Skip to main content

14.2.3 Precautions Concerning Hardware RAID

14.2.3 Precautions Concerning Hardware RAID
This section describes the precautions when you use hardware RAID.
Notes When Using Hardware RAID
  1. Perform backup of important data and programs on a regular basis. For some errors, you may need to restore data or programs from backup media by reconfiguring the hardware RAID.
  2. We recommend using an uninterruptible power supply (UPS) to ensure that the data in the system can be secured in the case of a power outage.
  3. If high availability such as duplication of controllers or data paths is required for the hardware RAID function, install a dedicated RAID system.
  4. Disk redundancy using hardware RAID is possible only under one SAS controller.
SAS2IRCU Utility
  1. To manage the hardware RAID environment of the SPARC M12/M10, we recommend using the SAS2IRCU utility for the following reasons.
    - SAS2IRCU: You can configure and manage a hardware RAID while logical domains are running.

    - OpenBoot PROM environment: You can configure a hardware RAID only when logical domains are stopped, but cannot manage it.
Note - If you use or plan to use the DR function in the SPARC M12-2S/M10-4S, be sure to install the SAS2IRCU utility. After DR integration, the SAS2IRCU utility is needed to enable RAID volumes.
  1. If the RAID volume fails, you can specify the failed disk by using the SAS2IRCU utility.
  2. The SAS2IRCU utility is used in the Oracle Solaris environment and requires root account privileges. Therefore, the SAS2IRCU utility should be operated by a system administrator.
  3. The SAS2IRCU utility is used in the logical domain where the SAS controller is assigned. Therefore, install the SAS2IRCU utility in the I/O root domain where the SAS controller on the motherboard is assigned.
Note - The assignment function for PCIe endpoint devices is not supported on the SAS controller in the SPARC M12/M10.
Notes When Configuring/Removing Hardware RAID
  1. When configuring or removing the hardware RAID, data reliability on a disk is not guaranteed. Be sure to back up data before you initially configure the hardware RAID for the system or remove an installed hardware RAID. After configuring the hardware RAID, you may need to install new data or restore data from backup media.
  2. Execution of hardware RAID reconfiguration changes the RAID volume-specific information (WWID value). So, if you perform an operation like activation device specification or Oracle Solaris mounting processing, the setting needs to be changed.
  3. The following table shows the standard time for synchronizing a 600 GB, 900 GB, or 1.2 TB disk drive in the unloaded state due to the configuration or maintenance of the hardware RAID. The synchronization time varies widely depending on the generation and usage state of the disk drives or the load state of systems.
Table 14-8  Standard Synchronization Time for the Hardware RAID
RAID Number of Disk Drives Synchronization Time
600 GB
Disk Drive
900 GB
Disk Drive
1.2 TB
Disk Drive
1 2
6 hours
8 hours
10 hours
10 (*1)
10 hours
15 hours
21 hours
16 hours
21 hours
29 hours
21 hours
31 hours
44 hours
1E 3
20 hours
29 hours
25 hours
35 hours
45 hours
32 hours
54 hours
65 hours
38 hours
*1 Only the SPARC M12 supports RAID10.  
  1. Do not perform tasks which require you to access disk drives, such as Oracle Solaris installation or data restore, while the disk drives are configuring or synchronizing the hardware RAID. From the point of view of RAID volume safety, we do not recommend rewriting the contents of disk drives before completing the RAID volume configuration. Be sure to perform these tasks after completing RAID volume configuration.
  2. Only one RAID volume can be configured at a time per SAS controller. When another set of commands to configure the RAID volume are issued for an SAS controller where RAID volume configuration is in process, completion of the first configuration task is followed by the next RAID volume configuration.
  3. When you configure the hardware RAID, the RAID volume becomes smaller in size than the original disk.
  4. If the system is restarted or a power outage has occurred and power is recovered while the hardware RAID is being configured or synchronized, start the configuration/synchronization procedure again for the SPARC M10. For the SPARC M12, the configuration/synchronization procedure is restarted from the point when the outage occurred.
Notes about Operating Hardware RAID
Your system may slow down because the hardware RAID controller cannot completely judge whether a disk has a failure. If your system has this problem, follow the procedures below. Be sure to first back up your data before performing the procedures because the hardware RAID will be removed.
1. Stop applications from using internal storage.

2. Remove the hardware RAID.

3. Identify whether the problem is related to the disk.

4. If the problem is not solved, replace all disks used for the hardware RAID.

5. Reconfigure the hardware RAID.

6. Restore data from backup media.
Next, the operation procedures for hardware RAID volumes are described. See the appropriate item indicated depending on the operation you will perform.
Table 14-9  Hardware RAID Operations and References
RAID Volume Operation References
Configuring a RAID volume "14.2.4 Preparation Before Hardware RAID Operation"
"14.2.5 Creating a Hardware RAID Volume"
"14.2.7 Managing a Hot Spare of a Hardware RAID Volume"
Deleting a RAID volume "14.2.4 Preparation Before Hardware RAID Operation"
"14.2.6 Deleting a Hardware RAID Volume"
Replacing disk drives that make up a RAID volume "14.2.9 Checking for a Failed Disk Drive"
"14.2.10 Replacing a Failed Disk Drive"
Re-enabling a hardware RAID after replacing CPU memory units (lower) of the SPARC M12-2/M12-2S/M10-4/M10-4S and motherboard units of the SPARC M12-1/M10-1 systems "14.2.11 Re-enabling a Hardware RAID Volume"
"14.2.12 Specifying a Hardware RAID Volume as a Boot Device"