advertisement
E Frequently Asked Questions
Smart Array 6400 Series Controller FAQ
Q: Are simultaneous internal and external connections supported for one channel?
A: No, simultaneous internal and external connections are not supported for one channel. For example, you cannot have both an internal connection for channel A and an external connection for channel A. However, you can have an internal connection for channel A and an external connection for channel B.
Q: How many Smart Array Controllers can I install in my system?
A: The maximum number of controllers that you can install depends on the system and several other factors specific to your configuration.
The maximum number of controllers is restricted to the number of PCI-X or PCI slots not used for other peripherals. The power rating of the system also limits the number of controllers. Each
Smart Array Controller requires 12 -18 W. The system must be capable of supplying each controller with 12 -18 W.
See the documentation for your system for more information.
Q: How many external drives can I attach to my system?
A: Each channel of a Smart Array controller supports up to 14 disk drives. The Smart Array 6402 has two channels supporting up to 28 disks. The Smart Array 6404 has four channels supporting up to 56 disks.
Q: Does the Smart Array 6402 Series Controller support SCSI tape drives and CD-ROM drives?
A: No
Q: What is the data transfer rate for the Smart Array Controllers?
A: Each channel of the Smart Array Controller supports Ultra320 speeds. Ultra320 SCSI has a bandwidth of 320 MB/s. Most server applications do not take advantage of the full Ultra320 bandwidth, so performance results vary from system to system.
Q: What is the difference between LVD and Ultra320 SCSI?
A: Low Voltage Differential (LVD) is a signaling level for SCSI protocols. Ultra320 is a SCSI protocol that uses LVD signaling. Ultra60 and Ultra2 also use LVD signaling, but earlier SCSI protocols use Single-Ended (SE) signaling.
Q: I have several hard drives in my server and in an HP StorageWorks Enclosure 4300. What
SCSI IDs do I assign for these drives?
A: If you are using hot-pluggable disks in a StorageWorks 4300 series enclosure, or an MSA 30, the SCSI IDs are set automatically according to the bay in which the drives are installed and do not need to be set manually.
Non-hot-pluggable disks must each have a unique SCSI ID value ranging from 0 to 15. Do not use ID 7, which is reserved for controller use. The SCSI ID value determines the priority given to the device when it attempts to use the SCSI bus.
Q: Can I use third-party (non-HP) drives with the Smart Array 6402 Controller?
A: No.
Q: Why do the drive activity LEDs light up on some drives when my system is idle?
A: The Smart Array 6402 Series Controller performs background activities on the drives when the controller is otherwise idle. For example, Auto-Reliability Monitoring (ARM) scans fault-tolerant volumes for defects and verifies the consistency of parity data, and Drive Parameter
Tracking periodically checks the performance of all drives on the controller (normally on an hourly basis).
Q: What is RAID ADG?
A: RAID ADG is an extension of RAID 5 that enables additional fault tolerance by using two different and independent parity schemes. Data is striped across a set of hard drives, just as with
Smart Array 6400 Series Controller FAQ 77
RAID 5, and the two sets of parity data are calculated and written across all the drives in the array.
RAID ADG provides an extremely high level of fault tolerance and can sustain two simultaneous drive failures without downtime or data loss. This fault tolerance level is useful for mission-critical data.
Q: What does the auto-fail missing disks at boot option control?
A: The auto-fail missing disks at boot option controls the power-on behavior of the
HP-UX RAID controller when some of the configured drives are missing. Auto-fail is enabled when the first logical drive is created by the saconfig utility. You can then disable it with the saconfig -F command.
As an example, consider the following scenario:
1.
The boot volume is on a RAID logical drive.
2.
The system is powered off.
3.
The cable for an enclosure containing configured disks is accidentally disconnected from the controller.
4.
All disks belonging to the boot volume are still connected to the controller.
5.
The system is then powered on.
If auto-fail is enabled in this scenario: During POST, the controller fails the missing disks.
Non-fault-tolerant logical drives are FAILED. Fault-tolerant logical drives either transition to
Interim Recovery (degraded) mode or to FAILED depending on the number of disks the logical drive is missing. The system then begins booting HP-UX.
During boot, the ciss init script detects the degraded/failed logical drives and generates an error that instructs you to run sautil. The sautil utility displays the degraded/failed logical drives and failed disks. At this point, you can either power off the system, reconnect the disks, and boot again; or reconnect the disks and run the sautil accept media exchange command to change the disks and logical drives back to OK state. For more information, see
<device_file> accept_media_xchg <logical_drive_number> Command” (page 63)
.
If auto-fail is disabled in the previous scenario:[HP/9000 system] The controller temporarily disables all logical drives, including the intact boot volume. The system fails to boot.
[HP Integrity system] During POST, you can choose one of the following options:
1.
Power off the system and reconnect the disks.
2.
Press F1. The controller temporarily disables all logical drives, including the intact boot volume. The system fails to boot.
3.
Press F2. The system takes the same actions as system with auto-fail enabled, as previously described.
NOTE:
HP recommends that you leave auto-fail enabled if the RAID controller is in an HP/9000 system that boots from a RAID volume.
78 Frequently Asked Questions
Glossary
array array capacity expansion
A set of physical disks configured into one or more logical drives. Arrayed disks have significant performance and data protection advantages over non-arrayed disks.
array accelerator
A component of some Smart Array Series controllers that dramatically improves disk read and write performance by providing a buffer. Data integrity is protected by a backup battery and
ECC memory.
See capacity expansion.
