a device in the Device Listing pane, tabs and panes are displayed for that device.

The following chart speciﬁes the tabs that are displayed for the indicated device type:

Tree Node Tab(s) Displayed

Local Host Basic Info, Notiﬁcations, SNMP

Controller Basic Info, Flash, RAID, RAID CLI, Tasks,

Advanced

Channel NVRAM, Basic Info

Device Basic Info, Flash, SES

1.0 ATTO Conﬁguration Tool Overview

Select the Local Host

The following tabs display in the Conﬁguration Options window when you select the local host in the Device Listing window:

• The Basic Info tab displays information about the booted operating system. See Figure 1 on page 3.

• The Notifications tab allows you to set up notiﬁcation of certain events in the Sonnet RAID controller. Refer to RAID Event Notiﬁcations on page 23.

• The SNMP tab allows you to conﬁgure SNMP monitoring and trap generation for the 3 Gb/s Sonnet RAID controller. Refer to SNMP Conﬁguration on page 45.

About window The About window, displayed when About is selected from the Help menu, lists the ATTO Conﬁguration Tool’s version number.

Select the Controller

The following tabs are displayed in the Conﬁguration Options pane when you select the Sonnet RAID controller in the Device Listing pane:

• The Basic Info tab provides information about the Sonnet RAID controller when it is highlighted in the Device Listing pane. You cannot make changes from this screen. See Figure 2 on page 3.

• The Tasks tab provides information about user-scheduled hard drive health maintenance. In this tab, you can reschedule or delete tasks. See Figure 6 on page 5.

• The Advanced tab does not function with the Sonnet RAID controller; clicking this tab merely displays a message.

Select a Channel

The following tabs display in the right pane when you select a speciﬁc channel in the Device Listing pane:

• When you select a speciﬁc channel under the Sonnet RAID controller in the Device Listing pane, the NVRAM tab displays the NVRAM parameters applicable to the Sonnet RAID controller and channel selected. Refer to RAID Controller NVRAM Settings on page 7, and Conﬁguration Tool Troubleshooting on page 51 for information about NVRAM settings.

• The Basic Info tab displays PCI information for the selected channel. See Figure 7 on page 6.

• The current status of the Configuration Tool is represented in the Status pane at the bottom of the window.

About window The About window, displayed when About is selected from the Help menu, lists the ATTO Conﬁguration Tool’s version number.

• The Flash tab provides information about the current ﬂash version programmed on the highlighted controller. See Figure 3 on page 4.

Click the Browse button at the bottom of the tab to search for new ﬂash ﬁles on your system such as FlashBundle_2007_02_27.R380. Once you’ve selected the ﬂash ﬁle, click the Update button to automatically update your Sonnet RAID controller.

• The RAID tab provides information about attached drives, their RAID group and Hot Spare associations, and their operating status. With the RAID tab selected, you can create, modify, and delete RAID groups. You may also schedule or start diagnostic and maintenance procedures, start individual drive performance testing, and send commands to attached enclosures to identify drives. See Figure 4 on page 4.

• The RAID CLI tab provides access to the command line interface, which, as an alternative to application menu-based commands, enables the use of ASCII-based commands to control conﬁguration and diagnostic tasks. See Figure 5 on page 5.

Support Note: The following feature does not work with

Fusion D400RAID, Fusion D800RAID, nor Fusion R800RAID

storage systems.

Select a Device

The following tabs display in the right pane when you select a speciﬁc device in the Device Listing pane:

• The Basic Info tab displays information about the selected enclosure or SAS expander.

• The Flash tab does not function with the Fusion storage systems; clicking this tab merely displays a message.

Support Note: The SES tab appears only when an

enclosure with one or two SAS expanders, such as Fusion

DX800RAID, or Fusion RX1600RAID, is used.

• The SES tab displays SES (SCSI Enclosure Services) status information for SES devices such as power supplies and fans. See Figure 8 on page 6.

1.0 ATTO Conﬁguration Tool Overview

Basic Info displayed when Local Host chosen in the Device Listing pane

Figure 1

Basic Info displayed when the Sonnet RAID controller is chosen in the Device Listing pane

Figure 2

1.0 ATTO Conﬁguration Tool Overview

Flash information displayed when the Sonnet RAID controller is chosen in the Device Listing pane

Figure 3

RAID group and associated drive information displayed when the Sonnet RAID controller is chosen in the Device Listing pane

Figure 4

1.0 ATTO Conﬁguration Tool Overview

RAID CLI displayed when the Sonnet RAID controller is chosen in the Device Listing pane

Figure 5

Scheduled tasks information displayed when the Sonnet RAID controller is chosen in the Device Listing pane

Figure 6

1.0 ATTO Conﬁguration Tool Overview

Basic Info displayed when a SAS expander is selected in the Device Listing pane

Figure 7

SES information displayed when a SAS expander is selected in the Device Listing pane

Figure 8

1.1 RAID Controller NVRAM Settings

The settings in the NVRAM tab vary depending upon the RAID controller and operating system.

Sonnet’s RAID controllers are designed to operate properly using factory settings. Entering invalid or incorrect settings when using an NVRAM conﬁguration utility such as the ATTO Conﬁguration Tool may cause your controller to function incorrectly.

See Figure 9 on page 8 for an example of the NVRAM pane.

WARNING: Back up system data when installing or

changing hardware configurations.

Use caution when making changes to NVRAM settings and only make changes to those with which you are familiar. Once you have made the desired changes, click Commit to save the changes. Click Save to name and save an NVRAM conﬁguration. Click Load to load a saved NVRAM conﬁguration. Changes do not take effect until you reboot the system.

If you do not want to make any changes, you may choose one of the following:

• Defaults: restores the controller to factory default settings. The Commit button must be clicked to save any changes.

• Restore: reverts to the NVRAM settings saved the last time the Commit button was used. Clicking Commit is not necessary.

Support Note: The SAS address is a globally-unique

identifier assigned to devices such as the Sonnet RAID

controller, and is similar to an Ethernet adapter’s MAC address.

SAS Address

Read only

NCQ (Native Command Queueing) Choices: enabled, scan only, disabled Default: disabled

When enabled, the Sonnet RAID controller’s driver sends multiple simultaneous commands to NCQ-capable SATA disk drives. Enabling NCQ may be useful when the Fusion storage system

is used in database applications, but may hinder performance in video editing and other applications.

Device Wait Time Choices: 1–255 seconds Default: 3

Speciﬁes the number of seconds that the driver waits for devices to appear.

Device Wait Count

Choices: 1–255 devices Default: 1

Speciﬁes the number of devices that must appear in order to cancel the Device Wait Time period.

Spinup Delay

Choices: 0-20 seconds Default: 0

Speciﬁes the number of seconds each SAS port waits for disk drives to spin up.

Multiplexing (6 Gb/s RAID Controller Only) Choices: enabled, disabled Default: disabled

Displays the SAS address assigned to the controller. The value cannot be modiﬁed.

Boot Driver Choices: disabled, enabled, scan only Default: disabled

If enabled and disk drives are detected during the bus scan, the BIOS driver remains resident. If disabled, the BIOS starts, resets the controller chip and unloads the driver.

If Scan Only is selected, the BIOS driver scans the bus and displays the devices attached, then unloads itself after a brief delay.

Heartbeat

Choices: enabled, disabled Default: enabled

When enabled, the Sonnet RAID controller’s ﬁrmware is required to respond to periodic activity. If the ﬁrmware does not respond, the system driver resets the ﬁrmware on the controller.

When enabled, multiplexing enables multiple 3 Gb/s devices to aggregate 6 Gb/s SAS bandwidth. In order to utilize this feature, devices must support multiplexing and conform to SAS 2.0 compliancy.

Phy Speed (6 Gb/s RAID Controller Only) Choices: 6 Gb/s, 3 Gb/s, 1.5 Gb/s Default: auto

Enables the user to manually adjust the PHY.

1.1 RAID Controller NVRAM Settings

NVRAM settings information shown when a Fusion drive enclosure is selected in the Device Listing pane

Figure 9

1.2 RAID Group Setup and Management

Support Note: In Fusion RAID storage systems

shipped from Sonnet with hard drives installed, the drives are formatted Mac OS Extended (Journaled), configured as a single RAID 5 or RAID 6 RAID group, and ready for use with Mac OS X-based systems. If you need to

change the configuration, delete the existing RAID group (see RAID Group Deletion on page 13), and use the Configuration Tool and the operating system software tools to reformat and reconfigure the drives. See page xiv for Drive Reformat

Instructions for Mac OS and Windows users.

The ATTO Conﬁguration Tool enables you to conﬁgure disk storage into RAID groups or Hot Spare drives.

Note: Even an individual JBOD disk is considered to be a RAID group.

Use the ATTO Conﬁguration Tool to set up RAID groups on your Sonnet RAID controller in one of the following RAID levels:

• JBOD

• RAID Level 0

• RAID Level 1

• RAID Level 4

• RAID Level 5

• RAID Level 6

• RAID Level 10

• DVRAID™

Support Note: DVRAID is a customized, protected

RAID 4 configuration. It is optimized for increased digital video playback performance when compared to that obtained from a RAID 5 configuration. DVRAID’s write per formance is decreased in order to accomplish this optimization.

DVRAID RAID groups may be set up automatically by the ATTO Conﬁguration Tool. All other RAID conﬁgurations require customized input; Sonnet recommends Custom RAID group setup.

Each RAID group may be divided into one or more partitions; each partition appears to the your computer as a virtual disk.

Preliminary Configuration Steps

1. Launch the ATTO Conﬁguration Tool application.

2. The Conﬁguration Tool main screen appears. In the Device Listing pane on the left side of the window, click ExpressSAS Rxxx under localhost; a new set of tabs appears in the right pane. See Figure 10 on page 14 for an overview.

3. Click the RAID tab; the application scans for drives. Attached drives are displayed in the top pane, while RAID groups and Hot Spares are displayed in the bottom pane.

Support Note: In the ATTO Configuration Tool’s Device

Listing pane, the Sonnet R AID controller is identified as

an ExpressSAS Rxxx.

Custom RAID Group Setup Steps

1. After completing Preliminary Conﬁguration Steps, select RAID Management > Create Group > Customized from the application menu.

2. Select the ﬁrst set of options to conﬁgure the new RAID group. See Figure 11 on page 14.

• Name: name the RAID group or use the one assigned by the

Conﬁguration Tool. The name must be unique, contain no spaces, and contain no more than 14 characters.

Support Note: Two RAID groups with the same name

may not be recognized. If you add another RAID group to your setup, you must make sure it does not have the same name as the existing one, and change it if it does.

• Level: select a RAID group level from the drop-down menu.

Support Note: Descriptions of RAID levels can be found

on the Wikipedia.org Web site at the following addresses: http://en.wikipedia.org/wiki/Standard_R AID_levels and http://en.wikipedia.org/wiki/ Nested_RAID_levels.

Windows Support Note: In order to create RAID

volumes larger than 2TB under Windows, you must do one of the following: Select the 4KB sector size when creating a custom RAID group (not DVRAID) and select Simple Volume as the formatting option in the Disk Management application.

-OR- Use the software configuration tools included with the Fusion storage to create volumes up to 2TB, span (link together in a virtual chain) the volumes, and then format as NTFS. - ORUse GPT formatting. Note that drives and volumes with GPT formatting are not visible to the 32-bit version of Windows XP Professional, nor to the 32-bit version of Windows Server 2003 SP1.

You may use the command line interface pane from the RAID CLI tab in the ATTO Conﬁguration Tool to set up or modify various parameters (Refer to Appendix A). However, using the menu-based procedures listed in this chapter is the preferred method for setting up RAID configurations for the Fusion storage system.

• Interleave: select an interleave value. The default value

is 64KB or 128KB, depending on the OS used. Sonnet recommends 1MB interleave size for maximum video editing performance.

Support Note: The interleave value chosen when

creating a RAID group makes a significant impact on performance. Fusion R AID storage systems shipped from Sonnet with pre-installed hard disk drives are now optimized for use with video editing (larger files) applications, typically with an interleave value of 512KB or 1MB selected. If you intend to use your storage system primarily for storage of smaller files (database, office documents, etc.), choose a smaller interleave value of 64KB or 128KB.

• Mirror Count: select the number of mirror groups when RAID 1 or RAID 10 RAID groups are created.

1.2 RAID Group Setup and Management

Custom RAID Group Setup Steps (continued)

• Initialize: select Advanced or Express; Sonnet recommends choosing Advanced (which is the default).

Support Note: When the Advanced Initialize option is

selected, parity blocks are calculated and the RAID group is thoroughly scanned and subjected to a complete Write/ Verify operation to map out any bad blocks on the drives before the RAID group is made available for use.

When the Express Initialize option is chosen, drives are not scanned and subjected to the Write/Verify operation, but parity blocks are calculated and the RAID group may be used during the initialization.

3. Click Next. Select the drives in the top pane and drag them into the device area in the bottom pane. See Figure 12 on page 15.

4. Click Next. Select the next set of options to conﬁgure the new RAID group. See Figure 13 on page 15.

• Sector Size: select a sector size from the drop down box.

The default is 512 bytes. Use 512 bytes unless you need to use the MBR partition scheme under Windows.

Windows Support Note: Choosing the 4K sector size

enables the creation and use of R AID volumes up to 16TB on systems running Windows XP 32-bit. Otherwise, the volumes are limited to 2TB.

5. If you want the RAID group to be presented as one virtual disk (partition), click Finish. If you want more than one virtual disk (partition), click Next (see Figure 14 on page 16), and then select one of the following options:

• leave as a single partition

• partition by count

• partition by size

If you choose to split the RAID group by count or capacity, you must enter additional information.

6. If you have not already done so, click Finish.

7. A conﬁrmation dialog box asks you to approve the settings you have chosen. Click Yes.

8. Select the RAID group in the Groups pane.

9. Select RAID Management > Properties from the application menu. In the Properties window, change the Prefetch value to 6. See Figure 15 on page 16.

10. Click accept.

11. Click the RAID CLI tab, type “get raidcommandtimeout” and then hit the return key; if the number that appears is 60000, skip to step 13.

• Speed Read: select Always, Adaptive, or Never. The default is Adaptive, but Sonnet recommends Always.

Support Note: For the Speed Read option, select Always

if you expect to work with large sequential files (video, for example), Never if you expect most of the files to be smaller in size (general storage, database, etc.), or Adaptive if you expect mixed use or don’t know.

• Rebuild Priority: select High, Same, or Low. The default is Same.

Support Note: Rebuild priority affects the performance

of your Fusion storage system when a drive is replaced and a degraded RAID group is rebuilt. Selecting Low rebuild priority enables you to continue working at the best performance level possible, but the RAID group will take much longer to rebuild.

• Auto Rebuild: on or off.

Support Note: If the Auto Rebuild option is not checked,

you will have to manually start a RAID group rebuild after replacing a faulted drive.

12. Type “set raidcommandtimeout 60000” and then hit the return key.

13. Every RAID group must ﬁnish initializing, and be formatted by your computer’s operating system software before it becomes available for use; Mac users will use Disk Utility, Windows

users will use Disk Management. For more information on drive formatting, see Mac OS Drive Formatting or Windows Drive Formatting starting on page 11.

1.2 RAID Group Setup and Management

Hot Spares Setup and Usage

If a drive in a parity RAID group becomes degraded or faulted, the RAID group will lose some redundancy until a new member (drive) is rebuilt into it. You can automate this procedure by designating one or more drives as Hot Spares. You may set up a pool of Hot Spare drives of different sizes appropriate for your RAID groups.

Support Note: Hard drives in the Hot Spare pool should

be of appropriate capacity to the RAID group so that

smaller drives are not replaced by much larger Hot Spare drives.

If the Sonnet RAID controller detects a faulted drive in a RAID group with a designated Hot Spare:

• The controller searches the Hot Spare pool for the smallest drive of sufﬁcient capacity to substitute for the faulted drive.

• The faulted drive is replaced with one from the Hot Spare pool.

• The controller begins an automatic rebuild of the RAID group.

Select RAID Management > New Hot Spare (or Delete Hot Spare) from the application menu, and then follow the instructions on the screen.

Mac OS Drive Formatting

1. Depending on how you conﬁgure your setup, a Disk Insertion window stating that there is an unreadable volume will appear at some point during the RAID group creation process; click Initialize, and then Disk Utility will open.

2. In the Disk Utility window, each RAID group you created using the ATTO Conﬁguration Tool will appear as a single volume. Select the volume, and then click the Erase tab at the top of the window.

Support Note for Power Mac G5 Users: When

creating RAID groups 16TB or larger, uncheck the Install Mac OS 9 Drivers checkbox; OS 9 drivers do not support volumes greater than 16TB.

Windows 7/Server 2008/Vista Drive Formatting

1. Click Start, then right-click Computer and select Manage.

2. In the Computer Management window, click Storage in the

left pane to expand the list (if necessary), and then click Disk Management.

