Sun Fire V215 Server
 
Sun Fire V245 Server
 

PLA-Support>Sun Systems>SPARC>Workgroup Servers>SN-SPARC: SF-V2x0
 
_Old GCS Categories>Sun Microsystems>Servers>Entry-Level Servers
 

Applies to:

Goal

This document provides the basic information of Hardware RAID functionality on the Sun Fire[TM] V215 and V245 servers

Solution

Steps to Follow

Hardware - LSI SAS1064 4 Phy-wide SAS HBA built into the motherboard.

The kernel module is "mpt".

Each LSI1064 HBA supports up to 2 RAID volumes at any one time.

Supported RAID Operations - Support for RAID0 (striping) and RAID1 (mirroring)

RAID1 - mirroring

• A 2-disk primary->secondary configuration
• All write I/Os are committed to both disks before returning to provide for data redundancy
• Read I/Os may be serviced by either disk
• Upon a single disk failure, the data is still available - the volume is in DEGRADED mode. The failed disk may be replaced and resync'd
• If the volume is disabled (mirror is broken) the drives need to be relabeled and all data is lost.
• When creating a RAID volume all data contained on the drives is lost. If mirroring of the boot drive is desired the RAID volume must be created before OS installation.

RAID0 - striping

• A 2 or more disk configuration
• No data is shared. Data is across all disks (concatenated) in exactly one location.
• Write I/Os are committed in a round-robin fashion across all member disks, performance increased in some workload scenarios
• Read I/Os are satisfied at their one location (performs as a single disk)
• Upon a single disk failure, data is lost
• If a RAID0 volume is disabled (broken), data is lost
• When creating a RAID volume all data contained on the drives is lost.

Management of RAID volumes by use of the Solaris raidctl command

• Create a mirror
  raidctl -c [-r 1] primary secondary

• Create a stripe
  raidctl -c -r 0 disk1 disk2 [disk3] ...

• Delete a RAID volume
  raidctl -d volume

• Display RAID volumes
  raidctl

• Update HBA controller firmware
  raidctl -F file controller

Examples

1.) Creating a mirror

• To create a RAID1 mirror
  raidctl -c [-r 1] primary secondary
  Example:

# raidctl -c c0t0d0 c0t1d0

• RAID volume c0t0d0 will be created
• Data on all drives will be lost
• No file systems may be mounted on either disk
• Secondary must be as large as primary
• RAID volume has the capacity of one disk

2.) Creating a stripe

• To create a RAID0 stripe
  raidctl -c -r 0 disk1 disk2 [disk3] ..
  Example:

# raidctl -c -r 0 c0t1d0 c0t2d0 c0t3d0

• RAID volume c0t1d0 will be created
• Data on all drives will be lost
• No file systems may be mounted on any disk
• RAID volume has the capacity of all disks
  combined - smallest common size is used

After RAID volume creation

• Member disk drives are no longer visible, only the RAID volume is presented
• The RAID volume may be addressed as any other LUN in Solaris
• The status of the RAID volume and it's member disks may be retrieved using raidctl(1M)
• Each LSI1064 HBA supports up to 2 RAID volumes at any one time
• Volumes must be labeled after creation before they can be used under Solaris

Labeling Volumes After Creation

After creating a RAID volume by using 'raidctl' or by using OBP level commands the volume must be labeled correctly before it can be used. This can be done using the 'format' command in Solaris:

As the root user run 'format' and choose the newly created volume from the list of drives.
(you may see some messages about the volume having a corrupt label or the block count being incorrect when booting or running format, this is due to the need for a correct label)
After the volume is selected chose "type" from the format menu, this will give you a list of available drive types to chose from.
Chose "0. Auto configure" from the list. This will define the volume with the correct type.
Then write the new label to the volume using the "label" command from the format> prompt. The system will ask you if you are ready to label the disk. type in "yes"

After you perform these steps the drive is ready for use. You can now partition the drive as needed and use the volume within Solaris normally.

Note also that after a RAID volume is deleted the drives that were previously part of the RAID volume should be re-labeled at that time for proper use also.

RAID Volumes Possible Status

OK - volume is optimal
DEGRADED - volume has a failed disk (mirror) or is otherwise functionally minimized
RESYNCING - volume is synchronizing (mirror)
FAILED - volume is offline

Disk Possible Status

OK - disk is operating normally
MISSING - disk is removed or otherwise offline
FAILED - disk is in a failed state and needs service

OBP Level Commands

PLEASE NOTE: OBP Level RAID commands are NOT SUPPORTED and are not intended for normal customer use. Future firmware updates may change command behavior and/or remove access to these commands altogether.

Before running the RAID OBP commands the auto-boot? variable needs to be set to false and the fcode-debug? variable needs to be set to true, then a reset-all is performed. After the reset finishes the controller must be selected. In the example below the "show-disks" command is used to determine the controller path. Note that the "/disk" was left off the path since it is the controller you want to select. (You can also use the probe-scsi-all command to find the path.)