Array
Configuration
Utility (ACU)
A configuration utility useful both for novices and for experienced RAID users.
Array Diagnostic
Utility (ADU)
A diagnostic tool that collects comprehensive information about the array controllers in a system and lists any problems detected.
Auto-Reliability Monitoring (ARM)
Automatic Data
Recovery cache capacity expansion capacity extension
CISS controller duplexing data guarding data striping drive mirroring
Error Correction and Checking
(ECC) memory fault tolerance
Also known as surface analysis. A fault management feature that scans physical disks for bad sectors. Data in the faulty sectors remaps onto good sectors. Also checks parity data consistency for disks in RAID 5 or RAID ADG configurations. Operates as a background process.
A process that automatically reconstructs data from a failed disk and writes it onto a replacement disk. Automatic Data Recovery time depends on several factors, but HP recommends that you allow at least 15 minutes per gigabyte. Also known as rebuild.
A high-speed memory component, used to store data temporarily for rapid access.
The addition of physical disks to a pre-existing disk array, and redistribution of existing logical drives and data over the enlarged array. The size of the logical drives does not change. Also known as array capacity expansion.
The enlargement of a logical drive without disruption of data. There must be free space on the array before capacity extension can occur. If necessary, create free space by deleting a logical drive or by carrying out a capacity expansion. Also known as logical drive capacity extension.
Command Interface for SCSI-3 Support Specification.
A type of fault tolerance that requires two Smart Array Series controllers. Each controller has its own set of disks, and the disk sets have identical data. When one controller fails, the other automatically takes over the servicing of requests. Controller duplexing can be done with LVM
MirrorDisk/UX.
See RAID.
Writing data to logical drives in interleaved chunks (by byte or by sector). Data striping improves system performance.
See RAID.
A type of memory that checks and corrects single-bit or multi-bit memory errors (depending on configuration) without causing the server to halt or to corrupt data.
flashing hot spare interim data recovery logical drive
The ability of a server to recover from hardware problems without interrupting server performance or corrupting data. Hardware RAID is most commonly used, but there are other types of fault tolerance, including controller duplexing and software-based RAID.
Updating the flash memory on a system. Flash memory is nonvolatile memory used to hold control code such as BIOS information. Flash memory is very fast because it can be rewritten block by block, rather than byte by byte.
See online spare.
If a disk fails in RAID 1, 1+0, 5 or ADG, the system still processes I/O requests, but at a reduced performance level.
A group of physical disks, or part of a group, that behaves as one storage unit. Each constituent physical disk contributes the same storage volume to the total volume of the logical drive. A
79
logical drive has performance advantages over individual physical disks. Also known as a logical volume.
See capacity extension.
logical drive capacity extension
Low Voltage
Differential
(LVD) online spare
PCI-X rebuild
Redundant Array of Independent
Disks (RAID)
A type of SCSI signaling that enables a maximum transfer rate of either 80 MB/s or 160 MB/s, conforming to either the Wide Ultra2 or Wide Ultra160 SCSI standards, respectively.
A fault-tolerant system that normally contains no data. When any other disk in the array fails, the controller automatically rebuilds the data that was on the failed disk onto the online spare.
Also known as a hot spare.
An enhanced PCI bus that enables operation at 133 MHz, equivalent to a data throughput of
1 GB/s. PCI-X is backward compatible with PCI systems and devices that operate at 66 MHz or 33 MHz.
See Automatic Data Recovery.
A form of fault tolerance. RAID 0 (no fault tolerance) uses data striping to distribute data evenly across all physical disks in the array, but has no redundant data. RAID 1+0 (disk mirroring) duplicates data from one disk onto a second disk. RAID 5 (distributed data guarding) distributes parity data across all disks in the array, and uses the parity data and data on remaining disks to reconstruct data from a failed disk. RAID ADG (advanced data guarding) is similar to RAID
5, but uses two independent sets of parity data. For more information, see the RAID Technology
Overview at:
http://docs.hp.com/en/netcom.html#Smart%20Array%20%28RAID%29
SCSI ID
Self-Monitoring,
Analysis, and
Reporting
Technology
(S.M.A.R.T.)
A unique ID number assigned to each SCSI device connected to a SCSI bus. The ID number determines the device priority on the SCSI bus; ID 7 is the highest priority and is always assigned to the SCSI controller.
Technology codeveloped by HP and the physical disk industry that provides warning of imminent disk failure. S.M.A.R.T. enables HP to offer Pre-Failure Warranty replacement of physical disks. S.M.A.R.T. supersedes the disk parameter tracking feature that was previously used, because the self-monitoring routines used in S.M.A.R.T. are more accurate than the disk parameter tracking tests. The self-monitoring routines are customized for each specific disk type and have direct access to internal performance, calibration, and error measurements.
Simple Network
Management
Protocol (SNMP)
Governs network management and the monitoring of network devices and functions.
Single-Ended (SE)
A type of SCSI signaling that enables a maximum transfer rate of 40 MB/s. Conforms to the
Wide-Ultra SCSI standard. Now being phased out in favor of LVD technology.
spare striping
See online spare.
See data striping.
surface analysis
See Auto-Reliability Monitoring.
Ultra, Ultra2,
Ultra160, Ultra320
A set of SCSI standards that support maximum signal transfer rates of 40 MB/s, 80 MB/s, 160
MB/s, and 320 MB/s respectively.
Very High
Density Cable
Interconnect
(VHDCI)
An external SCSI connector used by Ultra SCSI controllers.
80 Glossary
81
*J6369-90051*
Printed in the US
advertisement
Related manuals
advertisement