3. When the Initialize Disk window appears, select the RAID volume you created. Select the GPT partition style unless you need to access your RAID storage from a computer running 32-bit Windows XP Professional or 32-bit Windows Server

2003. Click OK.

4. In the Disk Management window, each RAID group you created will appear (listed as “unallocated”) as a single volume. Right-click where the word “unallocated” appears, and then select New Simple Volume.

5. When the Welcome to the New Simple Volume Wizard window appears, click Next to start the process.

6. When the New Simple Volume Wizard window appears, click Next.

7. When the Specify Volume Size window appears, click Next if you want all of the Fusion system’s capacity to remain as one block (volume). Otherwise, adjust the volume size to meet your needs, and then click Next.

8. When the Assign Drive Letter or Path window appears, select Assign the following drive letter, choose a letter, and then click Next.

9. When the Format Partition window appears, enter a new name for the volume table if you’d like. For RAID volumes up to 16TB, accept the default allocation unit size; for RAID volumes greater than 16TB, select 8192 from the drop-down menu. Select Perform a quick format, and then click Next.

Note: If you do not select the quick format option, this process will take

much longer to complete.

3. Click the Erase button; a window will appear asking you to approve your choice; click Erase.

4. Repeat steps 2 and 3 for each remaining unformatted RAID group, and then close Disk Utility.

5. Depending on how you conﬁgured the RAID groups, the volumes may already be mounted and present on the desktop. If you created a DVRAID, RAID 4, RAID 5, or RAID 6 RAID group, conﬁguration will take much longer. You may check on the progress by double-clicking the volume name in the lower pane of the ATTO Conﬁguration Tool window.

6. Once all the RAID groups have been formatted and ﬁnish building, they are ready to use.

10. When the next window appears, click Finish.

11. Repeat steps 4–10 for each remaining “unallocated” disk.

12. Depending on how you conﬁgured the RAID groups, the volumes may already be available to the system. If you created a DVRAID, RAID 4, RAID 5, or RAID 6 RAID group, conﬁguration will take much longer. You may check on the progress by double-clicking the volume name in the lower pane of the ATTO Conﬁguration Tool window.

13. Once all the RAID groups have been formatted and ﬁnish

building, they are ready to use.

1.2 RAID Group Setup and Management

Windows XP/Server 2003 Drive Formatting

1. Select Start > Control Panel from the Windows Start menu. In the Control Panel window, double-click Administrative Tools. In the Administrative Tools window, double-click Computer Management.

2. In the Computer Management window, click Storage on the left, and then click Disk Management beneath it.

3. When the Initialize and Convert Disk Wizard window appears, click Cancel.

4. In the Disk Management window, each RAID group you created will appear (listed as “unallocated”) as a single volume. Right-click one volume where the words “Not Initialized” appear, and then select Initialize Disk.

5. When the Initialize Disk window appears, select the RAID volume(s), and then click OK.

6. Back in the Disk Management window, right-click where the word “Online” appears, and then select Convert to GPT Disk.

7. Right-click where the word “unallocated” appears, and then select New Partition.

8. When the New Partition Wizard window appears, click Next.

15. Depending on how you conﬁgured the RAID groups,

the volumes may already be available to use. If you created a DVRAID, RAID Level 4, or RAID Level 5 group, conﬁguration will take much longer. You may check on the progress by double-clicking the RAID group name in the lower pane of the ATTO Conﬁguration Tool window.

16. Once all the RAID groups have been formatted and ﬁnish building, they are ready to use.

RAID Group Management Overview

The ATTO Conﬁguration Tool interface may be used to manage the replacement of a failed drive, add capacity to a RAID group, change a RAID group’s current RAID level conﬁguration to a new one, and change a RAID group’s properties.

WARNING: Data can be compromised or lost when

deleting storage or rearranging storage configurations.

The ATTO Conﬁguration Tool interface guides you step by step through many procedures which allow you to modify your storage and RAID conﬁgurations. Read all support notes and warnings carefully as you go to ensure the best performance and use of your storage. Many of these procedures may only be performed on drives that are not currently part of a RAID group, are not designated as a Hot Spare, or were ofﬂine when you initially set up RAID conﬁgurations.

9. When the Select Partition Type window appears, select Primary Partition, and then click Next.

10. When the Specify Partition Size window appears, click Next if you want all of the Fusion system’s capacity to remain as one block (volume). Otherwise, adjust the volume size to meet your needs, and then click Next.

11. When the Assign Drive Letter or Path window appears, choose a letter, and then click Next.

12. When the Format Partition window appears, enter a new name for the volume if you’d like. For RAID volumes up to 16TB, accept the default allocation unit size; for RAID volumes greater than 16TB, select 8192 from the drop-down menu. Select Perform a quick format, and then click Next.

Note: If you do not select the quick format option, this process will take

much longer to complete.

13. When the next window appears, click Finish.

14. Repeat steps 4–13 for each remaining “unallocated” disk.

Support Note: An unallocated drive or unallocated

storage is storage which is not part of a RAID group, not designated as a Hot Spare or was offline when you initially set up a RAID configuration using the ATTO Configuration Tool interface.

RAID Group Capacity Expansion

Select RAID Management > Expand Capacity from the application menu, and then follow the onscreen instructions. Depending on the RAID conﬁguration, you may need to add more than one drive at a time.

Mac User’s Support Note: Although this feature is

supported by the ATTO utility, as of this writing, Mac OS X does not support RAID group capacity expansion.

WARNING: Adding drives to an existing RAID group may

adversely impact performance. You cannot reverse this

operation unless you delete the R AID group.

1.2 RAID Group Setup and Management

RAID Group RAID Level Migration

Changing a RAID group from one RAID level to another is called migration. The following migration levels are supported:

• JBOD to RAID Level 0

• JBOD to RAID Level 1

• RAID Level 0 to RAID Level 10

• RAID Level 1 to RAID 10

Select RAID Management > Migrate RAID Level from the application menu, and then follow the instructions on the screen.

RAID Group Deletion

You may delete a group using the ATTO Conﬁguration Tool. Select RAID Management > Delete Group from the application menu, and then follow the instructions on the screen.

WARNING: Data can be compromised or lost when

deleting storage or rearranging storage configurations.

RAID Group Rebuilding

If a RAID group becomes compromised, you must rebuild it. Select RAID Management > Rebuild Group from the application menu, and then follow the instructions on the screen.

Support Note: A RAID group rebuild may take several

hours to complete, depending on the operating system,

drive capacities, and RAID configuration.

You may pause a RAID group rebuild by selecting the RAID group in the lower pane, and then selecting RAID Management > Pause Rebuild from the application menu. To resume the rebuild, select the RAID group in the bottom pane, and then select RAID Management > Resume Rebuild from the application menu.

RAID Group Properties Modification

Each RAID group has speciﬁc properties, and the value of each property remains with the RAID group when it is moved from one computer to another. Some of the properties can only be speciﬁed during RAID group creation (RAID level, interleave, and sector size), whereas others may be changed at any time during the life of the RAID group. See Figure 15 on page 16.

2. Select RAID Management > Properties from the application menu, and then view or change the current properties:

• Speed Read speciﬁes the cache policy to be used during

read operations. Once a read command is given, the ExpressSAS RAID code retrieves the next set of sequential data from the RAID group’s drives and caches it in the Sonnet RAID controller’s internal memory. If you select Never, read caching is never performed. If you select Always, read caching is always performed. If you select Adaptive, Speed Read is enabled or disabled depending on the sequential patterns detected in I/O requests.

• Auto Rebuild controls the replacement of a faulted

drive with any available unallocated drive. When you click the Auto Rebuild check box and the Accept button, Auto Rebuild is enabled. If a drive becomes faulted, the ExpressSAS RAID adapter replaces the drive with an unallocated drive.

• Rebuild Priority speciﬁes the ratio of rebuild I/O activity

to host I/O activity. A rebuild priority of Same (default value) indicates that rebuild I/O and host I/O are treated equally. A rebuild priority of Low indicates that host I/O is given a higher priority than rebuild I/O. A rebuild priority of High indicates that rebuild I/O is given a higher priority than host I/O. In practical terms, selecting Low enables you to continue working with full read performance, but the rebuild will take much more time to complete.

• Prefetch speciﬁes the number of stripes that are read when

Speed Read is enabled or set to adaptive. The valid values for prefetch are 0, 1, 2, 3, 4, 5 and 6; the default value is 1. This property can only be changed after the RAID group is created. To access this property, select the RAID group and view its properties.

3. Click Accept.

1. Select a RAID group in the Groups pane.

1.2 RAID Group Setup and Management

Configuration Tool main screen with the RAID tab selected

Figure 10

Options selection screen displayed when creating new, custom RAID groups

Figure 11

1.2 RAID Group Setup and Management

Drives selected to create a new RAID group

Figure 12

Additional options selection screen displayed when creating new, custom RAID groups

Figure 13

1.2 RAID Group Setup and Management

Select the number of partitions for the new RAID group

Select the RAID group’s properties

Figure 14

Figure 15

1.3 Drive and RAID Group Monitoring

The ATTO Conﬁguration Tool provides useful information on individual drives and associated RAID groups.

Use the ATTO Conﬁguration Tool to gather basic or detailed information about the drives connected to the Sonnet RAID controller, and operational status on the RAID groups created with them.

Basic Drive Information

Open the ATTO Conﬁguration Tool, expand the device tree in the Device Listing pane until ExpressSAS Rxxx appears, and then click to highlight it. In the Attached Drives pane, general information for all the drives is displayed. See Figure 16 on page 20.

• LED icon: Indicates operational status of the drives. Green = online, red = faulted

• Name: Displays the drive’s model number

• Vendor: Not used

• Address: Displays the SAS address generated by the Sonnet RAID controller

• Usage: Identiﬁes how the drive is being used. If it is part of a RAID group, the group name and member number are displayed. If it is a Hot Spare, it is listed as a Hot Spare.

• Capacity: Displays the drive’s formatted capacity.

Detailed Drive Information

In the Attached Drives pane, double-click a drive name to view detailed information. See Figure 17 on page 20.

• Index: Displays the drive’s RAID group index number. This number is used in CLI commands

• Capacity: Displays the drive’s formatted capacity

• Usage: Identiﬁes how the drive is being used. If it is part of a RAID group, the group name and member number are displayed. If it is a Hot Spare, it is listed as a Hot Spare.

• Sector Size: Displays the drive’s sector size

RAID Group Information

In the bottom pane, click the Groups tab to display RAID groups. See Figure 16 on page 19.

• Name: Displays the name of the RAID group

• Level: Indicates the RAID level formatting for the RAID group

• Capacity: Indicates the formatted, conﬁgured capacity of the RAID group

• Members: Indicates the number of drives in the RAID group

• Status: Displays the operating status for the RAID group. Refer to Deﬁnitions on page 35 for detailed information.

• Rebuild: Speciﬁes the general condition of the RAID group. None indicates no rebuild is taking place, nor is it necessary; Rebuilding indicates that the RAID group is degraded, and is in the process of rebuilding; Paused indicates that a rebuild was interrupted and needs to be restarted to ﬁnish. Refer to Deﬁnitions on page 35 for detailed information.

• Status: Displays the drive’s operating status. OK is displayed if it is functioning normally. If there is a problem, Faulted or Error is displayed.

• Type: Displays the type of media

• Name: Displays the drive’s model number

• Vendor: Not used; always displays Not Available

• Serial: Displays the drive’s serial number

• Address: Displays the SAS address generated by the Sonnet RAID controller

• Speed: Displays the drive’s interface speed

• Revision: Displays the drive’s ﬁrmware revision

• LUN: Displays the logical unit number, which is the number assigned to drive’s RAID group

Drive Identification

You may identify one or more drives using the ATTO Conﬁguration Tool to turn on LEDs in the Fusion drive enclosure.

1. Launch the ATTO Conﬁguration Tool application.

2. Expand the device tree to highlight the ExpressSAS Rxxx, and then click the RAID tab. Status for all drives connected to the Sonnet RAID controller will be displayed.

3. Click on one or more drives individually that you want to identify in the Attached Drives list.

4. Select RAID Management > Locate > Drive from the application menu. If the drive does not support this method of identiﬁcation, a message will appear in the bottom pane; go to the next step. Otherwise, look at the Fusion enclosure; the drive activity LED(s) for the speciﬁc drive(s) will be lit until you deselect Locate in the application menu.

(continued)

1.3 Drive and RAID Group Monitoring

Drive Identification (continued)

5. Double-click a drive in the top pane to display detailed information, and note the index number for the drive. Close the detailed drive information window.

6. Click the RAID CLI tab, and then type “Blockdevidentify x”, where x is the index number. Look at the Fusion enclosure; the drive activity LED for the speciﬁc drive will be lit.

Note: Type “Blockdevidstop” to turn off the LED.

S.M.A.R.T. Data

Self-Monitoring, Analysis and Reporting Technology, or S.M.A.R.T., is a monitoring system built into SATA drives to detect and report on various indicators of drive health. The S.M.A.R.T. feature keeps track of and reports on the status of SATA drive health using certain parameters recorded by the drives. Notiﬁcations can be sent when certain pre-determined values are exceeded.

Use the ATTO Conﬁguration Tool to view the ﬁles that record changes to an individual drive’s S.M.A.R.T. parameters. The ﬁles are permanent and can be viewed independently whether you have enabled monitoring or not.

S.M.A.R.T. Monitoring Enabling and Disabling

Monitoring is disabled by default; if you want to use the feature, you must enable it. You may enable or disable the monitoring feature at any time.

1. Launch the ATTO Conﬁguration Tool application.

3. The S.M.A.R.T. Status box displays.

If there has been a change from a previous S.M.A.R.T. status record, an arrow indicates the change direction, either higher or lower. See Figure 18 on page 21.

The S.M.A.R.T. status display also contains information such as the date and time the S.M.A.R.T. status was recorded, the total number of records for this drive, and the current monitoring status (enabled or disabled).

You may move to previous or subsequent records, query the drive or refresh the view using controls on the interface. Control-click (or right-click) a single drive in the Attached Drives pane, and select S.M.A.R.T. Status from the sub-menu to view the record.

• Use the left arrow or right arrow control to move between S.M.A.R.T. status records.

• Use the Refresh button to query the drive for the latest values. If any values are different from the most recent record, a new record is created and displayed.

S.M.A.R.T. Attribute Filtering

Each of the S.M.A.R.T. status attributes is assigned one or more classiﬁcation types:

• performance

• error rate

• event count

• critical

2. Expand the device tree and select the ExpressSAS Rxxx, and then click the RAID tab.

3. Select RAID Management > Monitor S.M.A.R.T. from the application menu to enable (indicated with a check mark) or disable monitoring.

S.M.A.R.T. Status Checking

The ATTO Conﬁguration Tool interface displays the latest S.M.A.R.T. status record for a selected drive. All attributes reported by the drive are listed with each attribute’s Threshold, Worst, Current and Raw value; the threshold value is the value at which notiﬁcation of a problem is generated by the software.

1. Select a single drive in the Attached Drives pane.

2. Control-click or right-click on the selected drive, and then select S.M.A.R.T. Status in the sub-menu.

The S.M.A.R.T. Status dialog box can be ﬁltered to display any combination of these types. The default view is to display all types.

1. Open the S.M.A.R.T. Status box, and then control-click (or right-click) in the table area where the attribute values are displayed.

2. Each classiﬁcation type that is visible has a check mark. Select any classiﬁcation type to change the check mark.

1.3 Drive and RAID Group Monitoring

S.M.A.R.T. Notifications

When S.M.A.R.T. monitoring is enabled, status is collected from each SATA drive at 60 minute intervals. If the data is different than the previous status, a S.M.A.R.T. status record is added to the S.M.A.R.T. status ﬁle for that drive. A notiﬁcation of the S.M.A.R.T. status difference is generated based upon the current settings in the Notiﬁcations pane. Refer to RAID Event Notiﬁcations on page 25.

The S.M.A.R.T. status’ notiﬁcation level is determined as follows:

• INFO: None of the status values was below the threshold value.

• WARNING: One or more of the status values was below a threshold value but none was classiﬁed as critical.

• CRITICAL: One or more of the status values was below a threshold value and one was classiﬁed critical.

1.3 Drive and RAID Group Monitoring

Drive information displayed with the RAID tab selected

Detailed information displayed for a specific drive

Figure 16

Figure 17

1.3 Drive and RAID Group Monitoring

S.M.A.R.T. status displayed for a specific drive

Figure 18

1.4 Enclosure (SES) Health Monitoring

Support Note: This feature does not work with Fusion

D400RAID, Fusion D800RAID, nor Fusion R800RAID

storage systems.

Many Fusion drive enclosures contain a SCSI enclosure processor which tracks enclosure health status, drive identiﬁcation and drive fault identiﬁcation. The ATTO Conﬁguration Tool recognizes drive enclosures that provide SCSI Enclosure Services (SES).