{1} ok setenv auto-boot? false
auto-boot? =            false
{1} ok setenv fcode-debug? true
fcode-debug? =          true
{1} ok reset-all
:
: 
:
{1} ok show-disks
a) /pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1/disk
b) /pci@1e,600000/pci@0/pci@1/pci@0/ide@1f/cdrom
c) /pci@1e,600000/pci@0/pci@1/pci@0/ide@1f/disk
q) NO SELECTION 
Enter Selection, q to quit: a
/pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1/disk has been selected.
Type ^Y ( Control-Y ) to insert it in the command line. 
e.g. ok nvalias mydev ^Y 
         for creating devalias mydev for    /pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1/disk

{1} ok select /pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1

The "show-volumes" command will show if there are any existing volumes and their state.

{1} ok show-volumes
Volume 0 Target 0  Type IM (Integrated Mirroring)
  Optimal  Enabled 
  2 Members                                         143112591 Blocks, 73 GB
  Disk 1 
    Primary  Online 
    Target 4        FUJITSU MAY2073RCSUN72G 0401
  Disk 0 
    Secondary  Online 
    Target 1        FUJITSU MAY2073RCSUN72G 0401
{1} ok 

If you wanted to create a volume you would select the controller as above and use the "create" command. The following example creates a mirrored volume with disks 0 and 1.

{0} ok 0 1 create-im-volume

  Target 0 size is 143243264 Blocks, 73 GB

  Target 1 size is 143243264 Blocks, 73 GB

  The volume can be any size from 1 MB to 69943 MB 

  What size do you want?  [69943]

  Volume size will be 143243264 Blocks, 73 GB

  PhysDisk 0 has been created for target 0 

  PhysDisk 1 has been created for target 1 

  Volume has been created

{0} ok

This command creates a striped volume using disks 0 and 1

{0} ok 0 1 create-is-volume

  Target 0 size is 143243264 Blocks, 73 GB

  Target 1 size is 143243264 Blocks, 73 GB

  The volume can be any size from 1 MB to 139886 MB

  What size do you want?  [139886]

  Volume size will be 286486528 Blocks, 146 GB

  PhysDisk 0 has been created for target 0 

  PhysDisk 1 has been created for target 1 

  Volume has been created

{0} ok

To unselect

{0} ok unselect-dev

when you have completed managing the RAID volumes you can set the auto-boot? and fcode-debug? variables back and reset the system:

{1} ok setenv auto-boot? true
auto-boot? =            true
{1} ok setenv fcode-debug? false
fcode-debug? =          false
{1} ok reset-all

Remember that even if you create a RAID volume at the OBP level you will still need to label the volume using format in Solaris before they can be properly used.

If you have existing RAID volumes and have to replace the system board the
following commands will need to be run to see the volume with the new board

{1} ok show-disks
a) /pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1/disk
b) /pci@1e,600000/pci@0/pci@1/pci@0/ide@1f/cdrom
c) /pci@1e,600000/pci@0/pci@1/pci@0/ide@1f/disk
q) NO SELECTION 
Enter Selection, q to quit: q
/pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1/disk has been selected.
Type ^Y ( Control-Y ) to insert it in the command line. 
e.g. ok nvalias mydev ^Y 
         for creating devalias mydev for  /pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1/disk

{1} ok select /pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1

{0} ok show-volumes ( Look for the inactive volume)
{0} ok X activate-volume (X is the volume number)
{0} ok unselect-dev 

If activating multiple volumes ensure they are activated in descending order:

{0} ok show-volumes
{0} ok 1 activate-volume
{0} ok 0 activate-volume
{0} ok unselect-dev
{0} ok probe-scsi-all

Failure to do so will result in the following error and will require a reset-all prior to activating the second volume:

{0} ok 0 activate-volume
Volume 0 is now activated
{0} ok 1 activate-volume
RAID volume request failed, ioc-status = 0007
Volume 1 could not be activate

The first volume activated will remain 'Degraded' until both volumes have completed syncing.

NOTE: IF the EEPROM variable "auto-boot?" was set to true you may get the following message.

{0} ok select /pci@1e,600000/pci@0/pci@a/pci@0/pci@8/scsi@1

Base SAS World Wide ID is 0!

This must be fixed immediately using set-sas-wwid
{0} ok 

If you do see this message set "auto-boot?" to false and issue a "reset-all"

Internal Comments
Audited/updated 12/03/09 - Ian.Macdonald@Sun.COM, Entry Level SPARC Content Team Member




raid, disk, mirror, stripe, raid0, raid1, raidctl, format, v215, v245, sync, degraded
Previously Published As 87435

Change History
No significant change history recorded
Date: 2007-07-06
User Name: 109125
Action: Update Canceled
Comment: *** Restored Published Content *** I have moved the document to other technical subject.
This update has been canceled. Sorry


Product_uuid
dd2799dd-e6e5-11da-ac3d-080020a9ed93|Sun Fire V245 Server

This solution has no attachment