Use the ATTO Conﬁguration Tool to gather the health status of the enclosure’s power supplies and fans. If the status of either of these sub-systems indicates a failure, the controller reports the problem. The Conﬁguration Tool shows the status of selected SES devices and reports the speciﬁc health of each sub-system.

SES Status Checking

1. Select the Fusion drive enclosure from the Device Listing

tree in the Conﬁguration Tool. See Figure 19 on page 24.

2. Select the SES tab at the top of the right pane.

3. View the overall status of each component across the top of

the right pane. See Figure 20 on page 24.

Note: Depending on the Fusion storage system used, some features may

not be fully supported.

4. Select a speciﬁc sub-system (power supply or fans) and view

the status of the reporting sub-system.

1.4 Enclosure (SES) Health Monitoring

Basic info information displayed when a SAS expander is selected

Figure 19

Power supply status displayed when the SES (SCSI Enclosure Services) tab is selected

Figure 20

1.5 RAID Event Notiﬁcations

The ATTO Conﬁguration Tool may be conﬁgured to issue notiﬁcations via audible and visual alerts when a RAID event occurs.

RAID events are divided into three categories:

• Critical events are ones in which a serious problem has occurred and the administrator of the RAID group should perform corrective action.

• Warning events are less serious but still warrant recording and notiﬁcation at some level.

• Information alerts provide supportive information about warnings or critical events.

Drop-down boxes on the Notiﬁcations pane allow you to choose the type of event which prompts an alert. See Figure 21 on page 26.

• Critical: only Critical events are reported

• Warning: all Warnings and Critical events are reported

• All: all Critical, Warning and Information events are reported

• None: no event is reported. The None level is useful in email

notiﬁcation because you can set up email addresses to which alerts might be sent at some future time.

You can choose any combination of notiﬁcations on the Notiﬁcation pane as needed. The notiﬁcations are speciﬁed at the host system level and apply to all Sonnet RAID controllers installed in the host system.

Logging

Logging notiﬁcation records the type of event as text in a log ﬁle you specify.

• The ATTO Conﬁguration Tool provides a default location for event log ﬁles which you can change in the Location ﬁeld found under Logging on the Notiﬁcations tab. The log ﬁle name is a combination of the adapter’s SAS address, an underscore and a 0 or 1.

• You may limit the size of the log ﬁle by entering a number greater than 0 in the Size Limit (KB) ﬁeld. Once the limit is reached, another log ﬁle is created. Once that log ﬁle’s limit is reached, the Conﬁguration Tool overwrites the ﬁrst log ﬁle and continues to rotate between the two ﬁles.

• If you do not want to limit the storage capability of the log ﬁle, enter a zero in the ﬁeld.

• Choose the type of event you want recorded in the event log from the Events: drop down box.

Email Alerts

Email notiﬁcation sends a message to designated email addresses when the event level from the drop down box next to the Notiﬁcation Address: ﬁeld is reached.

• You may specify several notiﬁcation addresses on each line in the email section of the Notiﬁcations pane, each separated by commas, for any event level.

• You must complete the IP address or name of the server and sender.

Basic Alerts

You can select an audible alert, a visual alert, or both for a particular category of events. Select a notiﬁcation level using the drop-down box next to the Audible and Visual labels on the Notiﬁcations screen.

Audible alert uses either the Sonnet RAID controller’s onboard buzzer (R6xx) or the computer’s speaker (R3xx) to sound an alarm. The alarm continuously sounds until you stop it. There is an icon in the system tray or the menu bar that is used to turn off the alarm. You must right-click (or option-click)the icon and then select Mute Audible Alert from the menu; this will stop the alarm.

Visual alert uses a system modal pop-up to display a message. You must close the pop-up using the pop-up’s button.

Support Note: Audible and visual alerts are not available

on systems running Linux.

• You may specify a user name and password for the mail server if one is required.

• A critical event email notiﬁcation is sent after a 10-second delay to allow several related events to be reported in the same message. All other notiﬁcation emails are sent at 15-minute intervals.

• You may specify if you are using a TLS/SSL enabled server (e.g.

Note: Contact your email provider for the appropriate Port Number.

™

Gmail

, Yahoo®, etc.) as well as the port used on that TLS/ SSL server. Enabling SSL enables you to connect to these email providers, as they usually require secure connections. Leave this box unchecked if you are using a server that doesn’t require a secure connection.

Choices: Enable SSL control = checked, unchecked, Port number control = 1–65535 Default values: Enable SSL control = unchecked, Port number control = 25

Using any other number will result in email notiﬁcation failure.

1.5 RAID Event Notiﬁcations

Configuration Tool Notifications screen when the local host is selected in the Device Listing pane

Figure 21

1.6 RAID Group Media Maintenance

The Media Scan feature scans disk drives for media errors and parity errors. All media errors are counted and ﬁxed. All parity errors are reported in the event log. The two options are described below.

Use the ATTO Conﬁguration Tool to check and repair media errors on parity RAID groups. You may start the media scan operations manually, or schedule the operations for a later time, or on a weekly basis.

Media Scan Options

Media Scan works with parity RAID group members, Hot Spare drives and unallocated drives. Media Scan reads the selected drives and, if a media error is found, Media Scan re-writes the disk with the media error. The drive will relocate the bad sector’s data to an alternate part of the drive. Media Scan rewrites the correct data for a disk that is a member of an online parity RAID group. Media Scan writes random data to Hot Spare drives and unallocated drives. The Media Scan feature records the number of media errors detected and corrected for each drive. These counts are stored persistently if the drive is a RAID group member or a Hot Spare drive. The counts are not persistently stored for any other drives.

Media Scan with Parity Verify is a variation of Media Scan that is available for online parity RAID groups only. Parity Veriﬁcation is performed on each RAID group stripe that has no detected media errors. The parity of the stripe is recalculated and compared to the original parity for the stripe. If there is a mismatch, an error is generated and recorded in the Event Log and System Log ﬁles.

Support Note: Media Scan operations significantly

impact your Fusion storage system’s performance, and may take several hours to complete. We suggest that you schedule Media Scan operations for times when the system in not in use.

Start a Media Scan

1. If your computer is running Mac OS X, drag the RAID group you want to scan to the trash (changes to an eject icon) to eject it. Otherwise, skip to step 2.

2. Select the RAID group, Hot Spare drive(s), or unallocated drive(s) you want to scan in the Groups pane; one RAID group, or multiple Hot Spare or unallocated drives may be selected.

3. Control-click (or right-click) the RAID group and select Scan Only or Parity Verify (Media Scan + Parity Verify) from the drop menu. See Figure 22 on page 29. You may also select RAID Management > Scanning > Scan Only (or RAID Management > Scanning > Parity Verify) from the application menu.

4. The scan starts immediately; the Attached Drives pane displays a rotating icon next to each drive being scanned, and the RAID Group pane displays the type of scan being performed and a status for percent complete.

Schedule a Media Scan

1. Select a RAID group in the Groups pane.

Note: Your computer must be on in order to perform manual or

scheduled Media Scan operations.

Modes of operation for:

• Online RAID Group – Media Scan performs SCSI Read commands on each stripe group. Parity veriﬁcation is performed on a stripe that has no media errors. Media Scan activity is scheduled in accordance with the Rebuild Priority Level. Media Scan is restarted after reboot if the Media Scan did not complete.

• Degraded or Ofﬂine RAID Group – Media Scan performs SCSI Read commands for each stripe group of the online drives. No parity veriﬁcation occurs. The Media Scan is restarted after a reboot if the Media Scan did not complete.

• Hot Spares & Unallocated Drives – Media Scan performs SCSI Read commands for each selected drive. Media Scan is not restarted after a reboot if the Media Scan did not complete.

2. Control-click (or right-click) the RAID group and select Schedule. See Figure 23 on page 30. You may also select RAID Management > Schedule from the application menu.

3. Select Scan Only or Parity Verify, and then click Next. See Figure 24 on page 30.

4. Select the time and frequency schedule for the task to be performed. See Figure 25 on page 31.

5. Click Finish, and then click Yes in the next dialog box to ﬁnish scheduling the task.

Note: You may schedule only one task per RAID group. You may not

schedule media scans for Hot Spare drives or unallocated drives.

1.6 RAID Group Media Maintenance

Cancel, Pause, or Resume a Media Scan

Media Scan functions for a parity RAID group may be paused, resumed and cancelled. The Conﬁguration Tool provides menu items to pause and resume, and CLI provides a command to cancel the Media Scan. The results of the Media Scan up to the time of the pause or cancel are saved persistently.

Media Scan functions for Hot Spare and unallocated drives can be cancelled, but cannot be paused. Media Scan of Hot Plug and unallocated drives are run immediately and they are not scheduled. The Conﬁguration Tool provides menu items to start and cancel Media Scan functions on these drives.

1. Select the RAID group being scanned in the groups pane.

2. Control-click (or right-click) the RAID group and select

Pause Media (or Parity) Scan to pause the Media Scan, or Cancel to stop the Media Scan. You may also select RAID

Management > Pause Media (or Parity) Scan, or RAID Management > Cancel from the application menu.

Note: You may resume a paused Media Scan operation by control-

clicking (or right-clicking) the RAID group that was being scanned and selecting Resume Media Scan, or by selecting RAID Management > Resume from the application menu.

Manage Scheduled Media Scans

1. Click the Tasks tab.

2. Select the RAID group scheduled for a Media Scan.

3. Control-click (or right-click) the RAID group and select Reschedule to change the Media Scan schedule, or Delete to delete the Media Scan. You may also select Task Management > Reschedule, or Task Management > Delete from the application menu. See Figure 26 on page 31.

Automatic Media Scan Cancellation

A Media Scan may be automatically cancelled for the following reasons:

• A parity RAID group being scanned requires a rebuild operation

View a Scan Report

1. Select the Sonnet RAID controller from the Device Listing pane.

2. Select the RAID tab in the right pane.

3. Select the RAID group, Hot Spare drive(s) or unallocated drive(s) whose scan report should be displayed. One RAID group can be selected or multiple Hot Spare and unallocated drives can be selected.

4. Select RAID Management > Scan Report from the application menu.

5. A Media Scan Error Report message box appears with the scan results. See Figure 27 on page 31. If a scan is currently in process, the Scanned column indicates the percent complete for the operation.

The Scan Report includes the counts for the most recent scan and the total results for all Media Scan operations. This information is persistent for RAID group drives and Hot Spare drives since it is stored in the drive’s metadata. The results may be erased from the drive when the metadata is rewritten, e.g. a Hot Spare’s information is lost when the Hot Spare is added to a RAID group. The Media Scan results for an unallocated drive are only available until a reboot occurs.

Media Scan provides the following metrics:

• Media errors detected on most recent scan in the Errors

column.

• Media errors corrected on most recent scan.

• Total stripe groups scanned on most recent scan. This data is

not displayed in Conﬁguration Tool.

• Media errors detected on all scans in the Total Errors column.

• Media errors corrected on all scans.

• A parity RAID group is being scanned and the last drive is marked “Faulted”

• A Hot Spare being scanned is needed for a RAID group rebuild

• An unallocated drive being scanned is needed for a RAID group rebuild

A Media Scan is not cancelled in the following scenario:

• An unallocated drive cannot be added to a RAID group when a Media Scan is active on the drive.

The Media Scan must be manually cancelled in the above scenario.

• Total number of all scans in the Scans column.

The Media Scan Error Report does not display the count of corrected errors. The corrected error count is displayed while the mouse pointer hovers over the Errors or Total Errors value for a speciﬁc drive. If the detected errors do not equal the corrected errors the warning icon appears to the left of device name column.

The Scan Report is updated by the Conﬁguration Tool every 10 seconds. The metrics are not updated when the Media Scan is paused.

1.6 RAID Group Media Maintenance

Selecting the Scan Only media scan operation for a RAID group

Scheduling a Media Scan operation for a RAID group

Figure 22

Figure 23

1.6 RAID Group Media Maintenance

Selecting the Media Scan type

Scheduling the time and frequency for a Media Scan operation

Figure 24

Figure 25

1.6 RAID Group Media Maintenance

Changing a scheduled Media Scan operation

Viewing a RAID group’s Media Scan report

Figure 26

Figure 27

1.7 Identify and Replace a Faulted Drive

If a drive in a RAID group fails, the RAID group’s status becomes degraded. This section will help you to identify and replace the bad drive.

Automatic Faulted Drive Identification

On supported Fusion storage systems, drive fault identiﬁcation is performed automatically by the Sonnet RAID controller when a RAID group member becomes degraded by exhibiting unrecoverable errors during I/O. The RAID controller reports the status of the drive and asks the Fusion enclosure to perform fault identiﬁcation, illuminating a blinking red LED as a result. The fault identiﬁcation continues until the drive is replaced or the RAID group is deleted. The ATTO Conﬁguration Tool will also issue visual, audible, and email notiﬁcations (only when conﬁgured to do so).

Support Note: Automatic faulted drive identification

does not work with Fusion D400RAID, Fusion D800RAID,

nor Fusion R800RAID storage systems.

Manual Faulted Drive Identification

When an error occurs that requires a drive to be replaced, the ATTO Conﬁguration Tool will issue visual, audible, and email notiﬁcations (only when conﬁgured to do so).

6. Click the RAID CLI tab, and then type “Blockdevidentify x”, where x is the index number. Look at the Fusion enclosure; the drive activity LED for the speciﬁc drive will be lit.

Note: Type “Blockdevidstop” to turn off the LED.

Faulted Drive Replacement with Auto Rebuild

A faulted drive is automatically replaced if Auto Rebuild is enabled and a suitable unallocated disk is available. Suitable unallocated drives are initialized, have a large enough capacity to replace the degraded drive, and cannot contain any RAID group information. The unallocated drive may be a pre-existing drive or a newly-installed drive.

If a Hot Spare Pool exists, the Sonnet RAID controller chooses a suitable Hot Spare drive before selecting an unallocated drive.

Manual Faulted Drive Replacement

Once you have identiﬁed the faulted drive, you must replace it and rebuild the affected RAID group.

1. Swap out the faulted drive.

2. Launch the ATTO Conﬁguration Tool application.

After a drive failure notiﬁcation has appeared,

1. Launch the ATTO Conﬁguration Tool application.

2. Expand the device tree to show the ExpressSAS Rxxx, and then click the RAID tab. Drive status for all drives connected to the Sonnet RAID controller will be displayed. The faulted or degraded drive will have a red LED icon next to it.

Support Note: If you have configured your setup to

include a Hot Spare drive, the ATTO Configuration Tool will automatically start rebuilding the RAID group using the Hot Spare drive.

3. Click on the faulted or degraded drive you want to identify in the Attached Drives list.

4. Select RAID Management > Locate > Drive from the application menu. If the drive is not supported for this method of identiﬁcation, a message will appear in the bottom pane; go to the next step. Otherwise, look at the Fusion enclosure; the drive activity LED for the speciﬁc drive will be lit until you deselect Locate.

5. Double-click the faulted or degraded drive in the top pane to display detailed information, and note the index number for the drive. Close the detailed drive information window.

3. Expand the device tree to show the ExpressSAS Rxxx, and then click to highlight the degraded RAID group.

4. Select RAID Management > Rebuild from the application menu; a tab for the RAID group will open, and you will be prompted to drag a free drive on top of the one being replaced. See Figure 28 on page 34.

5. After starting the rebuild, you may use the RAID group, but its write performance will be reduced until the rebuild is complete.

Support Note: A RAID group rebuild will take several

hours to complete, the total time depending on the rebuild priority, drive capacities, RAID group configuration, and operating system used.

1.7 Identify and Replace a Faulted Drive

Replacing a faulted drive with an unallocated drive to rebuild a degraded RAID group

Figure 28

1.8 Recover Data from Ofﬂine RAID Groups

Sonnet RAID controllers use ATTO Technologies’ DriveAssure™ technology to maintain error-free operation of RAID group member drives. This includes features like Drive Initialization, Media Scan, Parity Verify, Performance Monitoring, and algorithms for latency and response management. Occasionally, a drive will suffer a mechanical or fatal media error that makes it unusable. A connection error may cause multiple drives to be inaccessible, resulting in an array that cannot be accessed normally.

This chapter describes procedures to follow that may allow data to be partially recovered when drives fail or become inaccessible and cause a RAID group to go ofﬂine.

WARNING: Anytime a R AID group goes offline, its

data integrity has been compromised. It is imperative

that data consistency checks and/or file system repair tools be used to validate recovered data.

Definitions

RAID Group Status Definition

ONLINE The RAID group state is online, with all RAID group members available and fully operational.

DEGRADED The RAID group does not have full data protection. This occurs when a non-critical number of RAID

group members (drives) are unavailable, degraded or faulted (i.e., one member in a RAID 5 RAID group, two in a RAID 6 RAID group, and so forth). On a read command to inaccessible drive(s), the data is derived using redundancy or parity. A degraded RAID group may be initializing or rebuilding.

OFFLINE The RAID group’s data cannot be accessed because a critical number or drives have been faulted or

removed. For example, two or more drives in a RAID 5 RAID group have failed.

RECOVERY The RAID group is in basic data recovery mode. I/O may be limited to reads while in this state.

REC EXT The RAID group is in extreme data recovery mode, which returns data at all costs. I/O may be

limited to reads while in this state.

RAID Member Status Values Definition

ONLINE The drive is fully operational

DEGRADED Some of the drive’s data is not up to date

UNAVAILABLE Connection to the drive has been lost

FAULTED The drive has an unrecoverable error

Rebuild Status Values Definition

OK Data is up to date

REBLD The rebuild operation is in progress

A_INIT The advanced initialization operation is in progress

X_INIT The express initialization operation is in progress

MSCAN The media scan operation is in progress

Rebuild Status Suffixes Definition

…FAULTED The operation ceased due to an error

…HALTED The operation was stopped at the user request

…INTRUPTD The operation was interrupted

…% COMPL Percent completion for operation

1.8 Recover Data from Ofﬂine RAID Groups

RAID Group Failure Scenarios

RAID groups cannot be accessed normally when their member disks fail, and the RAID group is marked ofﬂine. RAID groups of different RAID levels are marked ofﬂine for different reasons, as follows:

RAID Level Reason(s) for Being Marked OFFLINE Recovery Method

JBOD and RAID 1 Any drive failure See Recovery from Faults on Critical Number of Drives on

page 38

Error during rebuild See Recovery from Failed Rebuild on page 37

RAID 1 and RAID 10

RAID 4 and RAID 5

RAID 6

Mistaken replacement of a good drive when its mirror has failed

Errors on two or more drives See Recovery from Faults on Critical Number of Drives on

Error during rebuild See Recovery from Failed Rebuild on page 37

Mistaken replacement of a good drive when another member of the RAID group has failed

Errors on three or more drives See Recovery from Faults on Critical Number of Drives on

Error during rebuild See Recovery from Failed Rebuild on page 37

Mistaken replacement of good drive(s) when another member of the RAID group has failed

See Recovery from Replacement of the Wrong Drive on page 39

page 38

See Recovery from Replacement of the Wrong Drive on page 39

page 38

See Recovery from Replacement of the Wrong Drive on page 39

Drive Replacement on a Failure Condition

Replace RAID Group Member Drives as Soon as They Fail

With parity and redundancy RAID levels, a RAID group can withstand the loss of one member, and the data is still valid and accessible. In this case, the RAID group goes into degraded mode and uses parity or redundancy to generate the data. Although the RAID group is fully operational, it is at risk because if any other drive fails, data integrity is called into question.

A Warning About Drive Replacement

A very common reason that an array goes from degraded mode to ofﬂine mode is when the wrong drive is replaced. By pulling out a perfectly good drive, a double-drive fault occurs and there are insufﬁcient drives to generate data. The following procedure is very important when you are considering removing a failed drive, to ensure the correct drive is pulled.

Identifying Failed Drives

Prior to replacing a drive, you must be very sure which one failed. If a failed drive is in an enclosure that supports SES (Fusion DX800RAID, RX1600RAID, RX1600 Expansion), the drive module's fault LED should be blinking. In that case, it is clear which drive should be replaced. If multiple drive modules’ LEDs are blinking, power cycling the enclosure(s) and reseating the drives can sometimes correct intermittent conditions.

Recovery Mode

Sometimes, despite careful operation and maintenance, drives will coincidentally fail in such a way that the RAID group integrity is compromised. After a RAID group has been marked ofﬂine because of problems with member drives, there is a way to possibly recover some of the data. The guidelines and commands listed on the following pages of this chapter can help recover data from an ofﬂine RAID group. The following descriptions refer to RAID 5 speciﬁcally, but the principles also extend to other RAID types.

The ATTO Conﬁguration Tool provides other methods to identify failed drives. Please refer to Identify and Replace a Faulted Drive on page 33 for details.

1.8 Recover Data from Ofﬂine RAID Groups

Recovery from a Failed Rebuild

Occasionally during a RAID group rebuild, one of the drives that was not replaced encounters a media error, which causes the rebuild to stop and forces the group ofﬂine. The following diagrams illustrate a scenario where the failed drive is removed and replaced. You may force the rebuild to continue. The error recovery procedure is also valid if there is a Hot Spare, or if there is an unallocated drive and Auto Rebuild is enabled.

A functioning RAID 5 RAID group:

321 4

encounters a drive failure, forcing the RAID group into a

degraded state:

321 4

The failed drive is either replaced with a Hot Spare or

unallocated drive, or is removed and replaced:

Barring major problems, the RAID group will be put back online at the end of the rebuild recovery. However, at the location corresponding to the media error, the data may be invalid!

Anytime a RAID group goes offline, data integrity has been compromised! It is imperative that data consistency checks

and ﬁle system repair tools are used to validate the data before trying to use it. See File System Repair Tools on page 39 for more information.

Recovery from a Failed Rebuild with a Second Unrecoverable Drive Failure

If a rebuild cannot continue to completion because of unrecoverable errors on multiple drives, use the Extreme recovery mode, described below, to try to recover data.

A functioning RAID 5 RAID group:

321 4

encounters a drive failure, forcing the RAID group into a

degraded state:

321 4A

If a media error is encountered during the rebuild process, the

rebuild is halted and the RAID group is forced ofﬂine:

321 4A

Use the ATTO Conﬁguration Tool to force the RAID group to continue the rebuild. First, display the RAID group information. See Figure 29 on page 43. Then right-click the ofﬂine RAID group that has the rebuild status rebuilding (faulted). Select the Start Recovery Rebuild option. See Figure 30 on page 43.

The RAID group status will change from ofﬂine to degraded. The rebuild status will change from rebuilding (faulted) to recovery rebuilding (n%), where n% is the completion percentage of the rebuild process.

Once the rebuild completes, the RAID group goes online:

321 4

The failed drive is either replaced with a Hot Spare or

unallocated drive, or is removed and replaced:

321 4A

If another drive fails with an unrecoverable failure during the

rebuild process, the rebuild is halted and RAID group is forced ofﬂine:

321 4A

Choose the Extreme Recovery option to try to recover data from

all drives.

321 4A 4

321 4A

(continued)

1.8 Recover Data from Ofﬂine RAID Groups

Recovery from a Failed Rebuild with a Second Unrecoverable Drive Failure (continued)

When a serious disk failure occurs during a rebuild, the rebuild cannot be forced to continue, but the Extreme recovery mode may be viable. If there is an available slot, reinstall the original failed drive (the drive provides another possible source of data even if the data is stale). This original failed drive will only be accessed if there is an error on one of the other RAID group members. Note that in this failure scenario, data may be extensively compromised.

Extreme recovery mode is only accessible from the Command Line Interface. Use the CLI command RGRecover Extreme to enter the “extreme, read-at- all-costs” recovery mode. See Using the CLI to Enable/Disable Recovery mode on page 41. Use a host application to read whatever data it can off of the drives and back it up to another location. The RAID group will stay in recovery mode. If you use the CLI command RGRecover Disable to disable this mode, the RAID group will return to the ofﬂine state. There is no mechanism to put this RAID group back online. Once you have the data backed up, use only good drives to create a new RAID group, and then restore the data to the new RAID group.

Recovery from Faults on a Critical Number of Drives

In this scenario, there is no Hot Spare drive, and either Auto Rebuild is not enabled or there is no spare drive for Auto Rebuild to use. When the ﬁrst drive fails, the array will continue to operate in DEGRADED mode and return data when requested. However, once another error occurs, the group is forced ofﬂine and data has been compromised. Basic recovery mode should be used.

A functioning RAID 5 RAID group:

321 4

Basic Recovery Mode

Use the ATTO Conﬁguration Tool or CLI commands to put the compromised RAID group into basic recovery mode. This allows you to read data to a backup location, replace all failed or marginal drives, and then restore the data to the new drives. However, at the location corresponding to the media error, the data may be invalid! Anytime a RAID group goes offline, data integrity has

been compromised. It is crucial that data consistency checks and ﬁle system repair tools are used to validate the data.

To enter recovery mode from the ATTO Conﬁguration Tool, select and right-click (or control-click) the ofﬂine RAID group to get display menu options and select the Basic Rebuild option. See Figure 31 on page 44.

The RAID group status will change from ofﬂine to recovery (basic). Once you enter recovery mode, use a host application to read whatever data it can off of the drives and back it up to another location. The RAID group will stay in recovery mode. If you use the CLI command RGRecover Disable to disable this mode, the RAID group will return to the ofﬂine state. There is no mechanism to put this RAID group back online.

Once you have backed up the data, use only good drives to create a new RAID group, and then restore the data to the new RAID group.

Extreme Recovery Mode

When a critical number of drives fail, or if a Rebuild recovery operation fails, the Extreme recovery mode may be an option. This will only work if one or both of the failed drives are still operational at some level, there are spare slots, and the data is not too far out of date. When this is the case, re-insert one or both drives into the array and select Extreme recovery mode. The original failed drives will only be accessed if there is an error on the other (non-failed) Refer to Recovery from a Failed Rebuild with a Second Unrecoverable Drive Failure on page 37.

encounters a drive failure, forcing the RAID group into a

degraded state:

321 4

When another media error occurs while the group is operating

in a degraded state, the group is forced ofﬂine:

321 4

1.8 Recover Data from Ofﬂine RAID Groups

Recovery from Replacement of the Wrong Drive

When a drive fails, but the wrong drive is replaced, a rebuild will not initiate and the RAID group will be taken ofﬂine. You should follow the procedure described in the previous chapter, Identify and Replace a Faulted Drive, and ensure the correct drive is replaced.

A functioning RAID 5 RAID group:

321 4

encounters a drive failure, forcing the RAID group into a

degraded state:

321 4

When the wrong drive is replaced, the RAID group is forced

ofﬂine!

3A21 4

The situation described above can be corrected. Replace the erroneously replaced drive with the original drive in the above example, then identify and replace the failed drive, and then start a rebuild.

File System Repair Tools

When a RAID group has been made accessible by setting the drive states appropriately, the host's ﬁle system may still have integrity problems. Supported operating systems have utilities that try to detect and repair ﬁle access issues.

Mac OS X Users: Disk Utility

1. Open Disk Utility, located in the Utilities folder in your

computer’s Applications folder.

2. Select the volume you want to repair in the column on the left, and then, click the First Aid tab on the right.

3. Click Repair Disk to verify and repair any problems on the selected volume(s). The results of the repair will be posted in the window. When no problems are found, it reports: “The volume “diskname” appears to be OK.” Occasionally, at the end of a scan, a message similar to the following will appear: “Scan complete. Problems were found, but Disk First Aid cannot repair them.”

Support Note: If there are problems found, you should

click Repair Disk to scan again. If the same problem(s) are found during the second scan, it means First Aid cannot repair the disk. This indicates that some of your files may be more severely damaged. To repair this damage, you can either back up your disk and reformat, or obtain a more comprehensive disk repair program.

Windows Users: CHKDSK

Chkdsk (Chkdsk.exe) is a Windows command line tool that checks formatted disks for problems., and then tries to repair any problems that it ﬁnds. For example, Chkdsk can repair problems related to bad sectors, lost clusters, cross-linked ﬁles, and directory errors. To use Chkdsk, you must log on as an administrator.

Linux Users: FSCK

The system utility fsck (“ﬁle system check”) performs a ﬁle system consistency check and interactive repair. It has options to do a quick check, preen ﬁles, preen and clean ﬁles, and others.

1.8 Recover Data from Ofﬂine RAID Groups

Using the CLI to Identify a Failed Drive

In addition to the ATTO Conﬁguration Tool’s graphical user interface, you may enter commands in the RAID CLI tab to identify failed drives.

BlockDevScan

The BlockDevScan command displays a list of all attached drives and their usage. Use this command to update the list of drive IDs, which are needed by other CLI commands.

blockdevscan 6 ;ID Vendor Product SerialNumber Capacity InUse Port ;--------------------------------------------------------------------- 0 ST3250310NS 9SF0800F 232.88GB r5 1 ST3250310NS 9SF076NJ 232.88GB r5 2 ST3160812AS-1 5LS25P51 149.10GB r5 3 ST3250620AS 6QF0LJ17 232.88GB r5

Ready.

For members of a RAID group, the InUse column shows the RAID group’s name.

RGDisplay

This is the RAID group display command. Use this command to display all RAID groups and their statuses.

rgdisplay 3 ;GroupName Type Interleave Capacity Partitions Members Status ;------------------------------------------------------------------------r5 RAID5 64 KB 447.0GB 1 4 DEGRADED

Ready.

Note: If RGDisplay shows a RAID group is ofﬂine, then data recovery methods should be initiated; skip this section and go directly to the

Using the CLI to Enable/Disable Recovery Mode on page 41.

If RGDisplay shows a RAID group is degraded, then it is important to replace the failed drive as soon as possible. RMStatus shows the status of each of the member drives. If a member is faulted or unavailable, it should be replaced. To match the displayed member to the physical drive, you can use SESIdentify, which is the preferred method, or BlockDevIdentify, if there is no SES capability.

RMStatus

Next, use RMStatus, the RAID member status command, to display information for each member of the RAID group in question.

rmstatus r5 6 ;ID Status RebuildStatus Vendor Product Rev SN ;----------------------------------------------------------------------- 0 ONLINE OK ST3250310NS SDN1 9SF0800F 1 ONLINE OK ST3160812AS-1 H 5LS25P51 2 FAULTED OK ST3250310NS SDN1 9SF076NJ 3 ONLINE OK ST3250620AS K 6QF0LJ17

Ready.

1.8 Recover Data from Ofﬂine RAID Groups

SESIdentify Drive (only valid if enclosure supports SES)

The drive’s fault LED should already be blinking. Use this CLI command to also blink the drive’s activity LED. The ID from the BlockDevScan command must be used, and not the member ID displayed with the RMStatus command. Match the S/N of the faulted drive (9SF076NJ) from RMSstatus to the BlockDevScan drive IDs, and use that number to select the correct drive to identify.

sesidentify drive 1

Ready.

sesidentifystop all

Ready.

BlockDevIdentify

Use this command to ﬂash the drive’s activity LED (if the drive is working enough so the activity LED can operate). The ID from the BlockDevScan command must be used, and not the member ID shown in the RMStatus command. Match the S/N of the faulted drive (9SF076NJ) from RMStatus to the BlockDevScan drive IDs, and use that number to select the correct drive to identify.

Sometimes, when BlockDevIdentify must be used, the activity LED won't ﬂash because the drive has failed so badly. One possible action is to blink all the other drives’ LEDs, and ﬁnd the failed drive by process of elimination.

blockdevidentify 1

Ready.

blockdevidstop

Ready.

Note: If SESIdentify is used, follow it up with the command SESIdentifyStop All to turn off the drive LEDs. When BlockDevIdentify is used,

follow it up with the command BlockDevIDstop (no parameters) to turn the drive LEDs off.

Using the CLI to Enable/Disable Recovery Mode

1. Display the RAID group and RAID group member drives:

rgdisplay

3 ;GroupName Type Interleave Capacity Partitions Members Status ;--------------------------------------------------------------------------- r5 RAID5 64 KB 447.0GB 1 4 OFFLINE

Ready.

rmstatus r5

6 ;ID Status RebuildStatus Vendor Product Rev SN ;-------------------------------------------------------------------------- 0 ONLINE OK ST3250310NS SDN1 9SF0800F 1 FAULTED OK ST3160812AS-1 H 5LS25P51 2 FAULTED OK ST3250310NS SDN1 9SF076NJ 3 ONLINE OK ST3250620AS K 6QF0LJ17

Ready.

(continued)

1.8 Recover Data from Ofﬂine RAID Groups

Using the CLI to Enable/Disable Recovery Mode (continued)

2. Put the RAID group into Basic or Extreme recovery mode with read and write access::

Set rgrecover r5 basic (or extreme or rebuild)

Ready.

3. Display the RAID group to see the results:

rgdisplay 3 ;GroupName Type Interleave Capacity Partitions Members Status ;---------------------------------------------------------------------------- r5 RAID5 64 KB 447.0GB 1 4 RECOVER

Ready.

4. Display the RAID group members to see the results:

rmstatus r5 6 ;ID Status RebuildStatus Vendor Product Rev SN ;--------------------------------------------------------------------- 0 ONLINE OK ST3250310NS SDN1 9SF0800F 1 ONLINE OK ST3160812AS-1 H 5LS25P51 2 ONLINE OK ST3250310NS SDN1 9SF076NJ 3 ONLINE OK ST3250620AS K 6QF0LJ17

Ready.

Once as much data as possible is retrieved from the drives, the recovery mode should be disabled and the faulted drives removed from the system.

5. Turn off the recovery mode:

set rgrecover r5 disable

Ready.

Using the CLI to Enable Recovery with Writes

On occasion, the data that has been disrupted is part of the ﬁle system on the disk drives. When this is the case, it is useful to run the ﬁle system repair tools for your operating system. This requires write access to the drives.

WARNING: Use the RGRecoverWithWrites command only when necessary, and use with caution!

Put the RAID group into Basic or Extreme recovery mode with read and write access:

set rgrecoverwithwrites r5 basic (or extreme)

Ready.

Run the appropriate ﬁle system repair utility prior to recovering data from the drives.

1.8 Recover Data from Ofﬂine RAID Groups

Degraded RAID group and its assosciated drive information displayed when the Sonnet RAID controller is selected

Figure 29

Starting a recovery rebuild operation on a faulted, offline RAID group

Figure 30

1.8 Recover Data from Ofﬂine RAID Groups

Starting a basic recovery operation on a faulted, offline RAID group

Figure 31

1.9 SNMP Conﬁguration

SNMP is a standard network protocol that provides the ability to monitor SNMP enabled systems from anywhere on the network. Once conﬁgured, 3rd party MIB (Management Information Base) browsers on a Management Station can be used to retrieve the current conﬁguration and receive TRAP messages when signiﬁcant events occur. Via the Conﬁguration Tool, SNMP is supported only for the Sonnet 3 Gb/s RAID controller (listed as an ExpressSAS R380).

The ATTO SNMP agent has two modes, depending on your conﬁguration. If the only SNMP agent on your server is the ATTO agent, the mode should be Enabled. If you have other SNMP agents running and you want them to integrate their data into a single tree of information or if you want to use advanced SNMP features, choose SubAgent mode.

Definitions

• SNMP - Simple Network Management Protocol

• Network Node - An addressable device attached to a computer network.

• Management Station - The host system that monitors network nodes.

• Trap - An alert that is sent to a management station by agents.

• Agent - A software process on the system being monitored that responds to queries using SNMP to provide status and statistics about a network node.

• SubAgent - A software process on the system being monitored that responds to SNMP queries from another agent, instead of a MIB (Management Information Base) browser. The request from the other agent is generated when it is contacted by a MIB browser.

Details

The ATTO SNMP agent supports protocol versions 1 and 2c. Communication with the agent requires UDP over IPv4. The ATTO speciﬁc information can be found under “attotech” in the “enterprises” node (1.3.6.1.4.1.4547). Read-only support is provided.

Enabled Mode

This mode of operation provides Management Stations with basic information for the standard OID (Object IDentiﬁer) tree, as well as the ATTO speciﬁc information. While this mode will also work if a system service is installed and running, the ATTO agent must be conﬁgured to listen on a different port than the system service to ensure both function properly. This mode is required to support SNMP if the system SNMP service (such as, Windows SNMP service or snmpd for OS X and Linux) is not already running.

SubAgent Mode This mode of operation relies on the operating system’s SNMP service being installed as a Master Agent and running. All Management Station requests must be directed to this Master Agent, not the ATTO agent. It is expected that only experienced SNMP administrators will use this mode, as installing and conﬁguring the operating system services are beyond the scope of this document. Examples are Windows SNMP services, or the open-source Net-SNMP package.

In this mode, the system service is automatically reconﬁgured so that when the service receives requests for the ATTO branch of the OID tree, the request is delegated to the ATTO agent through the loopback interface on a different port. While the primary beneﬁt of subagent mode is to allow the Master Agent to handle most requests directly (the operating system services implement many more MIBs than the ATTO agent), and to allow multiple subagents to be merged into a single OID tree, this mode can be used to enable features that are not supported by the ATTO agent.

For example, if the system service supports version 3 of the protocol, all communication over the network between the Management Station and the service can be encrypted using the desired privacy options of version 3. The system service will then communicate to the ATTO agent using one of the supported protocols.

Note: For security purposes, the ATTO agent will only respond to

requests through the loopback interface when in subagent mode. If the system service is stopped, the ATTO agent will be inaccessible from any other host on the network.

Note: Other features that are not directly supported by the ATTO agent,

but can be enabled with this conﬁguration include source address ﬁltering and stronger authentication than simple community strings provide. All of these features depend on properly conﬁguring the system service, and are beyond the control of the ATTO agent. Refer to the documentation for the system service to determine how to enable these features.

Note: On Windows Vista Service Pack 1 and Windows Server 2008, a

bug in the Microsoft SNMP service prevents subagent mode from working correctly. This is a known issue that requires a hotﬁx: http://support.microsoft.com/kb/950923. The bug has been ﬁxed in Service Pack 2. Enabled mode is not affected, nor is the Net-SNMP service if that is being used as a replacement for the Microsoft service.

Note: The Net-SNMP service that ships with Mac OS X 10.4 lacks the

functionality needed to support subagent mode. To work around this, you need to use the enabled mode or build the Net-SNMP package from source with proxy support enabled.

1.9 SNMP Conﬁguration

Disabled Mode

This mode disables the operation of the ATTO SNMP agent. Any network ports it has open to support SNMP are closed, making it inaccessible to any Management Station. The system service will not forward requests to it, and no TRAPs are sent out when events occur.

Basic Setup

The ATTO SNMP agent can be conﬁgured through the SNMP tab on the localhost node. When you ﬁrst enter the SNMP tab, and until you commit a new mode, the SNMP mode will be Disabled. The various settings can be manually edited, but the simplest way to enable SNMP support in this case is to click the Default button and then click Commit. This will load the proper settings based on your current system conﬁguration, and then reconﬁgure the agent to use them. The operational mode (enabled or subagent) and port will be selected based upon whether or not an operating system SNMP service is running.

Note: The System Service and TRAP Service states are not updated

dynamically as they change in the system. They can be manually updated at any time by clicking Refresh.

Note: In order to properly secure access to the agent from the local

machine, it is suggested you change the default string in the Communities list before committing the default settings.

This basic setup will allow you to browse the information that is made available by the agent with a 3rd party application, but will not send out TRAPs. TRAP destinations are network speciﬁc, and cannot be determined programmatically.

Configuration Options

Agent Port

This value speciﬁes the UDP port that the ATTO SNMP agent listens on for incoming Management Station requests. The port cannot be used by any other process on the system, or the behavior of both the agent and the other process is undeﬁned. The standard port value for SNMP is 161, but that is not the default value in subagent mode, because it is assumed that the system service is using that port.

Note: A binding error may not occur if the port is already in use. If you

are not sure if a port is in use, use the netstat command.

In enabled mode, this is the port the Management Station uses to communicate with the agent. In subagent mode however, the agent will not respond on this port to Management Station requests from a different machine. The port the Management Station needs to use is the system service’s port.

Communities

This is a list of community strings accepted by the agent when it receives an incoming request. If a Management Station makes a request and provides a community string that is not in this list, the request is dropped by the agent. If authentication traps are enabled, one will be sent to each conﬁgured destination. The list can be manipulated through the Add, Edit and Remove buttons below it. An existing community must be selected to edit or remove it. A valid community string has a length between 1 and 128 (inclusive) and can include any keyboard character. See Figure 32 on page 49

Note: In subagent mode, only a single community can be speciﬁed

since that is the community used by the Master Agent when talking to the ATTO SNMP subagent. The communities that Management Stations must use are conﬁgured through the system service.

Note: For added security in subagent mode, you should change the

default community string. The agent cannot tell the difference between a local Management Station request and a request from the system service, which means a local user can bypass the authentication checks done by the system service if they know this community string and the agent’s port.

Send Authentication TRAP

When checked, the agent will send a TRAP to the conﬁgured TRAP destinations indicating that a Management Station attempted to contact the agent and used a community string that is not in the community list.

Note: In subagent mode, this may only be useful for debugging. The

system service will perform Management Station authentication based on its conﬁguration, and only forward the request to the ATTO agent once the Management Station is authenticated. Since the system service is reconﬁgured as needed to use the community string on this pane when forwarding requests, there should not be authentication errors when the agent processes requests from the system service. The system service usually has the same option that can be enabled to see these authentication failures.

Enable TRAPs

When checked, the agent will send traps to the conﬁgured destinations, if any, if the agent itself is not disabled. See Figure 33 on page 49.

1.9 SNMP Conﬁguration

Trap Destination Table

This table lists the host address and community string used for each destination when a trap is triggered. The host column consists of the hostname or IP address of the Management Station to which the traps are sent, and the UDP port on which the trap receiver is listening. The community column displays the community string sent with the trap to that host. If the community string is not in the list of communities accepted by the receiver, the receiver will ignore the trap message. See Figure 34 on page 50.

The table can be manipulated through the Add, Edit and Remove buttons below it. An existing destination must be selected to edit or remove it. A valid community string has a length between 1 and 128.

Note: The trap conﬁguration is completely independent of the enabled

or subagent modes, but no traps are sent if the agent is disabled.

Many operating systems have a trap receiver service as part of their SNMP package, which can be conﬁgured to do various things when it receives a trap, such as writing to a log ﬁle or forwarding to another destination. In a similar manner to the way the system service and subagent mode can enable SNMPv3 support, the operating system trap service can be conﬁgured to translate traps it receives into SNMPv3 traps, with optional privacy, authentication and INFORM support. The ATTO agent then only needs to specify a single trap destination to this service in order to enable SNMPv3 support over the network. The manner in which the system service is conﬁgured is system speciﬁc, and beyond the scope of this document.

Test

This button will validate the settings displayed on screen, and then send a warm Start trap to each of the displayed trap destinations as if the displayed settings had been committed. Selecting disabled mode, unchecking the enable traps option or removing all trap destinations will prevent the test trap from being sent. Once an acceptable conﬁguration is entered, a commit is needed for it to take effect.

Note: The test is not performed by changing the settings used by the

agent at the time of the test, so testing will have no effect on any concurrent Management Station requests.

Save MIBs

This button will save the non standard MIBs implemented by the agent to the user speciﬁed directory as individual text ﬁles, therefore they can be loaded into a MIB browser. If one of the ﬁles already exists in that directory, the user is prompted to overwrite or skip the ﬁle, or cancel the operation.

Control Buttons

Commit

This button will reconﬁgure the ATTO agent to use the settings as they are currently displayed on the screen, after validating them to the extent possible. The settings are persistently stored, so that they are available after a reboot. If necessary, the system service will be reconﬁgured to support the agent’s current operational mode.

Restore

This button will update the display to reﬂect the settings the agent is currently using, rolling back any changes made in the pane that have not been committed. Since these are the current settings, they do not need to be committed.

Default

This button will update the display with the simplest conﬁguration, based on the current system conﬁguration. For example, if the operating system service is detected as running, the default mode is subagent. Otherwise, the mode is enabled and the registered SNMP port is displayed.

WARNING: Since it isn’t possible to determine trap destinations and

community strings programmatically, all trap destinations will be cleared.

1.9 SNMP Conﬁguration

Troubleshooting

Requests from an MIB browser time out:

• Ensure the agent is not disabled

• Ensure the Management Station is using the port for the system service if the agent is in subagent mode or the port displayed on the SNMP pane if the agent is enabled.

• Ensure the community string used by the Management Station is accepted by the system service when in subagent mode or is displayed in the communities list on the SNMP pane if the agent is enabled. Be sure the Send Authentication TRAP option is selected, committed, and then try again. If the community is a problem, each of the conﬁgured trap destinations will be notiﬁed.

• Ensure the Management Station is using SNMPv1 or SNMPv2c, unless the system service is being used.

• If in subagent mode, the system service may be misconﬁgured. Try enabling the agent and using the Management Station to talk directly to it. You will have to change the port value and likely the community string.

Traps are not received when testing

• Ensure the agent is not disabled

• Ensure the Enable TRAPs option is checked.

• Ensure a trap recipient is properly conﬁgured and running on each machine in the destination table, at the speciﬁed port.

• Ensure the community associated with the destination is in the trap recipient’s community list.

RAID OIDs are skipped when walking or the tables are empty

• Ensure the driver for the RAID controller is installed. This can be seen on the Basic Info pane for the controller.

Note: To properly work with Net-SNMP when conﬁgured as a

subagent, the Net-SNMP package has to be installed “With Windows Extension DLL Support”. This is an option during the installation of Net-SNMP.

Note: The architecture of Net-SNMP installed has to match the OS;

i.e. if you are using a 64-bit OS the 64-bit version of Net-SNMP must be installed.

1.9 SNMP Conﬁguration

SNMP tab with the Enable Traps option selected

Editing the Community string

Figure 32

Figure 33

1.9 SNMP Conﬁguration

Editing the trap destination

Figure 34

1.10 Conﬁguration Tool Troubleshooting

You may see an error message informing you about an unexpected event or incorrect information discovered by the application. Using the help text presented with the error message, correct the issue before proceeding.

Warnings and error messages are displayed in the Status pane.

Messages from NVRAM Tab Actions

• An error occurred loading NVRAM data. The ﬁrst time a channel is highlighted, the Conﬁguration Tool attempts to read NVRAM from the card. This message usually indicates that the Conﬁguration Tool could not communicate with the driver, probably because the application does not support the driver version in use.

• Warning: NVRAM could not be read, defaults returned. NVRAM is corrupt and the driver returns to the default conﬁguration. The defaults are presented via the graphical user interface. These defaults may be modiﬁed but the defaults or modiﬁcations must be committed (saved) in order to correct NVRAM.

• An error occurred updating the NVRAM. The driver cannot load the new settings on the card; no changes are made to the card.

Feature bounds checking

When the Commit button is clicked, each NVRAM feature is validated before being sent to the card. If any one of these features is deemed inappropriate based on the implemented checks, further NVRAM feature validation checks are stopped and the message is displayed.

Messages from Flash Tab Actions

• This is not a flash file, or it is corrupt. The ATTO-created ﬂash ﬁle is corrupt or the Conﬁguration Tool does not recognize the ﬁle as a ﬂash ﬁle. Only ATTO-created ﬂash ﬁles may be selected using the ﬂash ﬁle dialog box.

• This HBA is not compatible with the selected flash file. ATTO ﬂash ﬁles are created based on the type of card ﬂashed. Only certain ATTO ﬂash ﬁles are compatible with the Sonnet RAID controller. When a ﬂash ﬁle is selected, it is inspected to determine whether it is compatible.

• A valid file was not selected. You clicked the Cancel button on the ﬂash ﬁle selection dialog.

• An error occurred reading from the flash file, the file may be corrupt. You selected a compatible ﬂash ﬁle but the contents are corrupt.

• An error occurred updating the flash. You tried to ﬂash a card when the ﬁrmware was not able to accept a ﬂash.

• The card has been prepared for firmware updating, but the

machine must be rebooted for the changes to take effect. You need to repeat this process after rebooting to actually update the firmware.

Some ﬁrmware upgrades need to prepare the existing ﬁrmware in order to successfully update the controller. Rebooting enables the changes made during the preparation process to take effect, and the same ﬁle should be ﬂashed again.

• Execution Throttle is greater than the maximum allowable

value of 255. No NVRAM configuration changes have been made to your card. The exact message varies based on the ﬁrst

ﬁeld with an out-of-range value.

2.0 Drive Performance Testing

The ATTO Conﬁguration Tool’s drive performance testing feature records the performance level of a drive in a RAID group under an I/O (input/output) load generated by a host application on your computer. The information can be used to compare the performance of each drive relative to others.

The Conﬁguration Tool’s performance test application measures the data transfer rates in Megabytes per second (MB/sec.) which elapse from the time the start button is clicked until the stop button is clicked. Drive performance can be measured for any combination of single drives within a RAID group or across multiple RAID groups, or for all the drives in a selected RAID group.

Note: The performance test can only be performed on drives which

belong to a RAID group.

The Drive Performance dialog box displays individual statistics for each drive selected. After you start the I/O load application, the collection of these statistics begins when the Start button is pressed and continues until the Stop button is pressed. You may press the Reset button and restart a test at any time.

The Conﬁguration Tool application can be closed after starting the performance test and re-opened to show the statistics from the most recent test.

9. Stop the I/O any time after you have stopped the

performance test.

1. Start the I/O-generating application on your computer.

2. Select the Sonnet RAID controller from the Device Listing pane.

3. Select the RAID tab in the right pane.

4. Select the drives to be tested:

• Select individual drives from the Attached Drives pane.

• Select a single RAID group in the Groups pane.

5. Select RAID Management > Performance from the application menu.

6. The Performance dialog box displays. See Figure 35 on page 54.

7. Click the Start button.

8. Click the Stop button to halt monitoring.

Click the Reset button at any time to reset the performance

values to zero.

You can close the Performance box while monitoring is in progress, and then open it later to see the performance results.

2.0 Drive Performance Testing

Drive Performance window before starting the test

Figure 35

2.1 Windows Only - ATTO Disk Benchmark

The ATTO Utilities for Windows are installed from the CD that was included with your Sonnet RAID controller. Only one utility, Disk Benchmark, may be used with your Sonnet RAID storage system.

Disk Benchmark measures peak and sustained throughput for disk reads and writes. See Figure 36 on page 56. You may locate the Disk Benchmark application in the ATTO HBA Utilities folder within your system’s Applications folder.

1. Launch the application.

2. Select the drive letter for the disk to benchmark.

3. Select the transfer sizes to test.

4. Select the I/O option.

5. Click the Start button.

6. Wait for benchmark to run through the desired transfer sizes.

7. The Test Results Display at the bottom of the window is updated as the test progresses. The y-axis of the graph represents the transfer sizes in the selected range. The x-axis represents the transfer speeds in MB/sec. I/O speeds in KB/sec. for each transfer size are displayed textually to the right of the graph.

Support Note: Additional information on using Disk

Benchmark is available by accessing the Help menu in the

application.

Benchmark Fields

The benchmark ﬁelds include:

• Drive: Select the logical drive to benchmark. A test can be performed on any system drive.

• Transfer Size: Select the range of transfer sizes used for reading and writing data to the test drive. Transfer speeds are displayed for each size in the range. If the ﬁrst size is greater than the second size, the test is not performed for any transfer size.

• Total Length: Select the total size of the data ﬁle to be created on the test drive. This ﬁle is deleted when testing completes.

• Direct I/O: If this option is checked, ﬁle I/O on the test drive is performed with no system buffering or caching. Combine this option with Overlapped I/O for maximum asynchronous performance

Radio Button Group

• Overlapped I/O performs queued I/O. Upon selection, the Queue Depth option displays to select the maximum number of

read or write commands that may be executed simultaneously.

8. Click the Stop button to stop the test. When the test completes, the results can be saved or printed.

If errors were detected, a dialog box displays the errors in a table with the following four columns and a button:

• Benchmark Transfer Size: transfer size at which the error

occurred

• Buffer Index: index into the data block at which the error

occurred

• Actual Value: the value read from the ﬁle

• Expected Value: the value written to the ﬁle

• Log to File: Logs the error table to a *.log ﬁle and closes the

dialog. The ﬁle is given the same name as the test ﬁle and saved in the same directory. If the test was not previously saved, errors are logged to the generic ﬁle Bench32Error.log in the root of the test drive. If the log ﬁle already exists, the new errors are appended to the previously recorded errors. This is the only way to save detected errors. They are not saved in the test document ﬁle.

If the I/O comparison option was selected and errors were not detected, the message “No errors detected” is displayed.

• I/O Comparison compares the data read from the test ﬁle to the data written on a per block basis. You can select the data pattern for comparison from the Test Pattern drop-down box.

• Run Continuously runs the test continuously for a speciﬁed number of minutes. The test stops before the speciﬁed time if any errors are detected.

• Neither: Select if you do not want to perform overlapped I/O or I/O comparisons.

The following ﬁelds do not affect the benchmark but are informational, providing documentation of the test environment.

• Stripe Group: If the test drive is a stripe group, select its name from the list box. The names and quantities of drives in the stripe group are printed to the Description box. Select Clear to clear the contents of the Description box.

• Controlled by displays all Sonnet RAID controllers in the system.

• Description: Enter additional information about the test that can be saved or printed. Be sure to enter additional information after making a selection from the Stripe Group drop down box, as this erases the current description.

2.1 Windows Only - ATTO Disk Benchmark

Multiple Benchmark Testing

Disk Benchmark supports four command line parameters for uninterrupted testing:

• testfile opens and executes the test named testfile with the extension .bmk.

• textfile opens the text ﬁle named textfile. This ﬁle contains a list of test ﬁle names that have an extension of .bmk. Each test in this list is opened and executed in order. Stopping one test in the list prevents further tests from being executed. Error logging is the same as the command line parameter testﬁle, but all errors generated from all tests in the list are logged to one ﬁle: textﬁle.log.

Disk Benchmark screens before and after a test has been run

• /p testfile: Same as testfile, only the test is printed to the

default system printer instead of being executed.

• /p textfile: Same as textfile, only the tests in the list are printed to the default system printer instead of being executed.

Figure 36

2.2 ATTO Disk Benchmark Troubleshooting

The following suggestions may help if you encounter problems with Disk Benchmark.

• Use Windows Device Manager to check and verify that all drives are visible to the operating system.

• If drives are not listed, check the connections between the drive enclosure and the RAID controller card, and verify that all drives are fully seated in their bays.

• Make sure that the enclosure is powered up and has completed its self check before booting your computer.

• Reboot your system any time you make changes to a RAID group (after the RAID group has been rebuilt).

• As a last resort, you may use the ATTO Boot Conﬁguration Utility to low level format a troublesome device. However, this erases all information on the disk.

• Have you partitioned your drive, and then activated that partition?

• Did you format the drives for use with your operating system?

If problems persist, contact Sonnet customer service.

Appendix A - CLI ASCII-Based Interface

The RAID CLI tab enables experienced users to send Command Line Interface (CLI) commands to the Sonnet RAID controller to conﬁgure, test, and otherwise interact with attached storage.

WARNING: Do not use CLI unless you are directed to by

a Sonnet technician, as changing parameters may cause loss of data and/or disruption to performance and reliability of the Sonnet RAID controller. The ATTO Configuration Tool interface is the preferred method to operate and manage the Sonnet RAID controller.

The command line interface uses a set of ASCII-based commands to control conﬁguration and diagnostic tasks. See Figure 5 on page 5.

• CLI commands are context sensitive and generally follow a standard format [Get|Set] Command [Parameter1|Parameter2] followed by the return or enter key

• CLI commands are case insensitive: you may type all upper or all lower case, or a mixture. The use of mixed upper and lower case in this manual and the help screen are for clariﬁcation only.

• The get form returns the value of a parameter or setting and is an informational command.

• Responses to get commands are followed by Ready.

• The set form is an action that changes the value of a parameter or conﬁguration setting. It may require a SaveConfiguration command and a restart of the system before it is implemented. The restart can be accomplished using a separate FirmwareRestart command. A number of set commands may be issued before the SaveConfiguration command.

• Responses to set commands are either an error message or Ready. *. The asterisk indicates you must use a SaveConfiguration command to ﬁnalize the set command.

• Set commands which do not require a SaveConfiguration command, deﬁned as immediate commands, are immediately executed.

Support Note: Using certain CLI commands during

normal operation can cause a performance drop. Once command actions are complete, performance should return to normal levels.

• Commands generally have three types of operation: get, set and immediate.

Figure A-1 Symbols, typefaces, and abbreviations used to indicate functions and elements of the command line interface used in this manual.

Symbol Indicates

[ ] Required entry

< > Optional entry

| Pick one of

n – n A range (6 – 9 = 6, 7, 8, 9)

BlockDevID Index designation of a block device not assigned to any other RAID group; the index of a block device provided by the

BlockDevScan command. 0<=n<=63

GroupName The name of the RAID group to which the block device is assigned, or blank if the drive is available

MemberIndex Index designation of a RAID group member as found in the RMStatus command

PartIndex Index designation of a partition as found in the PartitionDisplay command

SASIndex Index designation of SAS drives as found in SASTargets

TID Target ID 0<=n<=255

Appendix A - CLI ASCII-Based Interface

CLI Error Messages

The following error messages may be returned by the Command line Interface

ERROR Invalid Command. Type “Help“ for command list.

ERROR Command Not Processed

ERROR Wrong/Missing Parameters

ERROR Invalid Hot Spare Serial Number

ERROR Invalid RAID GroupName

ERROR Invalid RAID Group State

ERROR Insufficient number of RAID Group members

ERROR RAID Group does not exist

ERROR No RAID Groups found

ERROR Invalid RAID Type

ERROR RAID Group is already unmapped

ERROR Invalid Block Device Index

ERROR Cannot perform operation. RAID Group has mapped Partitions

ERROR Cannot perform operation. RAID Group has Outstanding Commands

ERROR Block Device at specified index no longer available

ERROR Insufficient RAID Group members for RAID type

ERROR Incorrect number of RAID Group members for QuickVideo configuration

ERROR Invalid Virtual Drive ID

ERROR Specified capacity is invalid

ERROR Too many Indices specified

ERROR Only one add storage operation is permitted at any given time

ERROR No free block devices

ERROR Cannot benchmark a drive that is being initialized

ERROR Invalid RAID MemberIndex

ERROR Invalid RAID Member State

ERROR Missing RAID Member

ERROR Invalid RAID Member Capacity

ERROR Invalid Partition Index

ERROR Maximum number of RAID Groups exceeded

ERROR Maximum number of Partitions exceeded

ERROR Invalid number of Partitions

ERROR Maximum number of RAID Members exceeded

ERROR Maximum stripe width

ERROR Invalid number of Partitions specified

ERROR Invalid Span Depth specified

ERROR Cannot perform operation on mapped Partition

ERROR Specified drive is not being monitored

Appendix A - CLI ASCII-Based Interface

CLI Summary

The following chart summarizes the Command Line Interface commands, their defaults, and an example of how to enter the commands. Please note that commands which have no default values have a blank entry in that column of the table.

WARNING: Do not use the CLI unless you are directed

to by a Sonnet technician, as changing parameters may cause loss of data and/or disruption to performance and reliability of the Fusion storage system.

Command Default Example

AutoMap automap

AutoMapOnBoot enabled set automaponboot enabled

AutoResume rebuild = enabled, erase = disabled, initialize

= enabled

BlockDevClean blockdevclean 30

BlockDevIdentify blockdevidentify 30

BlockDevIDStop blockdevidstop

BlockDevInfo blockdevinfo 30

BlockDevScan blockdevscan

ClearEventLog cleareventlog

Date set date 03/03/2009

DeleteScheduledTasks deletescheduledtasks

DisplayScheduledTasks displayscheduledtasks

DriveAssureReport driveassurereport raidgroup1

DriveHealth disabled set drivehealth enabled

DriveHealthDisplay drivehealthdisplay all

DriveHealthStatus drivehealthstatus

DriveTest drivetest begin

DriveTestClearList drivetestclearlist all

set autoresume erase enabled raidgroup1

DriveTestConﬁg not initiated set drivetestconﬁg read

DriveTestList get drivetestlist all

DriveTestStatus get driveteststatus

DumpConﬁguration dumpconﬁguration

DumpEventLog dumpeventlog

EventLog enabled set eventlog disabled

EventLogFilter all all all set eventlogﬁlter gen info all

Help help eventlog

HSAdd hsadd 3

iii

Appendix A - CLI ASCII-Based Interface

Command Default Example

HSDisplay hsdisplay

HSRemove hsremove 3

Info info

IsReserved isreserved

Metrics metrics display all

OEMConﬁgFile get oemconﬁgﬁle

Partition partition alpha1 6 4 GB

PartitionDisplay partitiondisplay alpha1

PartitionMerge partitionmerge all

PartitionSplit partitionsplit alpha1 22 2

PartitionWriteCache set partitionwritecache enabled

RAIDRebuildPriority same set raidrebuildpriority low

RAIDSpeedWriteLimit 8 set raidspeedwritelimit 16

RebuildContinueOnError disabled rebuildcontinueonerror raidgroup1

Reserve reserve

RestoreConﬁguration restoreconﬁguration default

RGAddStorage rgaddstorage raidgroup1 span commit

RGAutoRebuild disabled set rgautorebuild all enabled

RGCancelAddStorage rgcanceladdstorage raidgroup1

RGCancelMediaScan rgcancelmediascan raidgroup1

RGCommit rgcommit all

RGCreate rgcreate raidgroup1 raid0

RGDiskWriteCache set rgdiskwritecache all disabled

RGDisplay rgdisplay all

RGErase rgerase raidgroup1

RGHaltConversion rghaltconversion raidgroup1

RGHaltErase rghalterase raidgroup1

RGHaltInitialization rghaltinitialization raidgroup1

RGHaltMediaScan rghaltmediascan raidgroup1

RGHaltRebuild rghaltrebuild raidgroup1

RGMediaScan rgmediascan raidgroup1 verify

RGMediaScanErrorReport set rghdparameter raidgroup1 30

RGMemberAdd rgmemberadd raidgroup1 30

RGMemberRemove rgmemberremove raidgroup1 30

RGPrefetch set rgprefetch raidgroup1 3

RGRebuild rgrebuild raidgroup1

Appendix A - CLI ASCII-Based Interface

Command Default Example

RGRecover rgrecover raidgroup1 rebuild

RGRecoverWithWrites rgrecoverwithwrites raidgroup1 basic

RGResumeConversion rgresumeconversion raidgroup1

RGResumeErase rgresumeerase raidgroup1

RGResumeInitialization rgresumeinitialization raidgroup1

RGResumeMediaScan rgresumemediascan raidgroup1

RGResumeRebuild rgresumerebuild raidgroup1

RGSectorSize 512 setrgsectorsize raidgroup1 8192

RGSpanDepth 1 set rgspandepth raidgroup1 22

RGSpeedRead all disabled set rgspeedread raidgroup1 enabled

RGUnmap rgunmap raidgroup1

RGWaitTimeout 3 rgwaittimeout 30

RMStatus rmstatus raidgroup1

Route route host 1 raid alpha1 6

RouteDisplay routedisplay 03 124

SASPortList sasportlist

SASTargets sastargets

SaveConﬁguration saveconﬁguration

SerialNumber get serialnumber

SES enabled set ses disabled

SESAlarmTest

SESDiskFailureAlarm disabled set sesdiskfailurealarm enabled

SESEnclosures sesenclosures

SESIdentify off set sesidentify all

SESIdentifyStop sesidentifystop all

SESMute

SESPoll 30 set sespoll 0

SESStartingSlot 1 set sesstartingslot 0

SESStatus

Time set time 03:32:30

TimeZone EST set timezone pst

VerboseMode enabled set verbosemode disabled

VirtualDriveInfo virtualdriveinfo

WrapEventLog enabled set wrapeventlog disabled

Appendix A - CLI ASCII-Based Interface

CLI Command Explanations

Command line interface commands are listed alphabetically with explanations of what they are used for, their defaults and syntax.

WARNING: Using CLI without contacting a Sonnet

technician is not recommended because changing parameters may cause loss of data and/or disruption to performance and reliability of the Fusion storage system.

• AutoMap (Immediate, Disabled on Error)

Automatically maps each RAID partition to a Target ID on the host system (maximum 256 maps allowed). If the optional passthrough parameter is entered then AutoMap stores maps for currently connected SAS/SATA Pass Through Devices.

AutoMap <passthrough>

• AutoMapOnBoot (Enabled, Disabled on Error) Regulates the automatic detection and mapping of RAID groups at startup.

set AutoMapOnBoot [enabled | disabled] get AutoMapOnBoot

• AutoResume Regulates the AutoResume features for interrupted rebuild, erase, and initialization operations at startup. Optional parameter GroupName speciﬁes the RAID group to operate on. If no GroupName is speciﬁed, the command operates on all existing RAID groups.

• BlockDevClean (Immediate, Disabled on Error) Removes any RAID conﬁguration data from the block device with the speciﬁed BlockDevID. BlockDevId is the index of a block device provided by the BlockDevScan CLI command.

BlockDevClean [BlockDevID]

• BlockDevIdentify (Immediate, Disabled on Error) Turns on a drive activity LED on the Fusion drive enclosure for a speciﬁed drive until a BlockDevIDStop command is given. Use either the RAID group name and member index, or BlockDevId. BlockDevId is the index of a block device provided by the BlockDevScan CLI command.

WARNING: The BlockDevIdentify command is intended

for diagnostic purposes only. Executing this command may adversely impact the performance and throughput of the Fusion storage system for the time that the LED is illuminated.

BlockDevIdentify [[Groupname MemberIndex] | BlockDevID]

• BlockDevIDStop (Immediate, Disabled on Error) Turns off the drive activity LED on the Fusion drive enclosure that was activated with the BlockDevIdentify command. Specify the RAID group name and member index, BlockDevId, or no parameters (which will turn off all previously identiﬁed drive’s LEDs). BlockDevId is the index of a block device provided by the BlockDevScan CLI command.

BlockDevIDStop <[GroupName MemberIndex] | BlockDevID>

• BlockDevInfo (Immediate, Disabled on Error) Lists detailed information about connected physical block devices along with any potential RAID group association.

BlockDevInfo [BlockDevID | all]

• BlockDevScan (Immediate, Disabled on Error) Lists all currently connected physical drives along with any potential RAID group association. Each block device listed is assigned a unique index at the time of the scan to identify drives for other CLI operations.

BlockDevScan

• ClearEventLog (Immediate) Clears the contents of the event log.

ClearEventLog

WARNING: All RAID group setup information is lost

when the BlockDevClean command is performed, therefore all data is lost. Back up your files before performing this command.

• Date Sets the current date. The date range is 01/01/2000 to 12/31/2099.

set Date [MM]/[DD]/[YYYY] get Date

• DeleteScheduledTasks (Immediate, Disabled on Error) Deletes a scheduled task with the ID returned by DisplayScheduledTasks.

DeleteScheduledTasks [Id]

Appendix A - CLI ASCII-Based Interface

• DisplayScheduledTasks (Immediate, Disabled on Error) Displays all outstanding scheduled tasks.

DisplayScheduledTasks

• DriveAssureReport (Immediate, Disabled on Error) Displays drive timeout error statistics for all member drives in a speciﬁed RAID Group.

DriveAssureReport [GroupName]

• DriveHealth (Disabled, Disabled on Error) Changes the system’s ability to acquire drive health data from connected drives. Issuing this command during I/O operations may adversely affect performance.

set DriveHealth [enabled | disabled] get DriveHealth

• DriveHealthDisplay (Immediate, Disabled on Error) Retrieves and displays S.M.A.R.T. (Self-Monitoring, Analysis and Reporting Technology) data from SATA drives, and medium defect and information exceptions counts from other drives. Issuing this command during I/O operations may adversely affect performance.

DriveHealthDisplay [BlockDevID | all]

• DriveHealthStatus (Immediate, Disabled on Error) Displays the current S.M.A.R.T. status of speciﬁed SATA disk drives, and medium defect and information exceptions counts from other drives.

DriveHealthStatus [BlockDevID | all]

• DriveTestConfig (Disabled on Error) Conﬁgures the drive test to perform one of the following operations: initialize (destructive write-only), read (nondestructive read-only), verify (destructive verify), mediascan (destructive for sectors with medium errors), or initverify (destructive write-read-verify). The test is not started until the DriveTest Begin command is given. A new conﬁguration may not be set while a drive test is being performed.

set DriveTestConfig [init | read | verify | mediascan | init-verify] get DriveTestConfig

• DriveTestList (Disabled on Error) Speciﬁes the drives to be included in the next drive test. DriveTestConfig should be set up prior to adding any drives into the test list. This command can be called with different eligible BlockDev IDs and each one will be added to the list. Drives which are part of a RAID group are only eligible for read drive tests. Additionally, Hot Spare drives are only eligible for media scan and read drive tests. The all parameter automatically chooses eligible drives. The test is not started until the DriveTest Begin command is given.

set DriveTestList [drive [BlockDevID] | all] get DriveTestList

• DriveTestStatus Displays the status of the currently running drive test but does not display performance metrics. If a block device ID is not running or cannot be found, its state will be idle and percent complete will be 0.

get DriveTestStatus <drive [BlockDevID]>

• DriveTest (Immediate, Disabled on Error) Starts or stops a drive test with the previously speciﬁed conﬁguration (refer to DriveTestConfig) and drive list (refer to DriveTestList). Drives being tested are not available for RAID conﬁguration or RAID operations. Only one test can be run at a time.

DriveTest [begin | cancel]

• DriveTestClearList (Immediate, Disabled on Error) Speciﬁes drives to be removed from the drive test list. The drive BlockDevID parameter removes the speciﬁed drive from the list. The all parameter removes all drives from the list.

DriveTestClearList [drive [BlockDevId] | all]

• DumpConfiguration (Immediate) Displays a unit’s conﬁguration to the management interface.

DumpConfiguration

• DumpEventLog Dumps the contents of the entire event log to the management interface.

DumpEventLog

• EventLog (Enabled, Disabled on Error) Regulates event logging. When enabled, records various system errors to the event log.

set EventLog [enabled | disabled] get EventLog

vii

Appendix A - CLI ASCII-Based Interface

• EventLogFilter (Enabled, Disabled on Error) Filters data from speciﬁc unit subsystems and levels when event logging is enabled. The speciﬁc entries supported are platformdependent. For set commands, the ﬁnal parameter indicates whether or not events from the speciﬁed subsystem and level are displayed.

• Help (Immediate) Displays a list of available commands. If you specify a command, detailed, command-speciﬁc information is displayed.

Help <command>

• HSAdd (Immediate) Assigns a drive to the Hot Spare pool.

HSAdd [BlockDevID]

• HSDisplay (Immediate) Lists all drives in the Hot Spare pool.

HSDisplay

• HSRemove (Immediate) Removes a drive from the Hot Spare pool

HSRemove [BlockDevID | all]

• Info (Immediate) Displays version numbers and other production information for key components. Use the optional brief parameter to display a more concise subset of system information.

Info <brief>

• IsReserved (Immediate) Displays the reservation status of the current services session or interface.

IsReserved

• MediaScanErrorReport (Immediate, Disabled on Error) Displays media scan error statistics for either a single drive or all drives in the system

MediaScanErrorReport [BlockDevID | all]

• Metrics (Immediate) Controls the collection of standard data metrics within a product based on the command parameters.

Metrics [start | stop | display] [drive BlockDevID] | all | running]

• OEMConfigFile (Disabled on Error) Displays the “name” (i.e., the contents of the ﬁrst record) of the OEM conﬁguration ﬁle stored in persistent memory.

get OEMConfigFile

• Partition (Immediate) Creates a speciﬁed partition with a speciﬁed capacity in Gigabytes (GB), Megabytes (MB), or blocks. The speciﬁed capacity must be smaller than the speciﬁed partition’s current capacity. A new partition is created to acquire the remainder of the original partition’s space.

Partition [GroupName] [PartIndex] [capacity] [GB | MB | blocks]

• PartitionDisplay (Immediate) Lists all the partitions available in the speciﬁed RAID group. The partitions are listed contiguously (as opposed to index order). GroupName is the ASCII name of the RAID group for which partitions will be displayed.

PartitionDisplay [GroupName]

• PartitionMerge (Immediate) Merges the speciﬁed contiguous partitions into one partition. GroupName is the ASCII name of the RAID group containing the partitions to merge. PartIdx is the index of a partition to merge, along with a number of contiguous partitions to merge to that index. will be merged into a single virtual disk. The RAID group must not be in a new state. None of the partitions to merge may be mapped.

PartitionMerge [GroupName] [[[PartIdx] [1-128]] | all]

• PartitionSplit (Immediate) Divides the speciﬁed partition into one or more partitions whose capacities are evenly distributed among the capacity of the original partition. GroupName is the ASCII name of the RAID group containing the partition to split. PartIdx is the index of the partition to split. The partition to split cannot be mapped and the RAID group must not be in a new state.

PartitionSplit [GroupName] [PartIdx] [1-128]

All indicates that all partitions in the RAID group

viii

Appendix A - CLI ASCII-Based Interface

• PartitionWriteCache Enables higher write performance with a small risk of data loss after a system failure. If disabled, provides a higher level of data integrity with lower write performance.

set PartitionWriteCache [GroupName] [PartIdx]

[enabled | disabled] get PartitionWriteCache [GroupName] [PartIdx]

• RAIDRebuildPriority Sets or displays the RAID rebuild priority. A RAID rebuild priority set to high gives higher priority to RAID rebuilds and lower priority to the processing of simultaneous read/write transactions. A RAID rebuild priority set to low gives lower priority to the rebuild and a higher priority to read/write transactions. Set to same, the RAID rebuild and processing of read/write transactions is the same. If all or no groups are speciﬁed, the system default and all of the individual RAID groups are set. If a group name is speciﬁed, only the speciﬁed group is set.

set RAIDRebuildPriority <GroupName | all> [high | low | same] get RAIDRebuildPriority <GroupName | all>

• RAIDSpeedWriteLimit (Disabled on Error) Sets or displays the limit on the coalescing factor. Warning: changing the default setting (8) may result in poor performance or timeouts. A lower setting is recommended when using multiple initiators. A higher setting may improve performance with multiple streams of sequential write I/O, but too high a setting will cause timeouts.

set RAIDSpeedWriteLimit [0 - 32] get RAIDSpeedWriteLimit

• RebuildContinueOnError Enables a rebuild to continue to completion even if media read errors are encountered during the rebuild operation.

set RebuildContinueOnError [GroupName1 | all] [enabled | disabled] get RebuildContinueOnError [GroupName1 | all]

• Reserve (Immediate) Reports the state of CLI reservation for the current CLI session. If the command reports that reservations are enabled, then another CLI session has control of parameter modiﬁcation.

Reserve

• RestoreConfiguration (Immediate, Disabled on Error) Forces the NVRAM settings to their original defaults when issued with the default option. The saved option undoes any changes made to this session since the last save.

RestoreConfiguration [default | saved]

• RGAddStorage (Immediate) Adds additional storage to an existing RAID group. GroupName is an ASCII name for the RAID group. MIRROR |STRIPE |SPAN speciﬁes the method used to expand the storage. Optional parameter list BlockDeviceID speciﬁes up to 10 indices of available block devices, provided by the BlockDevScan CLI command, to be added to the RAID group. If this list is omitted, the CLI command RGMemberAdd must be used. Optional parameter commit runs the RGCommit command automatically and all user data will be erased from each new member drive. If the parameter is omitted, the CLI command RGCommit must be entered. Any time before RGCommit is entered, the command RGCancelAddStorage can be used to cancel the process. Note: Mirrors cannot be added to a RAID 4, RAID 5, RAID 6, or DVRAID RAID group.

RGAddStorage [GroupName] [mirror | stripe | span] <BlockDeviceID | <commit>

• RGAutoRebuild Enables and disables Auto-Rebuild functionality for one or more RAID groups. Auto-Rebuild uses drives assigned as Hot Spares, followed by available drives, as automatic replacements for any member that fails. Auto- Rebuild is disabled by default.

set RGAutoRebuild [GroupName | all] [enabled | disabled] get RGAutoRebuild [GroupName | all]

• RGCancelAddStorage (Immediate, Disabled on Error) Cancels the RGAddStorage command.

RGCancelAddStorage [GroupName]

• RGCancelMediaScan (Immediate, Disabled on Error) Cancels a media scan that is running on the speciﬁed existing RAID group.

RGCancelMediaScan [GroupName]

• RGCommit (Immediate, Disabled on Error) RGCommit stamps a new RAID group’s conﬁguration to its member drives. Advanced Initialization is highly recommended for new drives; this erases and veriﬁes the drive media. The

RAID group is unavailable until the operation completes. Express Initialization performs a background initialization and the RAID group is immediately available for use. When RGCommit is issued after adding storage, it stamps an existing RAID group’s conﬁguration to the added drives, and initializes the drives if Advanced or Express are speciﬁed. GroupName is the ASCII name of the new RAID group to commit.

RGCommit <GroupName <advanced | express> | all <advanced | express>>

Appendix A - CLI ASCII-Based Interface

• RGCreate (Immediate) Creates a new empty RAID group. GroupName is an ASCII name for the RAID group (14 characters maximum, no spaces). The optional value after the RAID group type parameter represents the desired interleave for the RAID group. KB denotes interleave in kilobytes; without the KB sufﬁx, interleave is set in 512 byte blocks. If interleave is not provided, the system-default interleave is used.

RGCreate [GroupName] [RAID [ 0 | 1 | 10 | 4 | 5 | 6 ] | JBOD] <8KB | 16KB | 32KB | 64KB | 128KB | 256KB | 512KB | 1024KB | 2048KB>

• RGDiskWriteCache (Immediate)

Produces higher write performance with a small risk of data loss after a system failure. If disabled, drives are updated at the expense of some write performance.

set RGDiskWriteCache [GroupName | all] [enabled | disabled] get RGDiskWriteCache [GroupName | all]

• RGDisplay (Immediate) Displays status information for a single RAID group, or if all is used, all available RAID groups.

RGDisplay <GroupName | all>

• RGErase (Immediate) Erases the data from the speciﬁed existing RAID group. Warning: All data is lost when you use the RGErase command!

RGErase [GroupName]

• RGHaltConversion (Immediate) Stops the conversion on the speciﬁed existing RAID group.

RGHaltConversion [GroupName]

• RGHaltErase (Immediate) Stops the erase on the speciﬁed existing RAID group.

RGHaltErase [GroupName]

• RGHaltInitialization (Immediate) Stops the initialization process on the speciﬁed existing RAID group.

RGHaltInitialization [GroupName]

• RGHaltMediaScan (Immediate, Disabled on Error) Halts a media scan on the speciﬁed existing RAID group.

RGHaltMediaScan [GroupName]

• RGHaltRebuild (Immediate) Halts the rebuild(s) on the speciﬁed existing RAID group. Optional parameter MemberIndex speciﬁes the RAID member whose rebuild will be halted. For RAID 6 RAID groups, if a MemberIndex is speciﬁed, all rebuilding RAID members on the span with that MemberIndex will halt as well. If no MemberIndex is speciﬁed, all rebuilds on that RAID group will be halted.

RGHaltRebuild [GroupName] <MemberIndex>

• RGMediaScan (Immediate, Disabled on Error)

Initiates a Media Scan. A Media Scan reads all of a RAID group’s member drives and corrects media errors by calculating the expected data and rewriting it, so the drive can relocate it to a good sector. The verify option adds a data integrity check by verifying that the data and parity match. Verify plus fix causes the parity to be rewritten when a verify mismatch occurs. Enter the time (must be HH:MM in 24-hour time format) and the day of the week without the “daily/weekly“ option to schedule a one-time scan for a later date. Enter time, day of week and daily or weekly to schedule a scan on a recurring basis.

RGMediaScan [GroupName] <verify | verify fix> <[HH:MM] <day of week> <daily | weekly>>

• RGMediaScanErrorReport (Immediate, Disabled on Error)

Displays error statistics for either a single RAID group or all RAID groups in the system.

RGMediaScanErrorReport [GroupName | all]

• RGMemberAdd (Immediate) Adds available drives to a new RAID group or as part of an RGAddStorage operation. GroupName is the ASCII name of the RAID group to receive the RAID member. BlockDevID is the index of an available drives provided by the BlockDevScan CLI command. Up to 10 BlockDevIDs may be speciﬁed. If all is speciﬁed, then all available unused BlockDevIDs will be added

to the RAID group until the maximum number of RAID group members has been met. This command also resets the number of RAID group partitions to 1.

RGMemberAdd [GroupName | all] [BlockDevID]

• RGMemberRemove (Immediate)

RGMemberRemove removes a RAID member from a new RAID group. GroupName is the ASCII name of the new RAID group from which to remove the RAID member. MemberIndex is the index of the RAID member to remove. This also resets the number of partitions to 1.

RGMemberRemove [GroupName] [MemberIndex]

Appendix A - CLI ASCII-Based Interface

• RGPrefetch

Sets the prefetch for all or for the speciﬁed RAID group. This command will fail if the RAID group does not exist. GroupName is the ASCII name of the RAID group for which the parameter will apply.

set RGPrefetch [GroupName | all] [Value 0 to 6] get RGPrefetch [GroupName | all]

• RGRebuild (Immediate) Starts rebuilding the speciﬁed existing RAID Group. Optional parameters MemberN specify the members to rebuild. If no member is speciﬁed, all degraded members will be rebuilt. Optional parameters BlockDevIDN allows an available block device to be substituted for the RAID member currently assigned to the Member Index. RAID 6 RAID groups can rebuild two members using the optional and.

RGRebuild [GroupName] <Member1> <BlockDevID1> <and> <Member2> <BlockDevID2>

• RGRecover (Immediate) Provides mechanisms for assisting in the read-only access of data in offline RAID groups. Specify rebuild if the RAID group faulted while rebuilding. Basic forces a RAID group online with only up-to-date members and returns check condition on read errors. Extreme forces a RAID group online with older members and replaces read error data with zeros. Disabled turns off RGRecover for basic and extreme. Note: Before running

RGRecover, power off and power back on all drives in the affected RAID group drives to ensure the drives are at a known state and ready for data recovery.

RGRecover [GroupName] [rebuild | basic | extreme | disabled]

• RGRecoverWithWrites (Immediate) Performs the same functions as RGRecover, except writes to the affected RAID group are allowed to occur normally. Note: Care

must be exercised to minimize write activity. Proceed with writes at your own risk.

RGRecoverWithWrites [GroupName] [rebuild | basic | extreme]

• RGResumeConversion (Immediate) Continues the stopped conversion on the speciﬁed existing RAID group.

RGResumeConversion [GroupName]

• RGResumeInitialization (Immediate) Continues the initialization on the speciﬁed existing RAID group.

RGResumeInitialization [GroupName]

• RGResumeMediaScan (Immediate, Disabled on Error)

Resumes a media scan on the speciﬁed existing RAID group.

RGResumeMediaScan [GroupName]

• RGResumeRebuild (Immediate) Resumes the rebuild(s) on the speciﬁed existing RAID group. Optional parameter MemberIndex speciﬁes the RAID member whose halted rebuild will be resumed. For RAID 6 RAID groups, if a MemberIndex is speciﬁed, all halted RAID members on the span with that MemberIndex will resume as well. If no MemberIndex is speciﬁed, all halted rebuilds on that RAID group will be resumed.

RGResumeRebuild [GroupName] <MemberIndex>

• RGSectorSize

Sets the sector size of the speciﬁed RAID group. The desired RAID group sector size must be evenly divisible by the sector size of any member disk. 512 bytes is the default size for most operating systems. Use 4KB sectors to enable large volume support (greater than 2TB, up to 16TB) in Windows XP (32-bit).

set RGSectorSize [GroupName] [512-8192] get RGSectorSize [GroupName]

• RGSpanDepth Regulates the span depth on the speciﬁed existing new RAID group for all RAID conﬁgurations except JBOD, which implicitly supports spanning as members are added.

set RGSpanDepth [GroupName] [SpanDepth [1-16]] get RGSpanDepth [GroupName]

• RGSpeedRead Performs look-ahead during reads from RAID group member disks for all or the speciﬁed RAID group. GroupName is the ASCII name of the RAID group for which look-ahead reads will be performed. Auto will choose the algorithm based on each I/O command.

set RGSpeedRead [GroupName | all] [enabled | disabled | auto] get RGSpeedRead [GroupName | all]

• RGResumeErase (Immediate) Continues the erase on the speciﬁed existing RAID group.

RGResumeErase [GroupName]

Appendix A - CLI ASCII-Based Interface

• RGUnmap (Immediate) Removes all mapped partitions of the speciﬁed RAID group from the routing table. The partitions themselves will be unaffected, though they will now be inaccessible to any initiators.

RGUnmap [GroupName | all]

• RGWaitTimeout (Disabled on Error) Speciﬁes the maximum time in seconds that will elapse to discover previously conﬁgured RAID groups. The timeout is used during system boot time and when the BlockDevScan command is issued.

set RGWaitTimeout [1-300] get RGWaitTimeout

• RMStatus (Immediate) Displays the status of all RAID members within the speciﬁed RAID group or a speciﬁc RAID member (if speciﬁed) within the speciﬁed RAID group. This command will fail if the speciﬁed RAID group does not exist or a speciﬁed member index within the RAID group does not exist. GroupName is the ASCII name of the RAID group for which status will be displayed.

RMStatus [GroupName] <MemberIndex>

• SASTargets (Immediate, Disabled on Error) Lists the physical devices that are connected to all SAS ports.

SASTargets

• SaveConfiguration (Immediate, Disabled on Error) Saves conﬁguration changes. Please note that certain modiﬁcations require a system restart.

SaveConfiguration

• SerialNumber Displays the serial number. The serial number is a 13-character ﬁeld. The ﬁrst seven alphanumeric characters are an abbreviation representing the product name; the remaining six digits are the individual unit’s number.

get SerialNumber

• SES (Enabled, Disabled on Error)

Enables support for SES enclosures that have been discovered by the appliance.

set SES [enabled | disabled] get SES

• Route (Immediate, Disabled on Error) Maps a RAID partition to a Target ID on the host system. If a map with the speciﬁed Target ID already exists, then it will be overwritten. Use delete for a Target ID to remove the map.

Route host [tid] [ [RAID [GroupName] [PartIdx]] | [SAS [SasIdx]] | delete]

• RouteDisplay (Immediate) Displays a list of host protocol address to target destination device mappings. The optional tid parameter will limit the list to the maps which satisfy a search for the given Target ID. If the passthrough parameter is entered, then all SATA/SATA passthrough devices are displayed. Otherwise all mapped RAID partitions are displayed. If the persistent parameter is entered, then only persistent maps will be displayed. Otherwise, both persistent and non-persistent maps will be displayed.

RouteDisplay host <tid> | <<passthrough> <persistent>>>

• SasPortList (Immediate)

Lists the status of all available SAS ports.

SasPortList

• SESAlarmTest (Immediate, Disabled on Error)

Commands the speciﬁed enclosure’s audible alarm to be turned on at the speciﬁed warning level. Reset turns off the alarm that has been set at any warning level. Note that SESEnclosures must be executed prior to executing SESAlarmTest.

SESAlarmTest [EnclIdx] [set | reset] [info | non-crit | crit | unrecov]

• SESDiskFailureAlarm (Disabled, Disabled on Error)

Activates an audible alarm when the RAID controller determines that a RAID member disk drive has failed. The buzzer within the

enclosure which contains the failed disk drive will be activated, other enclosures will be unaffected.

set SESDiskFailureAlarm [enabled | disabled] get SESDiskFailureAlarm

• SESEnclosures (Immediate, Disabled on Error)

Displays a list of SES-enabled enclosures which have been discovered by the RAID controller.

SESEnclosures

xii

Appendix A - CLI ASCII-Based Interface

• SESIdentify (Disabled on Error)

Commands the appropriate SES enclosure to identify the speciﬁed element(s). All identiﬁes all disks. RAID and RAID group name identiﬁes all disks in a RAID group. If the MemberIndex is also speciﬁed, only that disk is identiﬁed. Drive and BlockDevID identiﬁes the speciﬁed disk.

set SESIdentify [all | RAID GroupName <MemberIndex> | drive BlockDevID] get SESIdentify

• SESIdentifyStop (Immediate, Disabled on Error)

Commands the appropriate SES enclosure to stop identifying the speciﬁed element(s). All stops identifying all enclosures’ drive slots. RAID and RAID group name stops identiﬁcation of all disks in a RAID group. Drive and BlockDevID stops identiﬁcation of the speciﬁed drive.

SESIdentifyStop [All | RAID GroupName <MemberIndex> | Drive BlockDevID]

• SESMute (Immediate, Disabled on Error)

Causes all known enclosures’ audible alarms to be set to either the mute or remind state. The default action is mute. The enclosure index qualiﬁer is optional. The optional parameter Remind may be speciﬁed to set the remind state, which causes an occasional audible reminder of the alarm condition (if supported). Note that SESEnclosures must be executed prior to executing SESMute.

SESMute <EnclIdx> <Remind>

• SESStatus (Immediate, Disabled on Error)

Displays the last polled status of the speciﬁed element type in the speciﬁed enclosure. SupportLevel indicates the SES features supported by the speciﬁed enclosure: fan, power, temp, alarm, drive LEDs. If no element type is speciﬁed, all status is displayed. Note that SESEnclosures must be executed prior to executing SESStatus.

• Time (Disabled on Error) Sets or displays the current time in 24 hour format.

set Time [HH: MM: SS] get Time

• TimeZone Sets or displays the time zone or an offset from GMT. GMT offset must be in the format +/-HH:MM

set TimeZone [[EST | CST | MST | PST] | [[+|-] [HH]:[MM]]] get TimeZone

• VerboseMode (Enabled) Controls the level of detail in CLI Help output and command response output.

set VerboseMode [enabled | disabled] get VerboseMode

• SESPoll (Enabled, Disabled on Error)

Speciﬁes the SES enclosure polling interval in seconds. At the speciﬁed interval, all known SES enclosures are polled for their current status. A setting of 0 disables SES enclosure polling.

Default: 60 set SESPoll [0 | 30-3600 get SESPoll

• SESStartingSlot (Disabled on Error)

Establishes the starting slot/ID number for all attached SES enclosures.

set SESStartingSlot [0 | 1] get SESStartingSlot

• VirtualDriveInfo (Immediate) Displays characteristics and statistics for all the available virtual drives or any available virtual drive identiﬁed by its virtual drive ID.

VirtualDriveInfo <Virtual Drive ID>

• WrapEventLog (Enabled, Disabled on Error) When enabled, the unit logs up to 2,048 event entries before wrapping (overwriting the ﬁrst entries). If disabled, the unit stops logging event entries when the buffer is full.

set WrapEventLog [enabled | disabled] get WrapEventLog

xiii

Appendix B - Drive Reformat Instructions

If your Sonnet Fusion RAID storage system shipped with pre-installed hard disk drives, the following information will assist you to reformat the drives per your needs.

In Fusion RAID systems shipped from Sonnet with hard disk drives installed, the drives are formatted Mac OS Extended (Journaled), and conﬁgured as a single RAID 5 or RAID 6 RAID group. If you need to change the conﬁguration, use the ATTO Conﬁguration Tool and the operating system software tools to reformat and reconﬁgure the drives.

Mac OS Users’ Instructions

1. Follow all the steps in the included documentation to install the software and the Sonnet RAID controller, and to set up and connect the Fusion RAID drive enclosure.

2. Start your computer, and then turn on the Fusion drive enclosure; the RAID volume should appear on the desktop.

3. Drag the volume to the trash (changes to an eject icon) to eject it.

4. Launch the ATTO Conﬁguration Tool.

5. Select ExpressSAS Rxxx in the Device Listing pane, and then click the RAID tab in the Conﬁguration Options pane.

6. Select RAID Management > Delete Group from the application menu.

11. Click Erase; a window will appear asking you to approve your choice.

12. Click Erase.

13. Repeat steps 10–12 for each remaining unformatted RAID group.

14. Close Disk Utility.

15. Depending on how you conﬁgured the RAID groups, the volumes may already be available to the system. If you created a DVRAID, RAID 4, RAID 5, or RAID 6 RAID group, conﬁguration will take much longer. You may check on the process by double-clicking the volume name in the lower pane of the ATTO Conﬁguration Tool window.

16. Once all the RAID groups have been formatted and ﬁnish building, they are ready to use.

Windows 7/Server 2008/Vista Users’ Instructions

1. Follow all the steps in the included documentation to install the software and the Sonnet RAID controller, and to set up and connect the Fusion RAID enclosure.

2. Start your computer, and then turn on the Fusion drive enclosure.

3. Launch the ATTO Conﬁguration Tool.

7. When the Delete Conﬁrmation window appears, click Yes.

8. Set up new RAID groups following the instructions on pages 9 and 10 of this manual.

WARNING: After selecting the DVRAID, RAID Level 4,

RAID Level 5, or RAID Level 6 option, configuration of the drives will take several hours to complete, with the total time depending on the operating system and drive capacities.

9. Depending on how you conﬁgure your setup, a Disk Insertion window will appear at some point stating that there

is an unreadable volume; click Initialize, and then Disk Utility will open.

10. In the Disk Utility window, each RAID group you created using the ATTO Conﬁguration Tool will appear as a single volume. Select the volume, and then click the Erase tab at the top of the window.

Support Note for Power Mac G5 Users: When

creating RAID groups 16TB or larger, uncheck the Install Mac OS 9 Drivers checkbox; OS 9 drivers do not support volumes greater than 16TB.

4. Expand the device tree in the Device Listing pane until ExpressSAS Rxxx is displayed, and then click ExpressSAS Rxxx.

5. Click the RAID tab.

6. In the Groups pane, you will see the RAID group that Sonnet created (not usable in a Windows system); click the group.

7. Select RAID Management > Delete Group from the application menu.

8. When the Delete Conﬁrmation window appears, click Yes.

9. Set up new RAID groups following the instructions on pages 9 and 10.

WARNING: After selecting the DVRAID, RAID Level 4,

RAID Level 5, or RAID Level 6 option, configuration of the drives will take several hours to complete, with the total time depending on the operating system and drive capacities.

xiv

Appendix B - Drive Reformat Instructions

Windows 7/Server 2008/Vista Users’ Instructions (continued)

10. Click Start, then right-click Computer and select Manage.

11. In the Computer Management window, click Storage in the

left pane to expand the list (if necessary), and then click Disk Management.

12. When the Initialize Disk window appears, select the RAID volume you created. Select the GPT partition style unless you need to access your RAID storage from a computer running 32-bit Windows XP Professional or 32-bit Windows Server

2003. Click OK.

13. In the Disk Management window, each RAID group you created will appear (listed as “unallocated”) as a single volume. Right-click where the word “unallocated” appears, and then select New Simple Volume.

14. When the Welcome to the New Simple Volume Wizard window appears, click Next to start the process.

15. When the New Simple Volume Wizard window appears, click Next.

Windows XP/Server 2003 Users’ Instructions

1. Follow all the steps in the included documentation to install the software and the Sonnet RAID controller, and to set up and connect the Fusion RAID enclosure.

2. Start your computer, and then turn on the Fusion drive enclosure.

3. Launch the ATTO Conﬁguration Tool.

4. Expand the device tree in the Device Listing section on the left side of the window until ExpressSAS Rxxx is displayed, and then click ExpressSAS Rxxx.

5. Click the RAID tab.

6. In the Groups pane, you will see the RAID group that Sonnet created (not usable in a Windows system); click the group.

7. Select RAID Management > Delete Group from the application menu.

8. When the Delete Conﬁrmation window appears, click Yes.

16. When the Specify Volume Size window appears, click Next if you want all of the Fusion system’s capacity to remain as one block (volume). Otherwise, adjust the volume size to meet your needs, and then click Next.

17. When the Assign Drive Letter or Path window appears, select Assign the following drive letter, choose a letter, and then click Next.

18. When the Format Partition window appears, enter a new name for the volume table if you’d like. For RAID volumes up to 16TB, accept the default allocation unit size; for RAID volumes greater than 16TB, select 8192 from the drop-down menu. Select Perform a quick format, and then click Next.

Note: If you do not select the quick format option, this process will take

much longer to complete.

19. When the next window appears, click Finish.

20. Repeat steps 13–19 for each remaining “unallocated” disk.

21. Depending on how you conﬁgured the RAID groups, the volumes may already be available to the system. If you created a DVRAID, RAID 4, RAID 5, or RAID 6 RAID group, conﬁguration will take much longer. You may check on the process by double-clicking the volume name in the lower pane of the ATTO Conﬁguration Tool window.

22. Once all the RAID groups have been formatted and ﬁnish building, they are ready to use.

9. Set up new RAID groups following the instructions on pages 9 and 10.

WARNING: After selecting the DVRAID, RAID Level 4,

RAID Level 5, or RAID Level 6 option, configuration of the drives will take several hours to complete, with the total time depending on the operating system and drive capacities.

10. Select Start > Control Panel from the Windows Start menu. In the Control Panel window, double-click Administrative Tools. In the Administrative Tools window, double-click Computer Management.

11. In the Computer Management window, click Storage on the left, and then click Disk Management beneath it.

12. When the Initialize and Convert Disk Wizard window appears, click Cancel.

13. In the Disk Management window, each RAID group you created will appear (listed as “unallocated”) as a single volume. Right-click one volume where the words “Not Initialized” appear, and then select Initialize Disk.

14. When the Initialize Disk window appears, select the RAID volume(s), and then click OK.

15. Back in the Disk Management window, right-click where the word “Online” appears, and then select Convert to GPT Disk.

Appendix B - Drive Reformat Instructions

Windows XP/Server 2003 Users’ Instructions (continued)

16. Right-click where the word “unallocated” appears, and then select New Partition.

17. When the New Partition Wizard window appears, click Next.

18. When the Select Partition Type window appears, select Primary

Partition, and then click Next.

19. When the Specify Partition Size window appears, click Next if you want all of the Fusion system’s capacity to remain as one block (volume). Otherwise, adjust the volume size to meet your needs, and then click Next.

20. When the Assign Drive Letter or Path window appears, choose a letter, and then click Next.

21. When the Format Partition window appears, enter a new name for the volume if you’d like. For RAID volumes up to 16TB, accept the default allocation unit size; for RAID volumes greater than 16TB, select 8192 from the drop-down menu. Select Perform a quick format, and then click Next.

Note: If you do not select the quick format option, this process will take

much longer to complete.

22. When the next window appears, click Finish.

23. Repeat steps 15–22 for each remaining “unallocated” disk.

24. Depending on how you conﬁgured the RAID groups, the volumes may already be available to use. If you created a DVRAID, RAID Level 4, or RAID Level 5 group, conﬁguration will take much longer. You may check on the process by double-clicking the RAID group name in the lower pane of the ATTO Conﬁguration Tool window.

25. Once all the RAID groups have been formatted and ﬁnish building, they are ready to use.

xvi

xvii

Remember to register your product online at http://registration.sonnettech.com to be informed of future upgrades and product releases. Software updates and links are available from the Sonnet web site at www.sonnettech.com. • Online support form available at http://supportform.sonnettech.com. Sonnet Technologies Customer Service hours are Mon.-Fri., 7 a.m.–4 p.m. Pacific Time • Customer Service Phone: 1-949-472-2772 • Email: support@sonnettech.com Sonnet Technologies, Inc., California USA • Tel: 1-949-587-3500 • Fax: 1-949-457-6349

©2010 Sonnet Technologies, Inc. All rights reserved. Sonnet, the Sonnet logotype, Simply Fast, the Simply Fast logotype, and Fusion are trademarks of Sonnet Technologies, Inc. Mac, the Mac logo, and Mac OS are trademarks of Apple Inc., registered in the United States and other countries. Other product names are trademarks of their respective owners. Product specifications subject to change without notice. Printed in the USA. MNL-RAID-SW-E-G-102610

Sonnet Technologies RAID User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual