Digital PDFs

EK-BA370-UG-C02

September 1998

206 pages

Original

3.9MB

Document:	DIGITAL StorageWorks UltraSCSI RAID Enclosure (DS-BA370-Series) User's Guide
Order Number:	EK-BA370-UG
Revision:	C02
Pages:	206
Original Filename:

OCR Text

Compaq StorageWorks
UltraSCSI RAID Enclosure
(DS–BA370 Series)
User's Guide
Part Number: EK–BA370–UG. C02/387403–001
September 1998
The DS–BA370 series UltraSCSI RAID rack-mountable enclosure is the
basic building block of UltraSCSI RAID subsystems. This guide and its
related publications comprise the basic documentation set for these
subsystems.

Compaq Computer Corporation
Houston, Texas

September 1998
While Compaq Computer Corporation believes the information included in
this publication is correct as of the date of publication, it is subject to change
without notice. Compaq makes no representations that the interconnection of
its products in the manner described in this document will not infringe on
existing or future patent rights, nor do the descriptions contained in this
document imply the granting of licenses to make, use, or sell equipment or
software in accordance with the description. Possession, use, or copying of the
software described in this documentation is authorized only pursuant to a valid
written license from Compaq, an authorized sublicensor, or the identified
licensor.
Compaq, DIGITAL, and the Compaq logo Registered in U.S. Patent and
Trademark Office.
CI, DSSI, and StorageWorks are trademarks of Compaq Computer
Corporation.
PROCOMM PLUS is a registered trademark of Data Storm Technologies, Inc.
© 1997, 1998 Compaq Computer Corporation. All rights reserved.
This equipment generates, uses, and may emit radio frequency energy. The
equipment has been type tested and found to comply with the limits for a
Class A digital device pursuant to Part 15 of the FCC rules, which are
designed to provide reasonable protection against such radio frequency
interference.
Operation of this equipment in a residential area may cause interference in
which case the user at his own expense will be required to take whatever
measures may be required to correct the interference.
Any modifications to this device—unless expressly approved by the
manufacturer—can void the user’s authority to operate this equipment under
part 15 of the FCC rules.

Note
Additional information on the need to interconnect the device
with shielded (data) cables or the need for special devices, such as
ferrite beads on cables, is required if such means of interference
suppression was used in the qualification test for the device. This
information will vary from device to device and needs to be
obtained from the EMC group or product manager.

Table of Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Manufacturer’s Declarations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
Chapter 1, Introducing the UltraSCSI RAID Enclosure
An UltraSCSI Bus Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–4
Product Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–5
UltraSCSI RAID Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–7
EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–7
PVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Master Enclosure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Master EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Expansion Enclosure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Expansion EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–9
I/O Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–9
SCSI Bus Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–9
Cache Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–10
ECB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–10
Enclosure Status. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–10
Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–11
Power. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–11
Redundant Power Supply Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–12
Optimum Power Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–12
Dual Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–13
UltraSCSI Bus Expansion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–13

UltraSCSI RAID Enclosure iii

Table of Contents

Chapter 2, Operating the UltraSCSI Enclosure
UltraSCSI Configuration Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–1
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–1
UltraSCSI RAID Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–2
UltraSCSI Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–3
UltraSCSI Enclosures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–4
Enclosure Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–4
Starting the Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–5
Applying Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–6
Turning On the Subsystem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–7
Shutting Down the Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–8
Standby Power Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–8
Full Power Shutdown. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–8
Using Standby Power Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–9
Turning Off the Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–10
Fault Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–12
Error Detection, Reporting, and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–12
Enclosure Status Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–13
SBB Shelf Status Indicators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–16
Power Supply Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–16
Disk Drive SBB Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–16
Disk Drive Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–17
Controller Shelf Status Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–18
Typical Controller Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–18
EMU Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–19
PVA Status LED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–19
I/O Module LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–20

Chapter 3, Environmental Monitoring Unit
Product Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Controller Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I/O Module Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Enclosure Configuration Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
EMU Front Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
EMU Communications Connectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
System Fault LED and Alarm Control Switch . . . . . . . . . . . . . . . . . . . . . . . . . .
Temperature Fault LED. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Power Status LED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

iv UltraSCSI RAID Enclosure

3–2
3–3
3–4
3–4
3–5
3–5
3–5
3–6
3–6

Table of Contents

Maintenance Terminal Connector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–6
Blower Fault LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–6
Configuring the EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–7
Connecting the EMU Communication Bus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–7
EMU Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–8
EMU Firmware Versions 1.0 and 1.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–9
EMU Firmware Version 1.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–9
Installing EMU Firmware Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–10
Setting the Temperature Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–10
Setting the Blower Speed Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–12
EMU Status Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–14
Alarm Control Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–15
EMU Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–16
Monitoring Power Supply Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–17
Error Condition Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Configuration Fault Code Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–21
Replacing an EMU. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–28

Chapter 4, Power Verification and Addressing
Product Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–2
Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–3
PVA Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–4
EMU Installation Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–4
Disk Drive SBB Device IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–4
Monitoring Power Supply Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–6
Setting the Enclosure Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–7
Master Enclosure Address. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–8
Expansion Enclosure Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–9
Implementing the Standby Power Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–10
Replacing a PVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–12

Chapter 5, Single-Ended I/O Module
I/O Module and Expansion Cable Error Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–3
Expanding the UltraSCSI Bus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–3
Controlling the Internal SCSI Bus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–4
Controlling the External SCSI Bus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–4
Controlling the Fault Bus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–5

UltraSCSI RAID Enclosure v

Table of Contents

Error Reporting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–5
Incompatible I/O Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–5
TERMPOWER Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–6
Rules for Expanding an UltraSCSI Bus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–7
Cabling an UltraSCSI Subsystem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–9
Shutting Down the Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–10
Cabling a Two-Enclosure Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–12
Turning On the Subsystem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–15
Cabling a Three-Enclosure Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–16
Replacing an I/O Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–21
Replacing an Expansion Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–23

Chapter 6, Power Distribution
Power System Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–2
Power Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–2
Standard (n+1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–3
Redundant (n+4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–3
Optimum (n+4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–3
AC Power Source. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–3
AC Input Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–4
Shelf Power Supplies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–5
AC Power Buses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–6
Power Distribution Configurations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–7
Standard (n+1) Power Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–7
Redundant (n+4) Power Supply Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–8
Optimum (n+4) Power Supply Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–9
Error Condition Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–10
Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–10
AC Input Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–11
Replacing a Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–12
Handling a Power Supply SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–12
Procedures for Replacing a Power Supply. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–13
Replacing an AC Input Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–17

Chapter 7, Storage Devices
Identifying a Storage Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–2
Disk Drive SBB Device Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–3
Disk Drive SBB Status Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–4

vi UltraSCSI RAID Enclosure

Table of Contents

Guidelines for Replacing a Disk Drive SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–6
The SBB Connector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–6
Prerequisites for Replacing a Disk Drive SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–7
Disk Drive SBB Replacement Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–7
Hot Swap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–8
Warm Swap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–8
Cold Swap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–8
Replacing a Disk Drive SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–9
Handling a Disk Drive SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–9
Procedures for Replacing a Disk Drive SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–10

Chapter 8, Enclosure Blowers
Function and Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–3
Error Condition Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–4
Replacing a Blower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–4

Appendix A, UltraSCSI Enclosure Specifications
Physical and Electrical Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–2
Operating Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–3
SBB Thermal Stabilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–5

Appendix B, Assembling an UltraSCSI RAID Subsystem
Installation Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–2
Installing the EMU. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–3
Installing the PVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–3
Installing the Disk Drive SBBs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–3
Installing the I/O Modules and Expansion Cables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–4
Installing Array Controllers and Cache Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–5
Installing the Power System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–9
Connecting Subsystem Cables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–10
Configuring the Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–10

Appendix C, Installing EMU Microcode
Microcode Upgrades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–1
Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–2
Preliminary Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–2
Preparing the EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–4

UltraSCSI RAID Enclosure vii

Table of Contents

Loading the Microcode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–4
Restarting the Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–6

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Glossary–1

viii UltraSCSI RAID Enclosure

List of Figures
Figure 1–1 DS–BA370 Series UltraSCSI Enclosure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–1
Figure 1–2 Typical DS–BA370 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–3
Figure 1–3 Typical Enclosure Bus and Device Addressing. . . . . . . . . . . . . . . . . . . . . . . . . 1–6
Figure 1–4 DS–BA370 Status LEDs Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–11
Figure 1–5 RA7000 Status LEDs Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–11
Figure 2–1 Power Supply Bezel Label. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–4
Figure 2–2 AC Power Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–6
Figure 2–3 DS–BA370 Status LEDs Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–14
Figure 2–4 RA7000 Status LEDs Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–14
Figure 2–5 Power Supply Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–16
Figure 2–6 Disk Drive Status LEDs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–17
Figure 2–7 Typical Array Controller OCP LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–18
Figure 2–8 EMU Status LED Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–19
Figure 2–9 PVA Status LED Location. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–19
Figure 2–10 I/O Module TERMPOWER LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–20
Figure 3–1 EMU. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–1
Figure 3–2 EMU Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–2
Figure 3–3 EMU Major Component Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–5
Figure 3–4 EMU Communications Bus Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–7
Figure 3–5 EMU Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–14
Figure 4–1 PVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–1
Figure 4–2 PVA Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–2
Figure 4–3 PVA Major Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–2
Figure 4–4 Master Enclosure Device IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–5
Figure 4–5 Master Enclosure Address 0: Shelf and SCSI Bus IDs . . . . . . . . . . . . . . . . . . . 4–8
Figure 4–6 Expansion Enclosure Address 2: Shelf and SCSI Bus IDs . . . . . . . . . . . . . . . . 4–9

UltraSCSI RAID Enclosure ix

List of Figures

Figure 4–7 Expansion Enclosure Address 3: Shelf and SCSI Bus IDs . . . . . . . . . . . . . . 4–10
Figure 5–1 Single-Ended I/O Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1
Figure 5–2 I/O Module Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–2
Figure 5–3 I/O Module Numbering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–2
Figure 5–4 I/O Module TERMPOWER LEDs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–6
Figure 6–1 Enclosure Power System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–1
Figure 6–2 AC Input Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–4
Figure 6–3 Shelf Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–5
Figure 6–4 Enclosure AC Power Buses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–6
Figure 6–5 Standard Power Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–7
Figure 6–6 Redundant Power Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–8
Figure 6–7 Power Supply Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–10
Figure 7–1 Typical Disk Drive SBB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–1
Figure 7–2 Typical Drive SBB Regulatory Label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–2
Figure 7–3 Typical Drive SBB Bezel Label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–3
Figure 7–4 Disk Drive Status LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–4
Figure 8–1 Enclosure Blower Assembly Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–2
Figure A–1 UltraSCSI RAID Enclosure Major Components. . . . . . . . . . . . . . . . . . . . . . . A–1
Figure A–2 UltraSCSI Enclosure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–2
Figure B–1 UltraSCSI RAID Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–1
Figure B–2 I/O Module Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–4
Figure B–3 Array Controller and Cache Module Locations. . . . . . . . . . . . . . . . . . . . . . . . B–5
Figure B–4 Enclosure Power System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–9

x UltraSCSI RAID Enclosure

List of Tables
Table 1–1 UltraSCSI Subsystems Capacities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–7
Table 2–1 Compaq-Supported PVA SCSI ID Switch Settings . . . . . . . . . . . . . . . . . . . . . . 2–5
Table 2–2 Turning On the Subsystem Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–7
Table 2–3 Turning Off the DC Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–9
Table 2–4 Restoring the DC Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–10
Table 2–5 Turning Off the Subsystem Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–11
Table 2–6 Enclosure Status Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–12
Table 2–7 Enclosure Status LED Displays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–15
Table 3–1 EMU Communication Cables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–8
Table 3–2 EMU Set Point Temperature Conversions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–11
Table 3–3 Temperature Set Point Rules—EMU Firmware Versions 1.0 and 1.1 . . . . . . 3–11
Table 3–4 Temperature Set Point Rules—EMU Firmware Versions 1.2 . . . . . . . . . . . . . 3–12
Table 3–5 Subsystem Status LEDs Displays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–16
Table 3–6 DC Voltage Ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–17
Table 3–7 EMU Status Displays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Table 3–8 EMU Fault Code Displays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–22
Table 3–9 Removing an EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–28
Table 3–10 Installing an EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–29
Table 4–1 DC Voltage Ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–6
Table 4–2 Setting the Enclosure Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–7
Table 4–3 Compaq-Supported PVA SCSI ID Switch Settings . . . . . . . . . . . . . . . . . . . . . . 4–8
Table 4–4 Turning Off the DC Power Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–11
Table 4–5 Turning On the DC Power Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–12
Table 4–6 Removing a PVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–12
Table 4–7 Installing a PVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–14
Table 5–1 TERMPOWER Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–7

UltraSCSI RAID Enclosure xi

List of Tables

Table 5–2 UltraSCSI Bus Expansion Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–8
Table 5–3 Preliminary Cabling Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–9
Table 5–4 UltraSCSI Bus Cables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–10
Table 5–5 Shutting Down the UltraSCSI Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–11
Table 5–6 Installing Two-Enclosure Subsystem Cables . . . . . . . . . . . . . . . . . . . . . . . . . 5–12
Table 5–7 Turning On the UltraSCSI Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–15
Table 5–8 Installing Three-Enclosure Subsystem Cables . . . . . . . . . . . . . . . . . . . . . . . . 5–17
Table 5–9 Removing an I/O Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–21
Table 5–10 Installing an I/O Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–22
Table 5–11 Removing an Expansion Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–23
Table 5–12 Installing an Expansion Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–24
Table 6–1 Power Supply Status LED Displays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–11
Table 6–2 SBB Handling Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–12
Table 6–3 Removing a Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–13
Table 6–4 Installing a Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–15
Table 6–5 Removing an AC Input Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–17
Table 6–6 Installing an AC Input Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–18
Table 7–1 Disk Drive Status LED Displays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–5
Table 7–2 SBB Handling Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–9
Table 7–3 Removing a Disk Drive SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–10
Table 7–4 Installing a Disk Drive SBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–12
Table 8–1 Removing a Blower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–5
Table 8–2 Installing a Blower. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–6
Table A–1 Optimum Operating Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–3
Table A–2 Minimum Operating Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–3
Table A–3 Shipping Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–4
Table A–4 SBB Thermal Stabilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–5
Table B–1 Installing Array Controllers and Cache Modules . . . . . . . . . . . . . . . . . . . . . . B–6
Table C–1 Preliminary Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–2
Table C–2 Preparing the EMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–4
Table C–3 Loading the Microcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–4
Table C–4 Typical PC Monitor Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–5
Table C–5 Restarting the Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C–6

xii UltraSCSI RAID Enclosure

Preface
The Compaq StorageWorks UltraSCSI RAID Enclosure (DS–BA370
Series) User’s Guide describes the functions, operations, components,
configurations, assembly, and specifications of this enclosure.

Intended Audience
This document is for use by personnel responsible for designing,
configuring, assembling, installing, and operating UltraSCSI RAID
subsystems using a DS–BA370 series enclosure mounted in either a
data center cabinet or a departmental server cabinet.

Structure
Manual organization is as follows:
Chapter 1
This chapter describes the UltraSCSI RAID enclosure functions, uses,
components, and features.
Chapter 2
This chapter describes how to configure and operate an UltraSCSI
RAID enclosure.
Chapter 3
This chapter describes the environmental monitoring unit (EMU)
operation, functions, uses, configuration, and replacement
procedures.

UltraSCSI RAID Enclosure xiii

Preface

Chapter 4
This chapter describes the power verification and addressing (PVA)
assembly operation, functions, uses, configuration, and replacement
procedures.
Chapter 5
This chapter describes the single-ended input/output (I/O) module
operation, functions, uses, configuration, and replacement
procedures.
Chapter 6
This chapter describes the enclosure power distribution components
operation, functions, uses, configuration, and replacement
procedures.
Chapter 7
This chapter describes the disk drive operation, functions, uses,
configuration, and replacement procedures.
Chapter 8
This chapter describes blower operation, status reporting, and
replacement procedures.
Appendix A
This appendix describes enclosure specifications, which include
physical, electrical, and environmental specifications.
Appendix B
This appendix describes the procedures for installing components in
an enclosure to create an UltraSCSI RAID subsystem.
Appendix C
This appendix describes general procedures for installing EMU
microcode into an UltraSCSI RAID subsystem.
Glossary

xiv UltraSCSI RAID Enclosure

Preface

Documentation Conventions
Table 1 describes the conventions for this publication:
Table 1

Documentation Conventions

boldface type

Boldface type indicates the first instance of terms
being defined in the text, the glossary, or both.
Hypertext links connect each glossary term within
a chapter to its definition in the glossary. Clicking
on a glossary term within a chapter activates the
link.

italic type

Italic type emphasizes important imformatioin and
indicates complete titles of manuals. Italic type
within the glossary indicates a cross-reference.

courier type
A/R

Courier type indicates a keyboard command or
screen display data.
As required.
The light emitting diode (LED) is OFF.

The LED is FLASHING (blinking in half-second
cycles).
The LED is ON or blinking rapidly (blinking less
than half-second cycles).

Warning
Information essential to the safety of personnel.

Caution
Information essential to avoid damaging software
or hardware.

UltraSCSI RAID Enclosure xv

Preface

Table 1

Documentation Conventions (Continued)

Note

Notes contain information that might be of special
interest to the user.
Single-ended SCSI bus.

Differential SCSI bus.

Related Documents
Table 2 lists publications that contain additional information about the
UltraSCSI RAID subsystem:
Table 2 Related Publications
Publication Title

Order Number

HSZ70 Array Controller HSOF Version 7.0 CLI
Reference Manual

EK–CLI70–RM

HSZ70 Array Controller HSOF Version 7.0
Configuration Manual

EK–HSZ70–CG

HSZ70 Array Controller HSOF Version 7.0
Service Manual

EK–HSZ70–SV

HSx801 Array Controller ACS Version 8.n2
User’s Guide

EK–HSx801–UG.x013

HSx801 Array Controller ACS V8.x4
Configuration and CLI Reference Guide

EK–HSx801–RG

HSx801 Array Controller ACS V8.x4
Maintenance and Service Guide

EK–HSx801–SV

HSx801 Array Controller ACS V8.x4
Illustrated Parts Map

EK–HSx801–MP

xvi UltraSCSI RAID Enclosure

Preface

Table 2 Related Publications (Continued)
Publication Title

Order Number
See system-specific
“Getting Started”
manual

Operating System Specific Release Notes

Improving UltraSCSI RAID Subsystem
Operation

EK–BA370–UP

Installing an External Cache Battery Cable

EK–HSZ70–TE

Installing a Host Bus Cable Ferrite Bead

EK–SWXES–IG

StorageWorks Solutions SBB User’s Guide

EK–SBB35–UG

UltraSCSI Subsystem Standby Power
Operation

EK–POWER–IG

UltraSCSI RAID Data Center Cabinet
(DS–SW600–Series) Installation and User’s
Guide

EK–SW600–UG

1. HSx80 array controllers include: HSZ80 and HSG80
2. Version 8.0, 8.1, or 8.2
3. Revision A01, B01, or higher
4. Version 8.3 or higher

UltraSCSI RAID Enclosure xvii

Manufacturer’s Declarations
This section discusses electromagnetic compatibility and accoustic
noise declarations.

Electromagnetic Compatibility
This CE–Mark Class A certified product can be installed in a
commercial or an office environment.
WARNING!
This is a Class A product. In a domestic environment
this product may cause radio interference in which case
the user may be required to take adequate measures.
ACHTUNG!
Dieses ist ein Gerät der Funkstörgrenzwertklasse A. In
Wohnbereichen können bei Betrieb dieses Gerätes
Rundfunkstörungen auftreten, in welchen Fällen der
Benutzer für entsprechende Gegenmaßnahmen
verantwortlich ist.
ATTENTION!
Ceci est un produit de Classe A. Dans un environnement
domestique, ce produit risque de créer des interférences
radioélectriques, il appartiendra alors à l’utilisateur de
prendre les mesures spécifiques appropriées.

UltraSCSI RAID Enclosure xix

Manufacturer’s Declarations

Acoustic Noise Declarations
Compaq® declares that the acoustic values of this product are as
shown in Table 1 and Table 2.
Table 1

Acoustics – Declared Values per ISO 9296 and ISO 7779

Product

Sound Power Level
LWAd, B

Sound Pressure
Level LpAm dBA
(bystander positions)

Idle

Operate

Idle

Operate

RAID Array 7000 Subsystem
(DS–SWXRA–HA) with:
2 — ac input boxes (DS–BA35X–HE)
8 — 180 W shelf power supplies
(DS–BA35X–HH)
24 — disk drives (2, 4, 9, or 18 GB)

6.1

N/A

48.4

RAID Array 7000 Subsystem
(DS–SWXRA–HA) with:
2 — ac input boxes (DS–BA35X–HE)
8 — 180 W shelf power supplies
(DS–BA35X–HH)
24 — disk drives (2, 4, 9, or 18 GB)

6.1

6.3

48.4

Note
Current values for sound power levels are available from Compaq representatives.
1 Bel = 10 dBA.

xx UltraSCSI RAID Enclosure

Manufacturer’s Declarations

Table 2

Schallemissionswerte – Werteangaben nach ISO 9296 und ISO 7779/DIN
EN27779
Schalleistungspege
l
LWAd, B

Schalldruckpegel
LpAm, dBA
(Zuschauerpositionen)

Leerlauf

Betrieb

Leerlauf

Betrieb

RAID Array 7000 Subsystem
(DS–SWXRA–HA) with:
2 — ac input boxes (DS–BA35X–HE)
8 — 180 W shelf power supplies
(DS–BA35X–HH)
24 — disk drives (2, 4, 9, or 18 GB)

6,1

N/A

48,4

RAID Array 7000 Subsystem
(DS–SWXRA–HA) with:
2 — ac input boxes (DS–BA35X–HE)
8 — 180 W shelf power supplies
(DS–BA35X–HH)
24 — disk drives (2, 4, 9, or 18 GB)

6,1

6,3

48,4

Gerät

Note
Aktuelle Werter für spezielle Austüstungsstufen sind über die Compaq Computer
Vertretungen erhältlich. 1 Bel = 10 dBA.

UltraSCSI RAID Enclosure xxi

1
Introducing the UltraSCSI
RAID Enclosure
The Compaq StorageWorks™ UltraSCSI RAID (redundant array
of independent disks) enclosure (DS–BA370 series) shown in Figure
1–1 is the basic building block for Compaq single-ended UltraSCSI
RAID subsystems.
Figure 1–1 DS–BA370 Series UltraSCSI Enclosure

CXO5797A

UltraSCSI RAID Enclosure 1–1

Introducing the UltraSCSI RAID Enclosure

This enclosure cannot function until:
1. It is installed in an UltraSCSI data center cabinet (DS–SW600
series) or an UltraSCSI departmental server cabinet
(DS–SWXRE series).
2. 3.5-inch StorageWorks building block (SBB) disk drives are
installed.
3. A compatible array controller is installed and connected to a
host computer or adapter.
4. The enclosure and the subsystem are configured for UltraSCSI
operation.
Note
The “DS–” model number prefix defines new Compaq
storage products. In some cases, these numbers might be
similar to an existing product number. For example, the
DS–BA35X–HE ac input box (also known as the ac
power controller) and the BA35X–HE ac input box have
similar model numbers but are not interchangeable.
Installing incompatible products in a StorageWorks
subsystem could result in degraded operation or
noncompliance with country-specific certifications.
See Figure 1–2 for the Compaq-supported enclosure applications.

1–2 UltraSCSI RAID Enclosure

Introducing the UltraSCSI RAID Enclosure

Figure 1–2 Typical DS–BA370 Applications

1
0

0
0

1
0

0
0

1
0

1
0
0

1
0

CXO6264A

UltraSCSI RAID Enclosure 1–3

Introducing the UltraSCSI RAID Enclosure

An UltraSCSI Bus Overview
UltraSCSI was developed by Digital Equipment Corporation (now
Compaq Computer Corporation) and subsequently standardized by
the ANSI (American National Standards Institute) SCSI (small
computer system interface) committee as standard X3T10.
UltraSCSI is a technology that incorporates several improvements
over a Fast 10 SCSI bus.
UltraSCSI increases the maximum transfer rate on a wide (16-bit)
SCSI bus from 10 megabits per second (Mb/s) to 20 Mb/s. This
change increases the maximum bus bandwidth from 20 megabytes
per second (MB/s) to 40 MB/s for a wide (16-bit) SCSI bus.
UltraSCSI incorporates smaller very high density cable
interconnect (VHDCI) cables and connectors. The connector is up
to half the size of the Fast SCSI HD68 (high-density 68-pin)
connector and the cables are thinner.
The higher speed of the single-ended UltraSCSI signals can cause
degradation of signal integrity over the same distances as
single-ended Fast SCSI signals. However, the UltraSCSI bus can run
over much greater distances than Fast SCSI in either single-ended or
differential mode. To attain these greater distances, UltraSCSI defines
and StorageWorks UltraSCSI implements the bus segment concept.
An UltraSCSI bus segment is an unbroken electrical path consisting
of conductors (in cables or on backplanes) and connectors. Every
UltraSCSI bus segment must have two terminators, one at each end
of the bus segment. Bus segments are either single-ended or
differential, depending on the type of terminators used in that
segment. In other words, an UltraSCSI bus segment corresponds to
what Fast SCSI referred to as an entire SCSI bus. As in Fast SCSI, a
bus segment only supports devices of the same type as its terminators.
Do not, for example, attempt to use a differential SCSI bus device on
a single-ended SCSI bus segment.
Note
The Compaq UltraSCSI RAID enclosure only supports
single-ended UltraSCSI buses.

1–4 UltraSCSI RAID Enclosure

Introducing the UltraSCSI RAID Enclosure

Product Description
The UltraSCSI enclosure is a rack-mountable storage subsystem that
accommodates the following components:
•

Depending upon the configuration, a minimum of zero, or a
maximum of two, UltraSCSI controllers with cache modules1,
such as the DS–HSZ70 series.

•

An environmental monitoring unit (DS–BA35X–EB), the
EMU, that monitors enclosure operation, detects and reports
error conditions, and can automatically initiate corrective actions.

•

A power verification and addressing assembly
(DS–BA35X–EC), the PVA, that establishes enclosure and
storage device addressing, monitors enclosure operation, and
detects and reports error conditions.

•

A minimum of one, or a maximum of two, ac input boxes
(DS–BA35X–HE), also called ac power controllers, that
distribute ac power to the shelf power supplies.

•

Either five or eight shelf power supplies (DS–BA35X–HH) that
provide dc power for enclosure operations.

•

Eight dual-speed blowers (DS–BA35X–ML) that cool the
components by drawing air in through the front and exhausting it
out the rear.

•

Six single-ended input/output (I/O) modules
(DS–BA35X–MN) for connecting enclosures together.

•

A maximum of 24 single-ended, 3.5-inch SBB disk drives for
data storage.

For enclosures with fewer than 24 disk drive SBBs or 8 power
supplies, installing the optional blank bezels (DS–BA35X–PB, this
model number provides six blank bezels) will improve air flow.

Each cache module requires an external cache battery (ECB),
DS–HS35X–Bx series. The batteries should be installed either in an
ECB shelf or a departmental server cabinet top cover.

UltraSCSI RAID Enclosure 1–5

Introducing the UltraSCSI RAID Enclosure

Expansion enclosures have filler panels installed in the controller
and cache module slots that improve air flow and reduce
electromagnetic interference (EMI) radiation.
The blank panels in the controller B and cache module B locations
improve air flow and control EMI in enclosures with only one
controller.
The enclosure backplane eliminates the need for cables to connect the
six vertical buses to the four horizontal drive SBB shelves. All the
devices on a shelf use the same SCSI bus identifier (ID). The PVA
address switch defines the SCSI bus ID of each shelf. See Figure 1–3
for bus and shelf identifiers when viewing the enclosure from the
front.
Figure 1–3 Typical Enclosure Bus and Device Addressing
Figure CALLOUT

1
1

SCSI bus

Shelf 4

Shelf 3

Shelf 2

Shelf 1

CXO6266B

The ac input boxes distribute ac power to the individual shelf power
supplies through ac power cables. The backplane distributes the dc
output voltages (+5 V dc and +12 V dc) from the power supplies to
the following components:

1–6 UltraSCSI RAID Enclosure

Introducing the UltraSCSI RAID Enclosure

•

EMU

•

PVA

•

UltraSCSI bus array controllers and cache modules

•

Disk drives

•

I/O modules

UltraSCSI RAID Subsystem
A subsystem is one or more departmental server cabinets or data
center cabinets with an enclosure. Each subsystem requires a master
enclosure with an UltraSCSI controller installed. Installing one or
two expansion enclosures and connecting them with cables creates an
“expanded” subsystem. See Table 1–1 for the capacities of the various
subsystems.
Table 1–1

UltraSCSI Subsystems Capacities

Enclosures

SCSI Buses

Disk Drives

Total Disk Drives

Note
Not all array controllers support configurations with
expansion cabinets. Refer to the array controller
documentation to ensure the controller supports expanded
configurations.

EMU
The EMU monitors the status of the UltraSCSI RAID enclosure to
include power, ambient (intake) air temperature, blower status, and
so forth. The EMU also detects error and fault conditions, displays
these conditions, reports the conditions to the user and the array
controller, and, in some cases, implements corrective actions.

UltraSCSI RAID Enclosure 1–7

Introducing the UltraSCSI RAID Enclosure

PVA
Each PVA functions only with its host enclosure and EMU. There is
no master PVA in the UltraSCSI RAID subsystem.
The PVA has the following primary functions:
•

Allows the user to select the enclosure UltraSCSI bus ID.

•

Enables the user to place the subsystem in a standby power mode
condition or return it to an operational status.

•

In conjunction with the associated EMU, ensures that the major
components are functioning properly and notify the user and the
array controller of error or fault conditions.

Master Enclosure
Each UltraSCSI RAID subsystem (whether it has one, two, or three
enclosures) is under the control of the master enclosure. The
enclosure with the array controllers is the master enclosure. The EMU
in the master enclosure is the only EMU that can communicate
directly with the array controllers. In a multiple enclosure, or
expansion configuration, the master enclosure also communicates
with the other enclosures using the EMU-EMU communications bus.
The master enclosure PVA SCSI bus address setting is always 0.
Using address 0 automatically assigns SCSI bus device addresses 0,
1, 2, and 3 to the four drive SBB shelves.

Master EMU
The EMU in the master enclosure is the master EMU. Only the
master EMU communicates directly with the array controller. The
master EMU communicates with the expansion enclosure EMUs over
the EMU–EMU communications bus. The master EMU controls the
operation of the entire subsystem to include error detection, reporting,
and status displays.

Expansion Enclosure
An UltraSCSI RAID enclosure connected to the master enclosure
with cables is an expansion enclosure. Depending upon the array
controller, a subsystem can have a maximum of two expansion
enclosures. The expansion PVA SCSI bus address setting is either:

1–8 UltraSCSI RAID Enclosure

Introducing the UltraSCSI RAID Enclosure

2 — Establishing SCSI bus device addresses 8, 9, 10, and 11
3 — Establishing SCSI bus device addresses 12, 13, 14, and 15
Note
Not all array controllers support an UltraSCSI RAID
subsystem with more than one enclosure. Refer to the array
controller documentation to ensure that the controller
supports expanded configurations.

Expansion EMU
An EMU installed in an expansion enclosure is an expansion EMU.
This EMU communicates with the master EMU over the EMU-EMU
communications bus. Upon loss of communications with the master
EMU, the expansion EMU assumes control of certain expansion
enclosure functions. In this state, the expansion EMU cannot
communicate with the UltraSCSI controller.

I/O Module
The I/O modules provide termination for the internal UltraSCSI buses
and are the connection points for the cables connecting the six buses
to the expansion enclosures.

SCSI Bus Controller
A hardware-software device that manages SCSI bus communications.
Controllers typically differ by the type of interface to the host and
provide functions beyond those the devices support. A host port
adapter connects the host to the controller. Other terms commonly
used for the SCSI bus controller include array controller, host
controller, or controller.

UltraSCSI RAID Enclosure 1–9

Introducing the UltraSCSI RAID Enclosure

Note
DS–HSx70 series array controllers require DS–HSx70
series cache modules to operate properly; likewise,
DS–HSx80 series array controllers require DS–HSx80
series cache modules.

Cache Module
A storage buffer that is required to support bus controller1 operations.

ECB
The ECB provides backup power to the cache module to protect data
integrity if there is a power failure.

Enclosure Status
When any of the multiple error detection circuits determine there is a
problem, an audible alarm can sound and one or more sets of status
LEDs (light emitting diodes) display an error code.
The enclosure has two LED indicators (see Figure 1–4). The green
LED is the System OK LED; the amber LED is the Fault LED. When
the Fault LED is ON, check the other status LEDs to determine the
cause of the error.
•

For an enclosure in a pedestal, these LEDs are part of the
pedestal (see Figure 1–5).

•

For an enclosure in a DS–SW600 series cabinet, the LEDs on the
cabinet door also display the enclosure status.

1–10 UltraSCSI RAID Enclosure

The cache module located in the cache A slot only supports the
array controller located in the controller A slot; likewise for cache
module B and controller B.

Introducing the UltraSCSI RAID Enclosure

Figure 1–4 DS–BA370 Status
LEDs Locations

Figure 1–5 RA7000 Status LEDs Locations

System OK
(green)
Fault
(amber)

System OK
(Green)

Fault
(Amber)

1
0

CXO5451B

CXO5751A

CALLOUTS For Both Figures
1.

System OK (green)

Fault (amber)

Options
The following sections describe the UltraSCSI RAID subsystem user
selectable options.

Power
The standard UltraSCSI enclosure power configuration consists of
one ac input box and five shelf power supplies. This configuration
provides the minimum power supply redundancy. A system power
failure will occur for any of the following conditions:
•

Loss of the ac power source

•

Failure of the ac input box

•

Failure of two power supplies

The user has the option of modifying the standard power
configuration to eliminate single points of failure.

UltraSCSI RAID Enclosure 1–11

Introducing the UltraSCSI RAID Enclosure

Redundant Power Supply Configuration
Adding three power supplies and a second ac input box provides
significantly more power redundancy than the standard configuration.
The redundant power supply configuration uses a single ac power
source. A system power failure will occur for any of the following
conditions:
•

Loss of the ac power source

•

Failure of both ac input boxes

•

Failure of five power supplies

This configuration provides additional power supply redundancy.
Optimum Power Configuration
Adding three shelf power supplies, a second ac input box, and a
second ac power source provides optimum power redundancy. With
separate ac power sources for each ac input box, there are no single
points of failure.
Note
For optimum operation, each ac input box should use a
different ac power source.
A system power failure will occur for any of the following conditions:
•

Loss of both ac power sources

•

Failure of both ac input boxes

•

Failure of five power supplies

•

Failure of the ac power source for bus B and one power supply on
bus A

•

Failure of the ac power source on bus A and one power supply on
bus B

•

Failure of the ac input box on bus B and one power supply on
bus A

•

Failure of the ac input box on bus A and one power supply on
bus B

1–12 UltraSCSI RAID Enclosure

Introducing the UltraSCSI RAID Enclosure

Compaq recommends using this configuration to eliminate all single
points of power failure.

Dual Controllers
Establishing a subsystem with a dual redundant configuration (two
UltraSCSI controllers, two cache modules, and ECBs) is the most
effective and efficient method of ensuring continuous subsystem
operation in the event of a component failure.

UltraSCSI Bus Expansion
Some controllers support expanding the UltraSCSI bus from the
master enclosure to two expansion enclosures. Refer to the array
controller documentation to determine if the controller supports this
feature.

UltraSCSI RAID Enclosure 1–13

2
Operating the UltraSCSI
Enclosure
Setting up an UltraSCSI RAID subsystem for operation involves
implementing the procedures described in the system-specific
“Getting Started” manual provided with each system. This chapter
supplements and expands on this information.
Other chapters in this manual contain detailed information about
individual component functions, operations, error conditions, and so
forth.

UltraSCSI Configuration Rules
Proper operation of the subsystem requires complying with both the
UltraSCSI and hardware configuration rules. In some instances,
operating system-specific configuration rules might apply.

General
The following rules apply to the UltraSCSI bus:
•

The UltraSCSI RAID subsystems can transfer data at a higher
rate (40 MB/s versus 20 MB/s) than that of the FAST 10
subsystems.

•

UltraSCSI subsystems require 68-pin VHDCI cables and
connectors. The physical difference in the contact spacing
precludes using FAST 10 cables.

UltraSCSI RAID Enclosure 2–1

Operating the UltraSCSI Enclosure

•

The following devices use a SCSI bus address (ID) and are SCSI
bus nodes:
— Adapters
— Controllers
— Storage devices

•

Every node on a bus must have a unique SCSI bus ID.

•

The SCSI bus ID for controller A is always 7.

•

The SCSI bus ID for controller B is always 6.

•

There are slots for six 3.5-inch disk drive SBB nodes in a shelf.
The address switch on the PVA module determines node SCSI
bus IDs. There are a maximum of 12 valid DS–BA370 SCSI bus
IDs—0 through 3 and 8 through 15.

•

A DS–BA370 series UltraSCSI bus does not support 5.25-inch
devices.

•

A DS–BA370 series enclosure does not support narrow devices.

•

Only UltraSCSI devices have a maximum transfer rate of 40
MB/s.

•

Installing a FAST 10 storage device reduces the bus average
transfer rate as follows:
— When the array controller communicates with a FAST 10
device, the transfer rate is at FAST 10 (the maximum rate this
device can handle).
— When the array controller communicates with a FAST 20
device, the transfer rate is at FAST 20 (the maximum rate this
device can handle).

UltraSCSI RAID Controllers
The following general configuration rules apply to the UltraSCSI
RAID array controllers:
•

The controller A SCSI bus ID is always 7.

•

The controller B SCSI bus ID is always 6.

2–2 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

•

Compaq UltraSCSI array controllers support FAST 10 (20 MB/s)
16-bit (wide) disk drives.

•

Compaq UltraSCSI array controllers support FAST 20 (40 MB/s)
16-bit (wide) disk drives.
Caution
If a conflict in controller configuration information exists
between this document and the controller-specific
documentation, use information contained in the
controller-specific documentation.

Note
Disk drive support is controller-specific. Refer to the array
controller documentation to determine compatible disk
drives.

UltraSCSI Nodes
A node is a SCSI device that uses a SCSI bus ID. In the UltraSCSI
RAID enclosure, the following rules apply:
•

SCSI bus ID 7 is reserved for the controller only.

•

SCSI bus ID 6 is reserved for the controller only.

•

SCSI bus IDs 4 or 5 cannot be used.

•

5.25-inch devices are not supported.

•

The valid device SCSI bus IDs are device type dependent,
meaning:
— An UltraSCSI device can use device addresses 0 through 3 or
8 through 15.
— A wide (16-bit), FAST 10, SCSI device (–VW suffix) uses
device addresses 0 through 3 or 8 through 15.

UltraSCSI RAID Enclosure 2–3

Operating the UltraSCSI Enclosure

UltraSCSI Enclosures
The following configuration rules apply to UltraSCSI RAID
subsystem enclosures:
•

The only compatible ac input box is model DS–BA35X–HE.

•

The only compatible power supply is model DS–BA35X–HH
(180 W). The bezel label shown in Figure 2–1 identifies these
power supplies.

Figure 2–1 Power Supply Bezel Label

CXO5926A

Enclosure Addressing
The following configuration rules apply to the UltraSCSI RAID
enclosure address settings:
•

The enclosure containing the array controller is the master
enclosure. The PVA SCSI ID switch setting for the master
enclosure is always 0.

•

UltraSCSI subsystem expansion enclosures use either PVA SCSI
ID switch setting 2 or 3:
— The first expansion enclosure switch setting is always 2.
— The second expansion enclosure switch setting is always 3.

2–4 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

See Table 2–1 for the Compaq-supported PVA SCSI ID switch
settings.
Table 2–1

Compaq-Supported PVA SCSI ID Switch
Settings

Total Number
of Enclosures

Master
Enclosure

First
Expansion

Second
Expansion

None

Starting the Subsystem
After assembling an enclosure and connecting enclosure cabling,
power is applied to start the subsystem.
Note
Before applying power to the subsystem, configure the
entire subsystem (including the host and the array
controller) as described in the appropriate publications.
Applying power to all the enclosures causes the EMU, PVA, and
array controller to automatically begin system operations.

UltraSCSI RAID Enclosure 2–5

Operating the UltraSCSI Enclosure

Applying Power
Initial power application is a matter of turning on the ac input boxes
located in the lower left and right corners of the enclosure (see Figure
2–2, labels A and B). Pressing the I on the ac power switch applies ac
power only to the shelf power supplies on the associated power bus
(bus A or B).
Note
Apply power in the following sequence:
1. All expansion enclosures
2. Master enclosure

Figure 2–2 AC Power Distribution
AC Input Box and SBB Shelf
Power Supply Locations

AC Input Box
1

1
0

5A
4A 19 20 21 22 23 24 4B

Shelf 4

3A 13 14 15 16 17 18 3B

Shelf 3

CXO5113A

10 11 12 2B

Shelf 2

Shelf 1

CXO5808B

2–6 UltraSCSI RAID Enclosure

Figure CALLOUTS
1.

Power bus A

Power bus B

Operating the UltraSCSI Enclosure

Turning On the Subsystem
After completing subsystem assembly, expansion, or movement of
the subsystem, complete the procedure in Table 2–2 to turn it back
O N.
Table 2–2
1.

Turning On the Subsystem Power

If there are no expansion enclosures, go to step 2.

On the master enclosure,
turn O N all the ac input
boxes by pressing I on each
ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn ON all the ac
input boxes for each
expansion enclosure by
pressing I on each ac power
switch (CALLOUT 1).

CXO5488B

Verify the array controller places the subsystem into an operational
state.

UltraSCSI RAID Enclosure 2–7

Operating the UltraSCSI Enclosure

Shutting Down the Subsystem
Shutting down the subsystem involves:
•

Clearing the cache module

•

Shutting down the controller

•

Either turning OFF the dc power for the subsystem (entering the
standby power mode) or going one step further and turning OFF
the ac input boxes (full power shutdown)
Caution
Failure to clear the cache module and shut down the array
controller before removing power will disrupt data
transfers. Therefore, always shut down the array controller
as described in the array controller documentation.
Some array controllers bypass the standby power mode
functionality from the PVA standby power switch. Refer to
the array controller documentation for availability of this
feature.

Standby Power Mode
Placing the subsystem in the standby power mode disconnects the dc
power from the backplane. However, ac input power is still applied to
the shelf power supplies. In this mode, the EMU alarm control switch
now functions as the standby power switch. To restore the dc power to
the backplane, press the alarm control switch on each enclosure
EMU.

Full Power Shutdown
Powering down an enclosure always involves shutting down the array
controller and then pressing the PVA standby power switch on each
expansion enclosure. If there are no expansion enclosures, shut down
the array controller and then press the PVA standby power switch on
the master enclosure.
Turning the subsystem off involves shutting down the array controller,
turning OFF the dc power, and then turning OFF the ac input boxes in
each enclosure.

2–8 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

Using Standby Power Mode
Complete the procedure in Table 2–3 to turn OFF the dc power
distribution, placing the subsystem in standby power mode. Complete
the procedure in Table 2–4 to restore dc power within the enclosures.
Table 2–3

Turning Off the DC Power

Clear the cache module and shut down the array controller per the
array controller documentation.

If there are no expansion enclosures, go to step 3.
If there are expansion enclosures, press and hold the PVA standby
power switch (CALLOUT 1) on an expansion enclosure until the
master EMU initiates a power shut down on all enclosures.
1

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1) on the
master enclosure until the EMU initiates a power shut down on the
master enclosure.

Note
The EMU alarm control switch (CALLOUT 1 shown below) on
each enclosure now functions as the dc power switch.
1

3
4

CXO6302A

UltraSCSI RAID Enclosure 2–9

Operating the UltraSCSI Enclosure

Table 2–4

Restoring the DC Power

If there are no expansion enclosures, go to step 3.

If there are expansion enclosures, momentarily press the EMU alarm
control switch (CALLOUT 1) on each expansion enclosure.
1

3
4

CXO6302A

Momentarily press the EMU alarm control switch (CALLOUT 1) on
the master enclosure.

Verify that the array controller restores the subsystem to an
operational state.

Turning Off the Subsystem
Adding enclosures or moving the subsystem might require removing
all power from the subsystem. Complete the procedure in Table 2–5
to remove all power.

2–10 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

Table 2–5

Turning Off the Subsystem Power

Clear the cache module and shut down the array controller per the
array controller documentation.

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1) on the
master enclosure until the master EMU initiates a power shut down
on the master enclosure.

If there are no expansion enclosures, go to step 5.

Turn O FF all the ac input
boxes for the master
enclosure by pressing O on
each ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn OFF all the
ac input boxes for each
expansion enclosure by
pressing O on each ac power
switch (CALLOUT 1).

CXO5488B

UltraSCSI RAID Enclosure 2–11

Operating the UltraSCSI Enclosure

Fault Detection
Fault detection involves identifying errors and providing an indication
to the user for analysis. The following sections address how faults are
identified, descriptions of status LED indications, and in some cases,
possible corrective actions.

Error Detection, Reporting, and Analysis
The EMU and PVA constantly monitor the operational status of the
enclosure and its components. In some instances, automatic
corrective action implementation occurs. Status changes are indicated
to the user through a combination of LEDs, an EMU audible alarm,
and an optional maintenance terminal connected to the array
controller. Table 2–6 lists the major enclosure components and the
status items monitored by each component.
Table 2–6

Enclosure Status Monitoring

Component
Enclosure

Power

Storage devices
(disk drive
SBBs)

Status Monitored
•

Intake air temperature

•

Exhaust air temperature

•

AC input power

•

Number of operational 180 W power supplies

•

Enclosure SCSI bus address

•

Output of dc power supplies

•

DC bus voltages

•

Termination I/O module dc voltage

•

Control signal response

•

Controller-detected fault condition

•

Removal

•

Installation

EMU

Installation

PVA

Installation

2–12 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

Table 2–6
Blowers

Enclosure Status Monitoring (Continued)
•

Operation

•

Operating speed

•

Installation

For a detailed description of status LEDs, and specific error
conditions, see the applicable chapter for each component.

Enclosure Status Indicators
The primary elements of the enclosure status indicators are the
enclosure power supplies and blowers. Loss of either a dc voltage or a
blower creates an error condition. The enclosure notifies the user of
the problem by sounding an audible alarm and changing the status
LED displays. When the enclosure is mounted in a data center cabinet
or a departmental server cabinet, these two status LEDs might be
connected to LEDs on the front door. The location of the enclosure
status LEDs is shown in Figure 2–3 and Figure 2–4.

UltraSCSI RAID Enclosure 2–13

Operating the UltraSCSI Enclosure

Figure 2–3 DS–BA370 Status
LEDs Locations

Figure 2–4 RA7000 Status LEDs Locations

1
2

1
0

CXO5451C

CXO5751B

C ALLOUTS For Figures
1.

System OK LED

System fault LED

CALLOUT 1 is the system OK LED and is green. Callout 2 is the
system fault LED and is amber. One or both of these two LEDs can be
ON at any time.
When the green system OK LED is ON, it indicates that the enclosure
is operational. This LED can be ON as long as a minimum number of
redundant components (such as blowers or power supplies) are
functioning correctly.
If the amber system fault LED is ON, it indicates one or more
components have an error or fault condition. To isolate an error
condition, observe the status LEDs on the EMU, PVA, power
supplies, disk drive SBBs, and array controllers. These component
status LEDs are discussed under Controller Shelf Status Indicators
later in this chapter. Table 2–7 lists possible enclosure status LED
displays.

2–14 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

Table 2–7

Enclosure Status LED Displays

Enclosure Type
Data
Center
Cabinet

LED is ON

Description and
Possible Corrective Actions

Dept.
Server
Cabinet

The enclosure is operating normally. There
are no detected errors.

There are one or more failed components:

LED is OFF

Observe the EMU front panel to see if
there is an overtemperature condition,
a blower problem, or a power problem.
Replace the defective blower or power
components.

Observe individual disk drive SBBs for
an error condition.

Check the array controller optional
system maintenance terminal for error
messages.

Either there is no power applied to the
enclosure or the enclosure is in a reset state:
1.

Ensure that ac power is present at the ac
input boxes and that the array controllers
are operational.

Ensure that there are at least four
operational power supplies.

Check the array controller optional
system maintenance terminal for error
messages.

UltraSCSI RAID Enclosure 2–15

Operating the UltraSCSI Enclosure

SBB Shelf Status Indicators
SBB shelf status indicators are provided by shelf power supplies and
disk drive SBBs.
Power Supply Status LEDs
The two green LEDs on each power supply (see Figure 2–5) display
the blower, the power bus, and the individual power supply status.
Normally, both these LEDs are ON.
Figure 2–5 Power Supply Status LEDs
Figure CALLOUTS
1.

Shelf status LED

Power supply status LED

1
2

CXO4787B

The green shelf status LED (CALLOUT 1) is ON when all the power
supplies and all the blowers are operational. If any blower or power
supply fails, this LED turns OFF.
The green power supply status LED (CALLOUT 2) is O N when the
power supply is operating properly. If the power supply fails, both its
power supply status LEDs turn OFF.
Disk Drive SBB Configuration
Part of the initial configuration procedure for the subsystem involves
configuring the disk drives. The disk drive SBB configuration tells the
array controllers the drive SBB device type, plus its physical and
logical location. The array controller requires this initial
configuration information before transferring data blocks to specific
disks to create storage sets. Once the configuration process is
complete, the system cannot function properly unless all the disks
remain in the same physical and logical location.

2–16 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

Removal of a disk drive SBB device is automatically sensed by the
array controller.
When installing a replacement device in a configured subsystem, the
array controller ensures that the replacement device type is identical
to the one removed. Only after establishing complete compatibility
can the array controller start configuring the disk drive SBB. As part
of the configuration process, the array controller will restore all data
that was on the original disk.
Disk Drive Status LEDs
Each disk drive SBB has two LEDs that display its operating status
(Figure 2–6). The green device activity LED (CALLOUT 1) is ON or
FLASHING whenever the disk drive SBB is active.
Figure 2–6 Disk Drive Status LEDs
Figure CALLOUTS
1.

Device activity LED

Device fault LED

1
2

CXO4652B

The amber device fault LED (CALLOUT 2) FLASHES whenever the
array controller detects that this device has an error condition. This
LED also FLASHES when the array controller issues a locate
command. This LED is only ON or FLASHING when there is a device
malfunction, except when a locate command is issued.
For detailed information about the disk drive SBB status LEDs, see
Chapter 7.

UltraSCSI RAID Enclosure 2–17

Operating the UltraSCSI Enclosure

Controller Shelf Status Indicators
Controller shelf status indicators are provided by the array controllers,
EMU, PVA, and I/O modules.
Typical Controller Status LEDs
The operator control panel (OCP) status LEDs (see Figure 2–7)
display the controller status. For a detailed explanation of each status
code, refer to the controller documentation.
Figure 2–7 Typical Array Controller OCP LEDs
1

Figure CALLOUTS

Reset button (contains the array
controller LED)

Port buttons (6 each)

Port/fault LEDs (6 each)

Host port power fault LED

4
CXO5732B

The green array controller LED (CALLOUT 1) inside the Reset button
indicates controller status. This LED flashes constantly once the
controller initialization is complete and the software is functioning.
Pressing this switch resets the controller.
The amber port LEDs (CALLOUT 3) are OFF when the bus is
functioning properly. A port LED that is ON or FLASHING indicates
that a device on that bus is not functioning properly.

2–18 UltraSCSI RAID Enclosure

Operating the UltraSCSI Enclosure

EMU Status LEDs
The EMU status LEDs (see Figure 2–8) have multiple functions (see
Chapter 3).
Figure 2–8 EMU Status LED Locations
Figure CALLOUTS
1

System fault LED
(amber)

Temperature fault
LED (amber)

Power status LED
(green)

Blower fault, error
condition, and fault
code LEDs (amber)

CXO6302A

PVA Status LED
The PVA (see Figure 2–9) has only one status LED, the power status
LED. For a detailed description of the PVA status displays and alarm,
see Chapter 4.
Figure 2–9 PVA Status LED Location
Figure CALLOUTS

The green power
status LED is part of
the standby power
switch.

CXO6303A

UltraSCSI RAID Enclosure 2–19

Operating the UltraSCSI Enclosure

I/O Module LEDs
Each I/O module (see Figure 2–10) has two green status LEDs that
display the status of both the internal (CALLOUT 2) and external
(CALLOUT 1) SCSI bus terminator power (TERMPOWER).
Whenever TERMPOWER is present, the applicable LED (internal or
external) is ON. During normal operation, both LEDs are ON. See
Chapter 5 for detailed information about these LEDs.
Figure 2–10 I/O Module TERMPOWER LEDs
Figure CALLOUTS

2
CXO5829B

2–20 UltraSCSI RAID Enclosure

External
TERMPOWER

Internal
TERMPOWER

3
Environmental Monitoring
Unit
The primary function of the EMU (Figure 3–1) is to monitor, process,
report, and display power distribution, temperature, blower,
configuration, SCSI addressing, I/O module, communications, and
microcode status information. The EMU and the array controller
exchange and process some of this information.
Figure 3–1 EMU

CXO-5497A-MC

Caution
Proper operation of an UltraSCSI subsystem requires an
operational EMU and PVA in each enclosure. Also,
communication links between EMUs are required in
multiple enclosure configurations to ensure proper
subsystem operation and error reporting.

UltraSCSI RAID Enclosure 3–1

Environmental Monitoring Unit

Product Description
The EMU detects enclosure error conditions and configuration faults,
and then notifies the user of existing or impending failures, using one
or more of the following error reporting systems:
•

EMU LEDs

•

EMU audible alarm

•

Error messages on the host interface

•

Enclosure LEDs

The EMU (CALLOUT 1) mounts directly above array controller “A”
(upper bus controller) as shown in Figure 3–2.
Figure 3–2 EMU Location
Figure CALLOUTS
1.

EMU

PVA

2
CXO5754B

In some instances [such as a blower failure, a high intake air
temperature, or a high internal enclosure (exhaust) temperature], the
EMU automatically initiates corrective action. For example, the EMU
will command the blowers to operate at high-speed to prevent
component damage due to an overtemperature condition. In a severe
case, the EMU might even turn OFF the enclosure power.

3–2 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Caution
Some array controllers bypass the standby power mode
functionality from the PVA standby power switch. Refer to
the array controller documentation for availability of this
feature.

Note
An overtemperature condition (greater than 50°C) causes
the EMU to turn the enclosure dc power OFF. When any
enclosure has an overtemperature condition, the cache
module data is preserved prior to turning OFF the enclosure
dc power.

Controller Status
The master EMU monitors the state of both array controllers. If the
EMU detects a controller fault, the EMU can implement the following
actions:
•

Sound an audible alarm

•

Turn the EMU system fault status LED to FLASHING

•

Display a controller fault code on the blower LEDs (only
displayed if the user momentarily presses the alarm control
switch)
Note
An error on the EMU-array controller communications path
might also cause a controller fault indication.

UltraSCSI RAID Enclosure 3–3

Environmental Monitoring Unit

I/O Module Status
The EMU verifies that the six I/O modules are present, properly
installed, and communicating with the backplane, and that
TERMPOWER is at the correct level. The EMU reports an error
condition whenever any one of these conditions are not met. The
EMU also reports the I/O module type to the array controller. If the
controller determines that all modules are not the same type, it
displays this information on the optional maintenance terminal, if
connected.
An integrated circuit on each I/O module functions as a bus
expander. The EMU and the array controller can either enable or
disable this circuit on an individual I/O module, thereby controlling
individual external SCSI buses.

Enclosure Configuration Information
The EMU maintains the following configuration information:
•

Enclosure number (the PVA SCSI ID switch setting)

•

EMU microcode revision

•

EMU message protocol revision

•

Temperature sensor set points

•

Number of installed power supplies by location

•

Number of installed disk drive SBBs by location

•

Number of operational blowers in each drive SBB shelf bank

•

Number and type of I/O modules installed

3–4 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

EMU Front Panel
The EMU user interface controls, connectors, and LED displays are
on the front panel (see Figure 3–3). The following sections describe
the function of each component.
Figure 3–3 EMU Major Component Locations
1

Figure CALLOUTS
2

CXO5774A

EMU communications connector
(labeled IIC)

System fault LED (amber) and alarm
control switch

Temperature fault LED (amber)

Power status LED (green)

Maintenance terminal connector

Blower fault LEDs (amber)

EMU communications connector
(labeled IIC)

EMU Communications Connectors
These connectors (CALLOUTS 1 and 7) provide for EMU–EMU
communications between the master EMU and expansion EMUs.
System Fault LED and Alarm Control Switch
The amber system fault LED is part of the alarm control switch
(CALLOUT 2). This LED turns ON if an error condition is detected.
The system fault LED starts FLASHING if the EMU has one or more
fault codes to display.
If the system fault LED is FLASHING, momentarily pressing the EMU
alarm control switch turns the audible alarm OFF and initiates a fault
code LED display.
Pressing the alarm control switch for at least 5 seconds clears all
stored active fault codes. New fault detection and display continues.
If the enclosure is in standby power mode, momentarily pressing the
alarm control switch restores dc power to the enclosure.

UltraSCSI RAID Enclosure 3–5

Environmental Monitoring Unit

Temperature Fault LED
If either the intake or exhaust air temperature exceeds user-defined
temperature setpoints, this amber LED (CALLOUT 3) turns ON. The
LED remains ON until the overtemperature condition is corrected.
Power Status LED
This green power status LED (CALLOUT 4) turns ON if all the
following normal operating conditions exist:
•

+5 V dc is greater than +4.7 V dc.

•

+12 V dc is greater than +11.4 V dc.

•

At least four power supplies are operational.

•

TERMPOWER is present on all six I/O modules.

The power status LED turns O FF if any one of the following error
conditions exist:
•

+5 V dc is less than +4.7 V dc.

•

+12 V dc is less than +11.4 V dc.

•

Fewer than four power supplies are operational.

•

TERMPOWER is missing from one or more I/O modules.

Maintenance Terminal Connector
This connector (CALLOUT 5) is used to connect the optional
maintenance terminal or a personal computer (PC) to the EMU for
displaying EMU:
•

Error messages

•

Information messages

This connector and a PC are also used to load microcode updates into
EMU memory.
Blower Fault LEDs
One or more of these eight amber blower fault LEDs (CALLOUT 6)
turn ON if one of the following error conditions exist:
•

One or more blowers are not operational.

•

One or more blowers are not operating at the correct speed.

•

A blower is removed.

3–6 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Configuring the EMU
Configuring the EMU requires connecting the EMU-EMU
communications bus and using the command line interpreter (CLI)
set emu command. This command establishes operating
temperature set points and desired blower speed operating mode
(high-speed or automatic).

Connecting the EMU Communication Bus
In an expansion configuration, an EMU-EMU communications bus
between the master EMU and each expansion EMU must be
established. To establish this communications bus, connect the EMU
communications connector using a BN26M series cable, as shown in
Figure 3–4. The center EMU in this figure is the master EMU.
Establishing this EMU-EMU communication bus ensures that the
master EMU can monitor and control all the enclosures within the
UltraSCSI subsystem.
Figure 3–4 EMU Communications Bus Connections

CXO5820A

Compaq recommends using a BN26M cable (see Table 3–1) no
longer than 4 m (13.1 ft) to connect the EMUs.

UltraSCSI RAID Enclosure 3–7

Environmental Monitoring Unit

Table 3–1

EMU Communication Cables
Length

Description

8-conductor 24 AWG, PVC, round
cable assembly with:
2 – 8 position (8MP) locking,
plug connectors

Meters

Feet

0.5
1.0
3.0
4.0

1.6
3.3
9.8
13.1

Part
Number
BN26M–0E
BN26M–01
BN26M–03
BN26M–04

CXO5747A

EMU Firmware
EMU firmware allows the user to use CLI commands to establish
limits for EMU functions such as temperature and blower speed
control. Different firmware versions determine the specific
commands or functions available. The master EMU compares the
temperature sensor readings and blower speed readings against limits
the user entered.
If the EMU is controlling blower speed (automatic mode), the
blowers are automatically placed into high-speed mode when any
sensor reports an airflow temperature that is within 4 degrees Celsius
(°C) or 7 degrees Fahrenheit (°F) of the sensor set point. For example,
if a temperature set point is 37°C (99°F), the EMU would cause the
blowers to operate at high-speed when the temperature reaches 33°C
(91°F). Operating the blowers in high-speed should prevent an
overtemperature condition from occurring and should prevent
shutting down the enclosure.
Compaq recommends using predetermined set points for EMU
firmware versions 1.0, 1.1, and 1.2, as identified in the following
sections.

3–8 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Caution
For EMU firmware versions 1.0 and 1.1, the command
set emu sensor_n_setpoint=default cannot be
used. This command establishes a temperature set point that
is too low for proper operation and might result in an
erroneous subsystem alarm.
Removing and replacing the master EMU automatically
changes the temperature set points. This situation requires
the user to reenter all the temperature set points using CLI
commands.

EMU Firmware Versions 1.0 and 1.1
To set the temperature set point for an EMU, the following CLI
commands must be used:
set emu sensor_1_setpoint=48
set emu sensor_2_setpoint=48
set emu sensor_3_setpoint=41
Where: 1 = backplane sensor 1 (exhaust)
2 = backplane sensor 2 (exhaust)
3 = EMU sensor (intake)

EMU Firmware Version 1.2
To set the temperature set point for an EMU, the following CLI
commands must be used:
set emu sensor_1_setpoint=default
set emu sensor_2_setpoint=default
set emu sensor_3_setpoint=default
Where: 1 = backplane sensor 1 (exhaust)
2 = backplane sensor 2 (exhaust)
3 = EMU sensor (intake)

UltraSCSI RAID Enclosure 3–9

Environmental Monitoring Unit

Note
Default temperature settings for EMU firmware version 1.2
are:
• Sensors 1 and 2 — 47
• Sensor 3 — 37

Installing EMU Firmware Updates
EMU firmware updates are installed by Compaq field service
personnel. Appendix C provides general microcode installation
procedures.

Setting the Temperature Sensors
Three temperature sensors are found in each UltraSCSI RAID
enclosure:
•

One mounted on the EMU module.

•

Two on the rear of the backplane, placed at the top middle.

The EMU sensor monitors the EMU intake air temperature, while the
two enclosure sensors monitor the backplane exhaust air temperature.
Naturally, the exhaust air temperature is higher than the EMU intake
air temperature during normal operation.
In an expansion enclosure configuration, only master EMU
temperatures are set. The temperature set points for the other
enclosures automatically change to match the master EMU setting.
Users have the option of setting the temperature at which an
individual sensor reports an overtemperature condition. The desired
temperature is entered in degrees Celsius within the range of 0°C
(32°F) through 49°C (120°F). See Table 3–2 for the relationship
between °C and °F.
Table 3–3 and Table 3–4 explains the rules to always follow when
setting temperature set points.

3–10 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Note
For detailed instructions about using the set emu
commands, refer to the array controller CLI documentation.

Table 3–2

EMU Set Point Temperature Conversions

°F

°C

°F

°C

°F

°C

°F

°C

°F

°C

104

106

108

109

111

113

115

117

100

118

102

120

Table 3–3

Temperature Set Point Rules—EMU Firmware
Versions 1.0 and 1.1

Enter all temperatures in degrees Celsius (°C). Do not use fractions,
decimals, or degrees Fahrenheit (°F).

The EMU set point (sensor_3) temperature must be a minimum of
6°C (11°F) greater than the highest expected ambient (intake)
temperature.

The backplane set points (sensor_1 and sensor_2) must be the same
temperature and a minimum of 13°C (23°F) greater than the highest
expected ambient temperature. This requires a minimum setting of
7°C (13°F) greater than the EMU set point.

UltraSCSI RAID Enclosure 3–11

Environmental Monitoring Unit

Table 3–3
4.

Temperature Set Point Rules—EMU Firmware
Versions 1.0 and 1.1 (Continued)

EMU firmware versions 1.0 or 1.1 cannot use the set emu
sensor_n_setpoint=default command. Instead, the
following command is required:
set emu sensor_n_setpoint=nn
(where nn is the desired temperature in °C).

Table 3–4

Temperature Set Point Rules—EMU Firmware
Versions 1.2

Enter all temperatures in degrees Celsius (°C). Do not use fractions,
decimals, or degrees Fahrenheit (°F).

The EMU set point (sensor_3) temperature must be a minimum of
2°C (4°F) greater than the highest expected ambient (intake)
temperature.

The backplane set points (sensor_1 and sensor_2) must be the same
temperature and a minimum of 12°C (22°F) greater than the highest
expected ambient temperature. This requires a minimum setting of
10°C (18°F) greater than the EMU set point.

Setting the Blower Speed Control
UltraSCSI enclosure blowers normally operate at low-speed and
provide sufficient air flow to cool the enclosure components and
ensure proper operation. Use the CLI set emu fanspeed
command as follows to change the operating speed of the blowers:
•

Enter set emu fanspeed=high and all blowers in all
enclosures operate at high-speed.

•

Enter set emu fanspeed=automatic and the local
EMU controls the blower speed in each enclosure.

3–12 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Note
For detailed instructions about using the set emu
commands, refer to the array controller documentation.
Compaq recommends using the
set emu fanspeed=automatic command, allowing
individual EMUs to implement corrective action if one of the
following error conditions occur:
•

A blower fails.

•

A blower is removed.

•

A blower is rotating too slowly to provide sufficient air flow in
the shelf.

•

The EMU intake air temperature exceeds the user-defined
temperature setting.

•

The backplane exhaust air temperature exceeds the user-defined
temperature setting.

Compaq suggests using the set emu fanspeed=high command
for either of the following conditions:
•

EMU intake air temperature is more than 38°C (100°F).

•

No air conditioning is available.

If the EMU detects a blower or temperature error condition, it
automatically switches all the operational blowers to high-speed. This
speed increases the air flow through the enclosure and reduces the
possibility of component failure. Simultaneously, the EMU initiates
the following actions:
•

Sounds the audible alarm on the EMU.

•

Turns ON the amber system fault LED if the temperature exceeds
an established set point.

•

Turns ON the amber temperature fault LED.

•

Notifies the array controller of the error condition.

UltraSCSI RAID Enclosure 3–13

Environmental Monitoring Unit

•

Starts the EMU timer if: (1) all blowers in a shelf bank are
defective, or (2) a blower is removed.

•

Approximately eight minutes after the EMU timer begins
operating, the array controllers will shut down and then the EMU
turns OFF the dc power in the enclosure to prevent continued
overheating and potential equipment damage.

Momentarily depressing the alarm control switch causes the EMU
audible alarm to turn OFF, but the fault status LEDs remains ON. At
this time, the blower LEDs display system fault codes that are
essential to pinpointing the fault. See the Configuration Fault Code
Reporting section within this chapter.

EMU Status Reporting
The EMU processes subsystem status reports and notifies the user of
problems via an EMU audible alarm and EMU status LEDs. Figure
3–5 identifies the numerous EMU status LEDs.
Figure 3–5 EMU Status LEDs

Figure CALLOUTS

5
CXO5748B

System fault LED

Alarm control
switch

Temperature LED

Power LED

Blower LEDs

The EMU processes two types of problems:
•

Error conditions—involves the failure of a component or
conditions outside the predetermined environment, such as an
overtemperature condition.
If an error condition is detected, the system fault LED
(CALLOUT 1) turns ON. For detailed information about other
system fault LED indications, see the Error Condition Reporting
section within this chapter.

3–14 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

•

Fault conditions—involves a subsystem configuration problem,
such as a Boot ROM checksum failure.
If a fault condition is detected, the system fault LED
(CALLOUT 1) starts FLASHING. For detailed information about
other system fault LED indications, see the Configuration Fault
Code Reporting section within this chapter.

Whenever the audible alarm sounds, check the EMU status LEDs to
determine the problem.
Caution
If the audible alarm sounds, but the status LEDs do not
indicate an error condition, it is possible that the error
condition no longer exists. However, always check the error
log on the optional maintenance terminal to verify that the
error no longer exists.
The master EMU reports the status of all enclosures, including error
conditions, to the array controller. The array controller records some
of this information in an error log and displays the log on the
controller console.

Alarm Control Switch
If a power, environment, blower error, or fault condition occurs, the
EMU turns ON its audible alarm and the amber system fault LED.
The alarm remains on until one of the following occurs:
•

The error condition is corrected.

•

The user momentarily presses the alarm control switch.

Momentarily pressing the alarm control switch (CALLOUT 2, Figure
3–5) turns OFF the audible alarm for all current fault conditions
within the enclosure. The alarm remains OFF until one of the
following conditions occur:
•

A different fault occurs.

•

An existing fault clears and reoccurs.

UltraSCSI RAID Enclosure 3–15

Environmental Monitoring Unit

•

The error condition still exists after 1 hour has passed. This
situation causes the audible alarm to continue sounding once
every hour for approximately 5 seconds.

To clear all the fault code displays, press and hold the alarm control
switch for at least 5 seconds.
Note
The alarm control switch does not affect either the error
condition or the error condition code stored by the EMU.
The only way to clear a stored error condition code is to
correct the problem.
If the enclosure is in standby power mode, the alarm control switch
functions as an enclosure power switch. Press the alarm control
switch on each enclosure to restore dc power to the enclosure.

EMU Status LEDs
Table 3–5 shows the three indications possible for EMU status LEDs.
If an LED’s blinking or flashing interval is more than 1 second, the
LED is considered ON.
Table 3–5

Subsystem Status LEDs Displays

Symbol

Condition
LED is OFF.
LED is FLASHING (blinking rapidly—blinks more than
once per second).
LED is ON or blinking slowly (blinks less than once per
second).

3–16 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Monitoring Power Supply Operations
Even with four operational power supplies, it is possible that a power
supply problem might cause one or both of the dc voltages to be too
low. Both the PVA and EMU monitor power supply operations to
ensure that dc voltages are within a specified range (see Table 3–6).
Table 3–6

DC Voltage Ranges

Nominal Voltage

+5 V dc

+12 V dc

Minimum Voltage

+4.7 V dc

+11.4 V dc

At least four operational power supplies are required to maintain the
minimum voltage requirements shown in Table 3–6. With less than
four operational power supplies, the EMU notifies the array
controllers of a fault condition. The array controllers shut down and
then the EMU initiates standby power mode.
Caution
Some array controllers bypass the standby power mode
functionality from the PVA standby power switch. Refer to
the array controller documentation for availability of this
feature.

Note
Only the EMU monitors the I/O module +5 V dc
TERMPOWER. The PVA does not monitor this voltage.

UltraSCSI RAID Enclosure 3–17

Environmental Monitoring Unit

Error Condition Reporting
EMU status LEDs identify system (enclosure) status, power status,
temperature status, blowers status, and error conditions. During
normal operations, only the power LED is ON. Refer to Figure 3–5 as
necessary to recall EMU status LED names and locations. Figures
contained in Table 3–7 describe:
•

EMU status LED display appearance

•

Enclosure status associated with each display

•

A probable cause for the detected error condition

•

Action recommended to correct the problem
Note
Presenting all possible LED display combinations is not
practical in this document. Therefore, Table 3–7 lists only a
representative sampling of the displays possible.

Table 3–7

EMU Status Displays
EMU Status

EMU LED Display

Operational Enclosure
The subsystem is fully operational.
The power supplies and blowers are
functioning properly. The EMU intake
and backplane exhaust air temperatures
are within user-specified limits.

3–18 UltraSCSI RAID Enclosure

CXO5778A

Environmental Monitoring Unit

Table 3–7

EMU Status Displays (Continued)

Nonoperational Enclosure
8

CXO5780A

Probable Cause 1—Either +5 V dc or 12 V dc out of range. Observe the individual power supply
status LEDs to determine the defective power supply and replace it.
Probable Cause 2—TERMPOWER less than +4.5 V dc on one or more I/O modules. Replace the
I/O modules displaying a TERMPOWER LED OFF indication.
Single Blower Fault
Operational blowers are running at
high-speed.

CXO5781A

Missing Blower Fault
or
Multiple Blower Faults

Operational blowers are running at
high-speed.
CXO5782A

Probable Cause 1—A blower is removed. Replace the missing blower.
Probable Cause 2—All blowers in one bank are not operating. Remove and install blowers correctly
within 8 minutes.
Probable Cause 3—Multiple blowers are installed incorrectly. Remove and install blowers correctly
within 8 minutes.
Comment—In this example, one of the following conditions exist in shelf bank 1:
•

A blower is removed.

•

Two or more blowers are not operating.

Failure to replace the missing or defective blowers within 8 minutes might result in the enclosure
entering standby power mode.

UltraSCSI RAID Enclosure 3–19

Environmental Monitoring Unit

Table 3–7

EMU Status Displays (Continued)

Configuration Error (Example 1)
8

CXO5783A

Probable Cause 1—Configuration error. See the Configuration Fault Code Reporting section for the
specific error condition.
Probable Cause 2—Slave EMU cannot communicate with the master EMU. See the Configuration
Fault Code Reporting section for the specific error condition.
Probable Cause 3—External SHELF_OK signal indicates error. See the Configuration Fault Code
Reporting section for the specific error condition.
Configuration Error (Example 2)
Configuration errors listed in probable
causes 1 through 4 also sound the
audible alarm.

CXO6546A

Probable Cause 1—Invalid cache module installed. Verify that the cache module is the proper
model.
Probable Cause 2—Defective cache module. Replace the defective cache module.
Probable Cause 3—Defective ECB. Replace the defective ECB.
Probable Cause 4—Misconfigured array controller setup. Setup as dual redundant, but only one
controller is running. Setup as nondual redundant, but two controllers are running. Reconfigure the
array controller setup.
Probable Cause 5—Defective or misconfigured disk drive. Replace or reconfigure the applicable
disk drive.

3–20 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Table 3–7

EMU Status Displays (Continued)

Overtemperature Condition
8

CXO5779A

Probable Cause 1—A temperature in one or more enclosures is above the user specified limit.
Determine and correct the cause of this condition quickly to prevent an overtemperature condition
from inducing a subsystem RESET or failure.
Probable Cause 2—A temperature set point is too low for this environment. Change the temperature
set point.

Configuration Fault Code Reporting
The system fault, temperature, power, and blower LEDs cannot
display all possible system faults. Therefore, if the system fault LED
is FLASHING, associate this indication with a system fault somewhere
within the enclosure. To identify a system fault, observe the blower
LEDs. They represent a hexadecimal code that is presented using the
a binary format. The upper LEDs (8 through 5) display the first digit
(the most significant digit) of the hexadecimal code; the lower LEDs
(4 through 1) display the second digit (the least significant digit) of
the hexadecimal code.
Note
Fault codes are not displayed by the blower LEDs until the
user momentarily presses the FLASHING alarm control
switch.

UltraSCSI RAID Enclosure 3–21

Environmental Monitoring Unit

Each time the alarm control switch is momentarily pressed, the EMU
initiates the following sequence of actions:
1. The blower LEDs display the first reported hexadecimal fault
code three times in succession.
— The LEDs display the first reported fault code (for the first
time) for 1 second and then turn OFF.
— The LEDs display the first reported fault code (for the second
time) for 1 second and then turn OFF.
— The LEDs display the first reported fault code (for the third
time) for 1 second and then turn OFF.
2. After the first reported fault code is displayed three times, the
LEDs sequence the display to present the next reported fault
code.
3. Steps 1 and 2 repeat for all reported fault codes.
For a detailed explanation of possible fault codes, see Table 3–8. The
numbered corrective actions represent the sequence to follow to
resolve the problem.
.

Table 3–8

EMU Fault Code Displays

Fault Code, Cause, Corrective
Action

EMU LED Display

Fault Code 01
One of the following Boot ROM error
conditions exists:

CXO6598A

A Boot ROM Signature failure. Replace the EMU.

A Boot ROM Checksum failure. Replace the EMU.

3–22 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Table 3–8

EMU Fault Code Displays (Continued)

Fault Code, Cause, Corrective
Action

EMU LED Display

Fault Code 02
Both microcode images are bad.
1.

Load microcode image into the
EMU from a PC.

Replace the EMU.

CXO5761A

Fault Code 03
Either a PVA SCSI ID switch setting is
invalid (set to 1, 4, 5, 6, or 7) or there is
no master enclosure identified (SCSI
bus ID = 0).

CXO5762A

Ensure PVA SCSI ID switches settings are as follows:
•

Master enclosure — 0

•

First expansion enclosure — 2

•

Second expansion enclosure — 3

If any PVA SCSI ID switch setting is changed, the controllers must be RESET to recognize the
new setting.

Replace each of the PVAs in sequence (first the master, then the expansions) until the fault
condition clears.

UltraSCSI RAID Enclosure 3–23

Environmental Monitoring Unit

Table 3–8

EMU Fault Code Displays (Continued)

Fault Code, Cause, Corrective
Action

EMU LED Display

Fault Code 04
Master EMU cannot communicate with
expansion EMU.

CXO5763A

Ensure PVA SCSI ID switches settings are as follows:
•

Master enclosure — 0

•

First expansion enclosure — 2

•

Second expansion enclosure — 3

If any PVA SCSI ID switch setting is changed, the controllers must be RESET to recognize the
new setting.

Ensure the EMU cables are properly connected.

Disconnect and reconnect each EMU cable one at a time. Replace the defective cable.

Replace the EMU.

Fault Code 05
Master EMU cannot communicate with
the array controller.
1.

Replace the EMU.

Replace the array controller.

CXO5764A

Fault Code 06
One or more temperature sensors
indicate an “invalid” temperature—one
that is greater than 49°C (120°F).

CXO5765A

Replace the EMU.

Instruct field service that temperature sensors need replacing.

3–24 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Table 3–8

EMU Fault Code Displays (Continued)

Fault Code, Cause, Corrective
Action

EMU LED Display

Fault Code 07
Invalid configuration.

Requires an array controller-specific
action (see the array controller
documentation).

CXO5766A

Fault Code 08
Two PVA SCSI ID switch settings are
the same.

CXO5767A

Ensure the PVA SCSI ID switch settings are as follows:
•

Master enclosure — 0

•

First expansion enclosure — 2

•

Second expansion enclosure — 3

If any PVA SCSI ID switch setting is changed, the controllers must be RESET to recognize the
new setting.

Replace each of the PVAs in sequence (first the master, then the expansions) until the fault
condition clears.

Fault Code 09
An EMU is installed in a PVA slot.
Remove the EMU from the PVA slot and
install it into the slot immediately to the
left.

CXO5768A

UltraSCSI RAID Enclosure 3–25

Environmental Monitoring Unit

Table 3–8

EMU Fault Code Displays (Continued)

Fault Code, Cause, Corrective
Action

EMU LED Display

Fault Code 0A
No PVA is installed.
Install a PVA.

CXO5769A

Fault Code 0B
Controller fault.

Requires an array controller-specific
action (see the array controller
documentation).

CXO5770A

Fault Code 0C
One or more defective power supplies.

Observe all power supply status LEDs.
Replace any power supply with the
lower (power supply status) LED turned
OFF.

CXO5771A

Fault Code 0D
The master enclosure SCSI bus ID
address is incorrect or an array controller
is installed in an expansion enclosure.

CXO5772A

Change the master enclosure SCSI bus ID to 0. The controllers must be RESET to recognize the
new setting.

If necessary, remove the array controller from the expansion enclosure and install it in the master
enclosure.

3–26 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Table 3–8

EMU Fault Code Displays (Continued)

Fault Code, Cause, Corrective
Action

EMU LED Display

Fault Code 0E
EMU memory test failed.

Replace the EMU.
CXO5773A

Fault Code 0F
Less than four operational power
supplies remain.
Replace the defective power supplies.
Comment—The EMU places the
system in the standby power mode for
this fault indication. After replacing the
defective power supplies, momentarily
press the alarm control switch to restore
dc power to the enclosure.
The initial display appears when a fault
is detected; the standby power mode
display appears after the EMU removes
dc power from the enclosure.

Initial Display

CXO5784A

Standby Power Mode Display

CXO5815A

Fault Code 10
EMU diagnostic program is loaded.
The audible alarm is on continuously
and it cannot be cleared by pressing the
alarm control switch.

CXO5814A

Load new microcode. The controllers must be RESET to recognize the new microcode.

Replace the EMU with one containing the proper microcode. The controllers must be RESET to
recognize the new microcode.

UltraSCSI RAID Enclosure 3–27

Environmental Monitoring Unit

Replacing an EMU
Complete the procedure in Table 3–9 to remove an EMU and
complete the procedure in Table 3–10 to install an EMU.
Table 3–9

Removing an EMU

Clear the cache module and shut down the array controller per the
array controller documentation.

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1) on the
master enclosure until the master EMU initiates a power shut down
on the master enclosure.

If there are no expansion enclosures, go to step 5.

Turn O FF all the ac input
boxes for the master
enclosure by pressing O on
each ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn OFF all the
ac input boxes for each
expansion enclosure by
pressing O on each ac power
switch (CALLOUT 1).

CXO5488B

3–28 UltraSCSI RAID Enclosure

Environmental Monitoring Unit

Table 3–9
6.

Removing an EMU (Continued)

Pull the EMU extractor
latches outward and
remove the EMU.

CXO-5497A-MC

Table 3–10 Installing an EMU
1.

Verify that the EMU is a
model DS–BA35X–EB.

CXO-5497A-MC

Align the EMU in the
top, left-hand slot
(CALLOUT 1) and insert
it.

2
CXO5754B

UltraSCSI RAID Enclosure 3–29

Environmental Monitoring Unit

Table 3–10 Installing an EMU (Continued)
3.

Pull the EMU extractor latches outward. Push the EMU into the slot
until the extractor latches engage the enclosure, then push the
extractor latches inward to seat and secure the EMU.

If there are no expansion enclosures, go to step 6.
For an expansion configuration, use a BN26M series cable (see Table
3–1) to create the EMU-EMU communications bus connections as
illustrated below.

CXO5820A

Turn O N all the ac input
boxes for the master
enclosure by pressing I on
each ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn ON all the ac
input boxes for each
expansion enclosure by
pressing I on each ac power
switch (CALLOUT 1).

CXO5488B

After the array controller is operational, verify that the subsystem is
functioning properly.

3–30 UltraSCSI RAID Enclosure

4
Power Verification and
Addressing
An enclosure’s PVA assembly (see Figure 4–1) and associated EMU
monitors the status of major UltraSCSI enclosure components to
verify their proper operation. If an error or fault condition occurs,
these modules notify the user, and in some instances, the array
controller, that a problem exists. The array controller can identify and
display information about some problems. The PVA and EMU status
LEDs indicate possible root causes of problems.
Figure 4–1 PVA

CXO-5737A

Caution
Proper operation of an UltraSCSI RAID subsystem requires
an operational EMU (model DS–BA35X–EB) and PVA
(model DS–BA35X–EC) in each enclosure. Also, EMU
communications bus links must be established between all
EMUs for proper subsystem operation and error reporting.

UltraSCSI RAID Enclosure 4–1

Power Verification and Addressing

The PVA and EMU mount next to each other in the enclosure—the
EMU (CALLOUT 1) is on the left and the PVA is on the right
(CALLOUT 2) as shown in Figure 4–2.
Figure 4–2 PVA Location
Figure CALLOUTS
1.

EMU

PVA

2
CXO5754B

Product Description
Figure 4–3 identifies the major PVA components.
Figure 4–3 PVA Major Component
1

Figure CALLOUTS
2

CXO5821A

4–2 UltraSCSI RAID Enclosure

Standby power
switch

Communications
port

SCSI ID switch

Power Verification and Addressing

The PVA has two status indicators:
•

A green power status LED that is part of the standby power
switch (CALLOUT 1). This LED is ON if there are at least four
operational power supplies. If an error condition occurs, the LED
turns OFF.
Caution
Some array controllers bypass the standby power mode
functionality from the PVA standby power switch. Refer to
the array controller documentation for availability of this
feature.

Note
The standby power switch initiates the standby power mode
when the array controller is shut down. In this mode, ac
power is applied to the ac input box and the power supplies.
However, dc power is disconnected from the backplane.
•

An audible alarm that sounds when the PVA detects removal of
the EMU. This alarm reminds the user to install an operational
EMU for restoring the subsystem to an operational status.

Functions
The primary functions of the PVA include the following:
•

Monitors the enclosure EMU status and notifies the user when the
EMU is missing.

•

Allows the user to select a Compaq-supported device SCSI bus
address (device ID) for each enclosure.

•

Ensures at least four operational power supplies are in the
enclosure.

•

Notifies the array controller if less than four operational power
supplies are functioning in the enclosure. The array controller
then removes power from the enclosure.

UltraSCSI RAID Enclosure 4–3

Power Verification and Addressing

This action—removing power from the enclosure—duplicates an
EMU function and ensures optimum subsystem performance and
integrity.

PVA Operation
Most PVA operations occur automatically and do not require user
intervention. The following sections define major PVA operations.

EMU Installation Monitoring
If the EMU is removed or installed incorrectly (for example, not fully
seated), the PVA detects this condition and sounds the PVA audible
alarm. This alarm continues until the PVA senses that an operational
EMU is present.
Caution
Removing the EMU and PVA at the same time will cause
the system to RESET and cause conflicts between the
enclosures due to duplicate SCSI bus ID settings.

Disk Drive SBB Device IDs
The 24 disk drive SBBs in each enclosure are arranged on 6 vertical
buses in 4 horizontal shelves. All six devices on the same horizontal
shelf use the same device address or ID, but each device uses a
separate vertical bus. See Figure 4–4 for the master enclosure disk
drive SBB device IDs.

4–4 UltraSCSI RAID Enclosure

Power Verification and Addressing

Figure 4–4 Master Enclosure Device IDs
Figure CALLOUT

SCSI bus

3
2
1
0
CXO5755B

Use the PVA SCSI ID switch to establish SCSI bus IDs for all devices
in the enclosure. This switch has eight (0 through 7) positions.
Caution
To preclude SCSI bus addressing conflicts, Compaq only
supports PVA SCSI ID switch settings of 0, 2, and 3.
Always set the master enclosure PVA SCSI ID switch setting to 0.

UltraSCSI RAID Enclosure 4–5

Power Verification and Addressing

Monitoring Power Supply Operations
Even with four operational power supplies, it is possible that a power
supply problem might cause one or both of the dc voltages to be too
low. Both the PVA and the EMU monitor power supply operations to
ensure that the dc voltages are within the ranges specified in Table
4–1.
Table 4–1

DC Voltage Ranges

Nominal Voltage

+5 V dc

+12 V dc

Minimum Voltage

+4.7 V dc

+11.4 V dc

With at least four operational power supplies, the minimum voltage
levels sufficient to run the system are present. Fewer than four
operational power supplies cannot maintain the minimum voltages. If
this occurs, the EMU notifies the array controllers so they can shut
down before standby power mode is initiated.
Caution
Some array controllers bypass the standby power mode
functionality from the PVA standby power switch. Refer to
the array controller documentation for availability of this
feature.

Note
The EMU monitors the system dc voltages and I/O module
+5 V dc TERMPOWER. The PVA only monitors the
system dc voltages.

4–6 UltraSCSI RAID Enclosure

Power Verification and Addressing

Setting the Enclosure Address
The PVA SCSI ID switch establishes the SCSI bus address (device
ID) for all disk drive SBBs in the enclosure. Table 4–2 identifies how
to set the PVA SCSI ID switch setting for each enclosure.
Table 4–2

Setting the Enclosure Configuration

To decrease the configuration number:
Press the upper switch (CALLOUT 1) to step the
address down one address at a time (decrement).

0
To increase the configuration number:
Press the lower switch (CALLOUT 2) to step the
address up one address at a time (increment).

2
CXO4851C

The following restrictions apply to establishing an enclosure address:
•

The master enclosure address is always 0.

•

There is only one master enclosure in any subsystem.

•

No two enclosures in a subsystem are allowed to have the same
address.

See Table 4–3 for the Compaq-supported enclosure addresses.
Caution
All devices on the same SCSI bus must have a unique SCSI
bus ID.
To preclude SCSI bus addressing conflicts, Compaq only
supports PVA SCSI ID switch settings of 0, 2, and 3.
If the PVA SCSI ID switch setting is changed for an
enclosure, the controllers must be RESET to recognize the
new setting.

UltraSCSI RAID Enclosure 4–7

Power Verification and Addressing

Table 4–3

Compaq-Supported PVA SCSI ID Switch
Settings

Total Number
of Enclosures

Master
Enclosure

First
Expansion

Second
Expansion

None

Master Enclosure Address
The master enclosure PVA SCSI ID switch setting is always 0. This
setting establishes the disk drive SBB SCSI bus IDs shown in Figure
4–5.
Figure 4–5 Master Enclosure Address 0: Shelf and SCSI
Bus IDs
Figure CALLOUT

3
0
CXO-4840A-MC

2
1
0
CXO5755B

4–8 UltraSCSI RAID Enclosure

SCSI bus

Power Verification and Addressing

Expansion Enclosure Addresses
Compaq supports only PVA SCSI ID switch settings of 2 (see Figure
4–6) and 3 (see Figure 4–7) for expansion enclosures:
•

2 — Used for the first expansion enclosure.

•

3 — Used for a second expansion enclosure.
Caution
To preclude SCSI bus addressing conflicts, Compaq array
controllers do not support enclosure address settings of 1, 4,
5, 6, or 7.

Figure 4–6 Expansion Enclosure Address 2: Shelf and
SCSI Bus IDs
Figure CALLOUT

SCSI bus

11
2
CXO-4842A-MC

10
9
8
CXO5776B

UltraSCSI RAID Enclosure 4–9

Power Verification and Addressing

Figure 4–7 Expansion Enclosure Address 3: Shelf and
SCSI Bus IDs
Figure CALLOUT

SCSI bus

15
3

CXO-4843A-MC

13
12
CXO5777B

Implementing the Standby Power Mode
A major function of the PVA is implementing the standby power
mode. This mode removes dc power from all enclosure components,
except for the EMU alarm control switch. In standby power mode, the
EMU alarm control switch now functions as a dc power switch when
pressed. Complete the procedure in Table 4–4 to place the subsystem
into standby power mode. Complete the procedure in Table 4–5 to
restore dc power to each enclosure.
Caution
Some array controllers bypass the standby power mode
functionality from the PVA standby power switch. Refer to
the array controller documentation for availability of this
feature.

4–10 UltraSCSI RAID Enclosure

Power Verification and Addressing

Table 4–4

Turning Off the DC Power Distribution

Clear the cache module and shut down the array controller per the
array controller documentation.

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1 in step 2)
on the master enclosure until the EMU initiates a power shut down
on the master enclosure.

Note
The EMU alarm control switch (CALLOUT 1 shown below) on
each enclosure now functions as the dc power switch.
1

3
4

CXO6302A

UltraSCSI RAID Enclosure 4–11

Power Verification and Addressing

Table 4–5

Turning On the DC Power Distribution

If there are no expansion enclosures, go to step 3.

If there are expansion enclosures, momentarily press the EMU alarm
control switch (CALLOUT 1) on each expansion enclosure.
1

3
4

CXO6302A

Momentarily press the EMU alarm control switch (CALLOUT 1) on
the master enclosure.

Verify that the array controller restores the subsystem to an
operational state.

Replacing a PVA
Complete the procedures in Table 4–6 to remove a PVA. Complete
the procedures in Table 4–7 to install a PVA.
Table 4–6
1.

Removing a PVA

Clear the cache module and shut down the controller per array
controller documentation.

4–12 UltraSCSI RAID Enclosure

Power Verification and Addressing

Table 4–6
2.

Removing a PVA (Continued)

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1) on the
master enclosure until the master EMU initiates a power shut down
on the master enclosure.

Turn O FF all the ac input
boxes by pressing O on each
ac power switch
(CALLOUT 1) on any
expansion enclosure and
then the master enclosure.

1
1
0

CXO5488B

UltraSCSI RAID Enclosure 4–13

Power Verification and Addressing

Table 4–6
5.

Removing a PVA (Continued)

Note the enclosure address setting (0, 2, or 3) on the PVA SCSI ID
switch (CALLOUT 3).
1

CXO5821A

Pull both extractor latches outward and remove the PVA.

Table 4–7
1.

Installing a PVA

Verify that the PVA being installed is a model DS–BA35X–EC.

CXO-5737A

Align the PVA (CALLOUT 2) with the
top, right-hand slot and insert it.

2
CXO5754B

4–14 UltraSCSI RAID Enclosure

Power Verification and Addressing

Table 4–7

Installing a PVA (Continued)

Pull the PVA extractor latches outward. Push the PVA into the slot
until the extractor latches engage the enclosure, then push the
extractor latches inward to seat and secure the PVA.

Ensure that the PVA SCSI ID switch is set to the valid number for the
enclosure:

•

Master enclosure — 0

•

First expansion enclosure — 2

•

Second expansion enclosure — 3

If there are no expansion enclosures, go to step 6.

Turn O N all the ac input
boxes for the master
enclosure by pressing I on
each ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn ON all the ac
input boxes for each
expansion enclosure by
pressing I on each ac power
switch (CALLOUT 1).

CXO5488B

After the array controller is operational, verify that the subsystem is
functioning properly.

UltraSCSI RAID Enclosure 4–15

5
Single-Ended I/O Module
The single-ended I/O module (see Figure 5–1) performs two
functions:
•

Provides termination for both the internal and external SCSI
buses

•

Establishes SCSI bus communications between enclosures

Figure 5–1 Single-Ended I/O Module

CXO5816A

Caution
I/O modules provide termination for both the internal and
external SCSI buses. Therefore, install the same type of
operational I/O modules on each of the six enclosure
connectors.

UltraSCSI RAID Enclosure 5–1

Single-Ended I/O Module

Note
Not all array controller firmware supports an UltraSCSI
RAID subsystem with more than one enclosure. Refer to
the array controller documentation to ensure that the
controller supports expanded configurations.
Six I/O modules mount on the bottom rear (see Figure 5–2,
CALLOUT 1) of each enclosure. These modules connect the array
controller to the devices in the expansion enclosures. In an UltraSCSI
RAID subsystem, the I/O module number corresponds to the SCSI
bus number. For example, SCSI bus 2 and I/O module 2 are different
elements of the same physical bus.
Figure 5–2 I/O Module Location
Figure CALLOUT
1.

2
4
6
5

SCSI bus
ports (1–6) for
I/O modules
(1–6),
respectively

3
1
CXO5805B

Odd numbered I/O ports are located in the bottom row and even
number ports are located in the top row (see Figure 5–3). Viewing
from the enclosure rear, port 1 is in the lower right corner and port 6
is in the upper left corner.
Figure 5–3 I/O Module Numbering
6

1
CXO5819A

5–2 UltraSCSI RAID Enclosure

Single-Ended I/O Module

I/O Module and Expansion Cable Error Conditions
Removal of an I/O module causes the EMU to report the following
error conditions to the controller:
•

Installation of different I/O modules

•

Loss of TERMPOWER

Either of these error conditions might cause the controller to cease
operation without implementing procedures to protect the data.
Before removing an I/O module or an expansion cable, completion of
the following procedures is mandatory to prevent inducing a
controller error condition:
1. Clear the cache and shut down the array controller.
2. Turn OFF the enclosure power.
Caution
Replacing an I/O module or an expansion cable without
first shutting down the controller can adversely affect data
integrity.

Expanding the UltraSCSI Bus
Each I/O module has an UltraSCSI single-ended bus expansion
integrated circuit that isolates the internal and external SCSI bus,
enhances SCSI signals, and allows the use of a longer UltraSCSI bus
cable.
Two 68-pin VHDCI receptacle connectors on the I/O module are the
connection points for expanding the SCSI buses between enclosures.
These connectors are wired in parallel and act as a trilink connector.
In an expansion configuration, Compaq recommends a maximum
cable length of 2 m (6.6 ft) between enclosures.

UltraSCSI RAID Enclosure 5–3

Single-Ended I/O Module

Controlling the Internal SCSI Bus
The I/O module controls the internal SCSI bus in the following
manner:
•

Isolates the internal SCSI bus from the external SCSI bus

•

Provides single-ended SCSI bus termination for each internal bus

•

Disconnects the internal SCSI bus from the external SCSI bus
when directed by the EMU

•

Distributes TERMPOWER (+5 V dc) to the internal SCSI bus

•

Turns ON the green internal TERMPOWER LED when internal
TERMPOWER is present

•

Turns OFF the green internal TERMPOWER LED when there is
either an internal TERMPOWER overcurrent condition or no
TERMPOWER

Controlling the External SCSI Bus
The I/O module controls the external SCSI bus in the following
manner:
•

Provides single-ended SCSI bus termination for the external bus

•

Distributes TERMPOWER (+5 V dc) to the external SCSI bus

•

Turns ON the green external TERMPOWER LED when external
TERMPOWER is present

•

Turns OFF the green external TERMPOWER LED when there is
either an external TERMPOWER overcurrent condition or no
TERMPOWER

•

Disables the external bus termination when a cable is connected
to the right VHDCI connector

5–4 UltraSCSI RAID Enclosure

Single-Ended I/O Module

Controlling the Fault Bus
The I/O module controls fault bus operation in the following manner:
•

Provides a fault bus driver for improved signal transmission

•

Distributes the FAULT_CLK and FAULT_DATA signals from the
master enclosure to the expansion enclosures

•

Distributes the SHELF_OK and SWAP_L signals from the
expansion enclosures to the master enclosure

Error Reporting
The EMU monitors the status of the I/O modules for the following
error conditions:
•

Installation of incompatible I/O modules

•

TERMPOWER problems

Incompatible I/O Modules
An I/O module error condition exists if any of the following
conditions exist:
•

One or more nonoperating I/O modules

•

Installation of fewer than six operational I/O modules

•

All I/O modules are not the same type

When the EMU detects an I/O module error condition, it
automatically reports the problem to the array controller. Any
operational controller will automatically turn OFF without
implementing procedures to clear the cache module. The result is that
the controller will cease to operate and be in an error state, a state that
can adversely affect the integrity of the data.
Note
Existance of an I/O module error condition prevents the
subsystem from operating.

UltraSCSI RAID Enclosure 5–5

Single-Ended I/O Module

Only when all I/O modules are the same type can an UltraSCSI RAID
subsystem function properly. The Compaq UltraSCSI controller only
supports single-ended buses. Therefore, use only single-ended I/O
modules (DS–BA35X–MN). The enclosure does not support the
following conditions:
•

Improperly installing a single-ended I/O module

•

Installing fewer than six single-ended I/O modules

•

Installing a termination I/O module

•

Installing a differential I/O module

TERMPOWER Errors
The I/O module displays a TERMPOWER status indication for both
the internal and external bus as shown in Figure 5–4.
Figure 5–4 I/O Module TERMPOWER LEDs
Figure CALLOUTS

2
CXO5829B

External
TERMPOWER

Internal
TERMPOWER

The nominal TERMPOWER for both buses is +5 V dc.
The green internal (CALLOUT 2) and external (CALLOUT 1)
TERMPOWER LEDs are ON during normal operation.
If a TERMPOWER overcurrent condition exists or there is no
TERMPOWER, the associated LED is OFF.

5–6 UltraSCSI RAID Enclosure

Single-Ended I/O Module

See Table 5–1 for possible causes of TERMPOWER problems.
Table 5–1

TERMPOWER Problems

TERMPOWER Error

Possible Causes

Internal

I/O module
Power supply

External

I/O module
Power supply
External SCSI bus cable

Removing an I/O module disconnects the SCSI bus termination,
resulting in a loss of TERMPOWER available to the internal SCSI
bus. Without TERMPOWER, the bus cannot function correctly.

Rules for Expanding an UltraSCSI Bus
By connecting the master enclosure to one or two expansion
enclosures, each SCSI bus is expanded from a 4-device bus to an 8- or
12-device bus, respectively. See Chapter 4 for a discussion of valid
enclosure SCSI bus addresses and the associated array controller and
storage device SCSI bus IDs.
Note
Wide UltraSCSI supports a maximum of 16 devices per bus
(2 array controllers and 14 disk drives). However, Compaq
does not support any configuration with more than 12 disk
drives.
To expand a SCSI bus, use the 68-pin VHDCI receptacle connectors
on the I/O module and a BN37A series cable. See Table 5–2 for a list
of the rules for expanding an UltraSCSI bus.

UltraSCSI RAID Enclosure 5–7

Single-Ended I/O Module

Caution
The Compaq UltraSCSI controller does not support
connecting or disconnecting a SCSI bus cable when both
data and power are present on the connector—a “hot
swap.” Only connect or disconnect SCSI bus cables using
the “cold swap” method.

Table 5–2
1.

UltraSCSI Bus Expansion Rules

The I/O module numbers (1 through 6) correspond to the enclosure
SCSI bus numbers.
6

1
CXO5819A

Always connect cables between enclosures to the same I/O module
number. That is, connect I/O module 1 on the master enclosure to I/O
module 1 on an expansion enclosure.

The first cable on any I/O module always connects to the left
connector.

Note
Connecting a cable to the right I/O module connector removes
the external SCSI bus termination from the module.

5–8 UltraSCSI RAID Enclosure

The second cable on any I/O module always connects to the right
connector.

Single-Ended I/O Module

Cabling an UltraSCSI Subsystem
This section provides procedures for adding a second and third
enclosure (expansions) to a subsystem. The first expansion enclosure
connects to the master enclosure. A second expansion enclosure
connects to either the master enclosure or the first expansion
enclosure. Before beginning any cabling installation, complete the
preliminary procedures in Table 5–3.
Caution
Not all array controllers support an UltraSCSI RAID
subsystem with more than one enclosure. Refer to the array
controller documentation to ensure the controller supports
expanded configurations.
The Compaq UltraSCSI controller does not support
connecting or disconnecting a SCSI bus cable when both
data and power are present on the connector—a hot swap.
Only connect or disconnect SCSI bus cables using the cold
swap method.

Table 5–3

Preliminary Cabling Procedures

Determine how many enclosures will be connected to the master
enclosure (0, 1, or 2).

Determine if a second master enclosure will connect to the master
enclosure or the first expansion enclosure.

Determine the optimum route for each UltraSCSI bus cable.

Use Table 5–4 to select the UltraSCSI cable for each segment—based
upon the distances between the enclosures.

Caution
To ensure proper UltraSCSI bus operation, Compaq
recommends a maximum cable length of 2 m (6.6 ft).

UltraSCSI RAID Enclosure 5–9

Single-Ended I/O Module

Table 5–4

UltraSCSI Bus Cables
Length

Description

68-conductor SCSI cable with:
2 — VHDCI straight plug
connectors with jack screws

Meters

Feet

0.3
0.5
1.0
1.5
2.0

1.0
1.6
3.3
4.9
6.6

Order
Number
BN37A–0C
BN37A–0E
BN37A–01
BN37A–1E
BN37A–02

CXO5702A

Shutting Down the Subsystem
To protect the data and devices, complete the procedures in Table 5–3
and Table 5–5 before beginning cable installation procedures for a
two- or three-enclosure subsystem.

5–10 UltraSCSI RAID Enclosure

Single-Ended I/O Module

Table 5–5

Shutting Down the UltraSCSI Subsystem

Clear the cache module and shut down the array controller per the
array controller documentation.

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1) on the
master enclosure until the master EMU initiates a power shut down
on the master enclosure.

If there are no expansion enclosures, go to step 5.

Turn O FF all the ac input
boxes for the master
enclosure by pressing O on
each ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn OFF all the
ac input boxes for each
expansion enclosure by
pressing O on each ac power
switch (CALLOUT 1).

CXO5488B

UltraSCSI RAID Enclosure 5–11

Single-Ended I/O Module

Cabling a Two-Enclosure Subsystem
Complete the procedures in Table 5–6 to connect the master enclosure
to one expansion enclosure. Perform this procedure for the I/O
modules on each SCSI bus, starting with SCSI bus 1 in Figure 5–3.
Each SCSI bus requires one BN37A cable for a total of six cables.
Caution
Not all array controllers support an UltraSCSI RAID
subsystem with more than one enclosure. Refer to the array
controller documentation to ensure the controller supports
expanded configurations.
Overtightening the I/O module spring-loaded mounting
screws can damage the threads.

Table 5–6

Installing Two-Enclosure Subsystem
Cables

Complete Table 5–5 to shut down the UltraSCSI subsystem.

Use a screwdriver to loosen two spring-loaded mounting screws on
the I/O module.

Grasp the I/O module by
the cable support bracket
and pull it straight
outward until removed
from the enclosure
shroud.
CXO5816A

Repeat steps 2 and 3 for all the I/O modules.

5–12 UltraSCSI RAID Enclosure

Single-Ended I/O Module

Table 5–6
5.

Installing Two-Enclosure Subsystem
Cables (Continued)

On the master enclosure
I/O module, align the
Cable A (CALLOUT A)
connector with the left
connector. Gently insert
the cable connector into
the module connector.
Tighten both cable
connector thumb screws
to fully seat the
connector.

CXO5854B

Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.

On the master enclosure, align the module that has Cable A
(CALLOUT A) attached with the Port 1 backplane connector. Gently
insert the module and fully seat it. Use a screwdriver to tighten the
spring-loaded mounting screws.
6

On the expansion I/O
module, align the Cable A
(CALLOUT A) connector
with the left connector.
Gently insert the cable
connector into the module
connector. Tighten both
cable connector thumb
screws to fully seat the
connector.

CXO5854B

Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.

UltraSCSI RAID Enclosure 5–13

Single-Ended I/O Module

Table 5–6

Installing Two-Enclosure Subsystem
Cables (Continued)

10. On the expansion enclosure shroud, align the module that has Cable
A (CALLOUT A) attached with the Port 1 backplane connector.
Gently insert the module and fully seat it. Use a screwdriver to
tighten the spring-loaded mounting screws.
6

A
6

1
CXO5840B

11. Repeat steps 5 through 10 for the remaining SCSI buses (ports 2
through 6).
12. Turn ON the subsystem using Table 5–7 and verify that it is operating
properly.

5–14 UltraSCSI RAID Enclosure

Single-Ended I/O Module

Turning On the Subsystem
Complete the procedures in Table 5–7 to turn ON the subsystem.
Table 5–7

Turning On the UltraSCSI Subsystem

If there are no expansion enclosures, go to step 3.

Turn O N all the ac input
boxes for each expansion
enclosure by pressing I on
each ac power switch
(CALLOUT 1).

1
1
0

On the master enclosure,
turn O N all the ac input
boxes by pressing I on each
ac power switch
(CALLOUT 1).

CXO5488B

Verify that the array controller places the subsystem into an
operational state.

UltraSCSI RAID Enclosure 5–15

Single-Ended I/O Module

Cabling a Three-Enclosure Subsystem
Complete the procedures in Table 5–8 to connect the master enclosure
to two expansion enclosures. The second expansion enclosure
connects to either the master enclosure or the first expansion
enclosure. Table 5–8 connects the second expansion enclosure to the
first expansion enclosure. Perform this procedure for the I/O modules
on each SCSI bus, starting with SCSI bus 1 (identified in Figure 5–3).
Each SCSI bus requires two BN37A cables for a total of 12 cables.
Caution
Not all array controllers support an UltraSCSI RAID
subsystem with more than one enclosure. Refer to the array
controller documentation to ensure the controller supports
expanded configurations.
Overtightening the I/O module spring-loaded mounting
screws can damage the threads.

Note
The procedure in Table 5–8 connects a second expansion
enclosure to the first expansion enclosure. To connect the
second expansion enclosure to the master enclosure, do the
following:
• Substitute the first expansion enclosure for the master
enclosure in steps 5 through 7.
•

5–16 UltraSCSI RAID Enclosure

Substitue the master enclosure for the first expansion
enclosure in steps 8 through 12.

Single-Ended I/O Module

Table 5–8

Installing Three-Enclosure Subsystem
Cables

Complete Table 5–5 to shut down the UltraSCSI subsystem.

Use a screwdriver to loosen the two spring-loaded mounting screws
on the I/O module.

Grasp the I/O module by
the cable support bracket
and pull it straight
outward until removed
from the enclosure
shroud.
CXO5816A

Repeat steps 2 and 3 for all the I/O modules within the three
enclosures.

CXO5854B

Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.

UltraSCSI RAID Enclosure 5–17

Single-Ended I/O Module

Table 5–8
7.

Installing Three-Enclosure Subsystem
Cables (Continued)

On the master enclosure shroud, align the module that has Cable A
(CALLOUT A) attached with the Port 1 backplane connector. Gently
insert the module and fully seat it. Use a screwdriver to tighten the
spring-loaded mounting screws.
6

On the first expansion I/O
module, align the Cable A
(CALLOUT A) connector
with the left connector.
Gently insert the cable
connector into the module
connector. Tighten both
cable connector thumb
screws to fully seat the
connector.

CXO5854B

Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.

10. Using the expansion I/O module in step 8, align the Cable B
(CALLOUT B) connector with the right connector. Gently insert the
cable connector into the module connector. Tighten both cable
connector thumb screws to fully seat the connector.
B

CXO5855B

5–18 UltraSCSI RAID Enclosure

Single-Ended I/O Module

Table 5–8

Installing Three-Enclosure Subsystem
Cables (Continued)

11. Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.
12. On the first expansion enclosure shroud, align the module that has
Cable B (CALLOUT B) attached with the Port 1 backplane connector.
Gently insert the module and fully seat it. Use a screwdriver to
tighten the spring-loaded mounting screws.
6

B
6

13. On the second expansion
enclosure I/O module,
align the Cable B
(CALLOUT B) connector
with the left connector.
Gently insert the cable
connector into the module
connector. Tighten both
cable connector thumb
screws to fully seat the
connector.

2
A

CXO5854B

14. Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.

UltraSCSI RAID Enclosure 5–19

Single-Ended I/O Module

Table 5–8

Installing Three-Enclosure Subsystem
Cables (Continued)

15. On the second expansion enclosure shroud, align the module that has
Cable B (CALLOUT B) attached with the Port 1 backplane connector.
Gently insert the module and fully seat it. Use a screwdriver to
tighten the spring-loaded mounting screws.
6

B
6

1
CXO5841B

16. Repeat steps 5 through 15 for the remaining SCSI buses (ports 2
through 6).
17. Turn ON the subsystem using Table 5–7 and verify that it is operating
properly.

5–20 UltraSCSI RAID Enclosure

Single-Ended I/O Module

Replacing an I/O Module
Complete the procedure in Table 5–9 to remove an I/O module.
Complete the procedure in Table 5–10 to install an I/O module.
Caution
To prevent interrupting a data transfer or losing data, shut
down the subsystem before removing an I/O module.
Overtightening the I/O module spring-loaded mounting
screws can damage the threads.

Table 5–9

Removing an I/O Module

Shut down the subsystem using Table 5–5.

Use a screwdriver to loosen two spring-loaded mounting screws on
the I/O module.

Grasp the I/O module by
the cable support bracket
and pull it straight
outward until removed
from the enclosure
shroud.
CXO5940A

Cut the wire tie securing the left cable. Loosen the thumb screws and
remove the cable.

Label the left cable connector.

If a right cable is attached, cut the wire tie securing the right cable.
Loosen the thumb screws and remove the cable.

Label the right cable connector.

Place the I/O module into an electrostatic bag.

UltraSCSI RAID Enclosure 5–21

Single-Ended I/O Module

Table 5–10 Installing an I/O Module
1.

Verify that the I/O
module being installing is
a model
DS–BA35X–MN.

CXO5816A

Note
Jumpers are factory installed—do not modify jumper settings.

Align the cable with the
left I/O module connector.
Gently insert the cable
connector into the module
connector. Tighten both
cable connector thumb
screws to fully seat the
connector.

CXO5940A

Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.

If attaching a cable to the right I/O module connector is necessary,
repeat steps 2 and 3 using the right I/O module connector.

Align the I/O module with the backplane connector in the enclosure
shroud. Gently insert the module and fully seat it. Use a screwdriver
to tighten the spring-loaded mounting screws.

Turn ON the subsystem using Table 5–7 and verify that it is operating
properly.

5–22 UltraSCSI RAID Enclosure

Single-Ended I/O Module

Replacing an Expansion Cable
Complete the procedure in Table 5–11 to remove an expansion cable.
Complete the procedure in Table 5–12 to install an expansion cable.
Caution
Not all array controllers support an UltraSCSI RAID
subsystem with more than one enclosure. Refer to the array
controller documentation to ensure that the controller
supports expanded configurations.
To prevent interrupting a data transfer or losing data, shut
down the subsystem before removing an I/O module.
Overtightening the I/O module spring-loaded mounting
screws can damage the threads.

Table 5–11 Removing an Expansion Cable
1.

Shut down the subsystem using Table 5–5.

Use a screwdriver to loosen the two spring-loaded mounting screws
on the I/O module connected to the cable that is being removed.

Grasp the module by the cable support bracket and pull it directly to
the rear and remove it from the enclosure shroud.

Repeat steps 2 and 3 for the I/O module attached to the other end of
the cable that is being removed.

Cut the wire ties on that
secure the cable to the
I/O modules. Loosen the
thumb screws and
remove the cable.

CXO5940A

UltraSCSI RAID Enclosure 5–23

Single-Ended I/O Module

Table 5–12 Installing an Expansion Cable
1.

Verify that the expansion cable is
the proper type and length per
Table 5–4.
CXO5702A

Align the cable with the
left I/O module
connector. Gently insert
the cable connector into
the module connector.
Tighten both cable
connector thumb screws
to fully seat the
connector.

CXO5940A

Install a wire tie through the holes on the cable support bracket and
around the cable. Tighten the wire tie.

Repeat steps 2 and 3 for the other end of the cable. This end of the
cable might connect to a left or right I/O module connector,
depending on the enclosures involved.

Align the I/O module with the backplane connector in the enclosure
shroud. Gently insert the module and fully seat it. Use a screwdriver
to tighten the spring-loaded mounting screws.

Turn O N the subsystem and verify that it is operating properly.

5–24 UltraSCSI RAID Enclosure

6
Power Distribution
This chapter describes the UltraSCSI RAID enclosure power
distribution to include the following:
•

Enclosure power system overview (see Figure 6–1)

•

Power distribution configurations (standard, redundant, and
optimum)

•

Error condition reporting

•

Replacement procedures (power supply and ac input box)

Figure 6–1 Enclosure Power System

CXO5800A

UltraSCSI RAID Enclosure 6–1

Power Distribution

Note
Proper operation of the controller cache modules requires
ECBs. Refer to the array controller documentation for
discussion on how to use the ECBs.

Power System Overview
In any RAID subsystem, the elimination of single points of failure is
taken into consideration. Designing systems with redundant power
capabilities is one way to eliminate single points of failure.
The following major components compose the power system:
•

AC power sources

•

AC input boxes

•

Power cords

•

Shelf power supplies
Note
The ac input box power cords are integral to the data center
cabinets. For the departmental server cabinets, the
user-supplied power cords are country-specific.
The drive SBB shelf ac power cords are integral parts of the
UltraSCSI enclosures.

Power Configurations
For all StorageWorks products with the capability of using redundant
power, the power configurations definitions follow a n+x format. In
this format, n is the minimum number of power supplies required for
operation and x is the number of redundant power supplies. Each
UltraSCSI RAID enclosure requires a minimum of four (n) power
supplies to operate and a specified number of power supplies for
redundancy purposes.

6–2 UltraSCSI RAID Enclosure

Power Distribution

Three configurations exist:
•

Standard (n+1)

•

Redundant (n+4)

•

Optimum (n+4)

Standard (n+1)
This configuration has one redundant power supply for a total of five
power supplies. A single ac input box distributes ac power to all five
power supplies using power bus A. This configuration requires only
one ac power source.
Redundant (n+4)
This configuration has four redundant power supplies for a total of
eight power supplies. Two ac input boxes distribute ac power to four
power supplies on power bus A and four power supplies on power
bus B. This configuration requires only one ac power source that
supplies ac power to both ac input boxes.
Optimum (n+4)
This configuration has four redundant power supplies for a total of
eight power supplies. Two ac input boxes distribute ac power to four
power supplies on power bus A and four power supplies on power
bus B. This configuration requires two ac power sources, each
supplying ac power to a different ac input box.

AC Power Source
The ac input box requires one of the following ac power sources:
•

100–120 V ac, 60 Hz, single-phase, 12A

•

220–240 V ac, 50 Hz, single-phase, 6A
Note
The optimum power configuration requires two separate ac
power sources, one for each ac input box. These ac power
sources cannot be on the same power distribution leg, use
the same circuit breaker, nor the same ac wall receptacle.
They must have a common ground.

UltraSCSI RAID Enclosure 6–3

Power Distribution

AC Input Box
The ac power distribution is provided through the ac input box (see
Figure 6–2) to the shelf power supplies. The ac power switch on each
ac input box controls the ac distribution to the associated power bus
(A or B). These ac input boxes are completely interchangeable and
have no physical or electrical differences between them. AC input
boxes do not have error indicators on them, nor do they generate any
error signals. Without an optional second ac input box, an ac input
box failure disables the enclosure.
Figure 6–2 AC Input Box
Figure CALLOUTS

AC power switch

AC power receptacle

1
0

CXO5488B

Caution
The only compatible ac input box is the model
DS–BA35X–HE; model BA35X–HE cannot be used in an
UltraSCSI RAID enclosure.

6–4 UltraSCSI RAID Enclosure

Power Distribution

Shelf Power Supplies
Each shelf power supply (see Figure 6–3) converts the incoming ac
voltage into +5 V dc and +12 V dc and distributes these voltages
throughout the RAID enclosure.
The maximum capacity for the UltraSCSI RAID enclosure is eight
power supplies, four of which provide power redundancy.
The standard power configuration uses five power supplies, one of
which provides power redundancy.
Caution
UltraSCSI RAID enclosures require shelf power supplies
rated for at least 180 W, such as model DS–BA35X–HH.
Other power supplies, such as the 150 W power supply,
model BA35X–HF, cannot be used.

Figure 6–3 Shelf Power Supply
Figure CALLOUTS
1.

Shelf status LED

Power supply status LED

1
2

CXO4787B

UltraSCSI RAID Enclosure 6–5

Power Distribution

AC Power Buses
AC power distribution to the UltraSCSI shelf power supplies is
distributed over two power buses—bus A and bus B (see Figure 6–4).
Each power bus has an ac power source, a separate ac input box,
power supplies, and power cords.
Figure 6–4 Enclosure AC Power Buses
n+1 Configuration

n+4 Configurations
1

Figure CALLOUTS

4A 19 20 21 22 23 24 5A

Shelf 4

4A 19 20 21 22 23 24 4B

Shelf 4

3A 13 14 15 16 17 18

Shelf 3

3A 13 14 15 16 17 18 3B

Shelf 3

10 11 12

Shelf 2

10 11 12 2B

Shelf 2

Shelf 1

PVA
EMU
Controller A
Controller B
Cache A
Cache B

CXO6301B

Power bus A

Power bus B

CXO6482B

Although the standard (n+1) power configuration can operate the
subsystem via power bus A (CALLOUT 1), Compaq recommends
using both power buses with two ac power sources—the optimum
power configuration (n+4).
Power bus A (CALLOUT 1) uses black power cords and power bus B
(CALLOUT 2) uses grey power cords. These colored cords indicate the
power bus being used:
•

AC input box A controls power bus A—power supply positions
1A through 5A (n+1 configuration in Figure 6–4)

•

AC input box B controls power bus B— power supply positions
1B through 4B (n+4 configuration in Figure 6–4)

6–6 UltraSCSI RAID Enclosure

Power Distribution

Power Distribution Configurations
This section describes the three power configurations for the
UltraSCSI RAID enclosure.

Standard (n+1) Power Configuration
This power configuration provides minimum power supply
redundancy. AC power distribution is provided over power bus A to
five power supplies located in positions 1A through 5A (see Figure
6–5). Power bus A uses four black power cords on the left side of the
enclosure and one black power cord on the upper right corner.
Figure 6–5 Standard Power Configuration
Figure CALLOUT

1.
4A 19 20 21 22 23 24 5A

Shelf 4

3A 13 14 15 16 17 18

Shelf 3

10 11 12

Shelf 2

Shelf 1

Power bus A

CXO5929A

CXO6301B

The standard power configuration has the following items:
1 — ac power source
5 — shelf power supplies (locations 1A through 5A)
1 — ac input box (location A)

UltraSCSI RAID Enclosure 6–7

Power Distribution

Any one of the following power related error conditions will cause
this UltraSCSI RAID enclosure to cease operation:
•

Loss of the ac power source

•

Failure of two shelf power supplies

•

Failure of the ac input box

Redundant (n+4) Power Supply
Configurations
This power configuration provides more power redundancy than the
standard configuration. AC power distribution is provided over both
power buses (A and B, CALLOUTS 1 and 2, respectively) as shown in
Figure 6–6. Each bus has four power supplies for a total of eight.
Figure 6–6 Redundant Power Configuration
1

CXO5797A

Figure CALLOUTS

4A 19 20 21 22 23 24 4B

Shelf 4

3A 13 14 15 16 17 18 3B

Shelf 3

10 11 12 2B

Shelf 2

Shelf 1

PVA
EMU
Controller A
Controller B
Cache A
Cache B

Power bus A

Power bus B

CXO6482B

AC input box A distributes the ac power through four black cords at
the left side of the enclosure. AC input box B distributes the ac power
through four grey cords on the right side of the enclosure.

6–8 UltraSCSI RAID Enclosure

Power Distribution

To implement this configuration from an n+1 configuration, order the
following additional components:
•

3 — power supplies (DS–BA35X–HH)

•

1 — ac input box (DS–BA35X–HE)

This power configuration provides data protection through maximum
redundancy of all enclosure power system components. Any one of
the following power-related error conditions will cause this UltraSCSI
RAID enclosure to cease operation:
•

Loss of the ac power source

•

Failure of five shelf power supplies

•

Failure of both ac input boxes

Optimum (n+4) Power Supply Configurations
The only difference between this configuration and the redundant
(n+4) configuration is the use of two separate ac power sources. One
ac power source provides power to ac input box A; the other ac power
source provides ac power to ac input box B.
To implement this configuration from a standard n+1 configuration,
order the following additional components:
•

3 — power supplies (DS–BA35X–HH)

•

1 — ac input box (DS–BA35X–HE)

To implement this configuration from n+1 or an n+4 configuration
requires a second ac power source.
Compaq recommends using this power configuration to eliminate all
single points of power failure. The optimum power configuration
provides complete data protection with maximum redundancy of all
enclosure power system components and the ac input power source.
Any one of the following power related error conditions will cause
this UltraSCSI RAID enclosure to cease operation:
•

Loss of both ac power sources

•

Failure of both ac input boxes

•

Failure of five power supplies

UltraSCSI RAID Enclosure 6–9

Power Distribution

•

Failure of the ac power source for bus B and one power supply on
bus A

•

Failure of the ac power source on bus A and one power supply on
bus B

•

Failure of the ac input box on bus B and one power supply on
bus A

•

Failure of the ac input box on bus A and one power supply on
bus B

Error Condition Reporting
This section describes the power supply and ac input box error
condition reporting.

Power Supply
The two green LEDs on each power supply (see Figure 6–7) display
the blower, the power bus, and the individual power supply status.
Normally, both of these LEDs are ON.
Figure 6–7 Power Supply Status LEDs
Figure CALLOUTS
1.

Shelf status LED

Power supply status LED

1
2

CXO4787B

The green shelf status LED (CALLOUT 1) is ON when all the power
supplies and all the blowers are operational. If any blower or power
supply fails, this LED turns OFF.

6–10 UltraSCSI RAID Enclosure

Power Distribution

The green power supply status LED (CALLOUT 2) is O N when the
power supply is operating properly. If the power supply fails, both
power supply LEDs (CALLOUTS 1 and 2) turn OFF.
Table 6–1 identifies the possible fault displays and provides a
description of each state.
Table 6–1

Power Supply Status LED Displays

LED
Display

Description and Possible Corrective Actions
This power supply is functioning properly.

LED is ON

LED is OFF

A blower or power supply fault occurred:
•

Check the operating condition of each blower.
Replace the defective blower.

•

Check all power supplies for one with both LEDs in
the OFF state.
Replace this power supply.

A power fault has occurred:
1.

There is no ac power.
Check the ac power source.

This power supply has failed.
Replace the power supply.

AC Input Box
There are no ac input box status indicators. However, if all power
supply status LEDs turn OFF for a specific power bus, this indication
points to an ac input box failure or possibly an ac power source
failure.

UltraSCSI RAID Enclosure 6–11

Power Distribution

Replacing a Power Supply
This section describes how to properly handle power supply SBBs
and the procedures for replacing a power supply.

Handling a Power Supply SBB
Table 6–2 defines the proper methods for handling a power supply
SBB to prevent damaging it.
Table 6–2

SBB Handling Rules
Do . . .

Do NOT . . .

Set an SBB on its side.

Set an SBB on its edge.

CXO6549A

CXO6550A

Set SBBs side by side.

Stack SBBs.

CXO6551A

CXO6552A

Set an SBB down gently.

Drop an SBB.

CXO6553A

CXO6554A

Push an SBB into the shelf
gently, but firmly.

Force the SBB into the shelf.

CXO6555A

6–12 UltraSCSI RAID Enclosure

CXO6556A

Power Distribution

Procedures for Replacing a Power Supply
The basic procedure for removing and replacing shelf power supplies
is the same as for replacing disk drives.
There are two methods for replacing shelf power supplies—the hot
swap method and the cold swap method. Use the power supply status
LED indication, either operational or nonoperational, to determine
which swap method to use.
•

The hot swap method is normally used to replace one shelf power
supply without removing power from the enclosure.

•

The cold swap method is normally used only during initial
installation. This method requires making all devices inactive and
removing ac power from the UltraSCSI RAID enclosure.

Complete the procedures in Table 6–3 and Table 6–4 to replace a
shelf power supply.
Caution
To protect this sensitive electronic device from electrostatic
discharge (ESD), use the following precautions:
• Wear an ESD wrist strap.
•

Do not touch the printed circuit board or the backplane
connector.

•

Do not lay the device on a work surface; place it on an
electrostatic mat.

•

Place the device in an electrostatic bag for shipment.

To prevent ESD damage to a power supply SBB, do not
touch the SBB connector.

Table 6–3

Removing a Power Supply

To perform a hot swap, go to step 9.

To perform a cold swap, continue with step 3.

Clear the cache module and shut down the array controller per the
array controller documentation.

UltraSCSI RAID Enclosure 6–13

Power Distribution

Table 6–3
4.

Removing a Power Supply (Continued)

If there are no expansion enclosures, go to step 5.
If there are expansion enclosures, press and hold the PVA standby
power switch (CALLOUT 1) on an expansion enclosure until the
master EMU initiates a power shut down on all enclosures.
1

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1) on the
master enclosure until the master EMU initiates a power shut down
on the master enclosure.

If there are no expansion enclosures, go to step 7.

Turn O FF all the ac input
boxes for the master
enclosure by pressing O on
each ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn OFF all the
ac input boxes for each
expansion enclosure by
pressing O on each ac power
switch (CALLOUT 1).

CXO5488B

Disconnect the power cord from the front of the power supply.

6–14 UltraSCSI RAID Enclosure

Power Distribution

Table 6–3
9.

Removing a Power Supply (Continued)

Press the two mounting
tabs together to release
the power supply from
the shelf. Using both
hands, pull the power
supply out of the shelf.

CXO-4314A-MC

10. Place the power supply into an electrostatic bag.
11. Lay the power supply on its side on a flat surface.

Table 6–4
1.

Installing a Power Supply

Check the bezel label to verify that the power
supply being installed is a 180 W power supply.

CXO5926A

Lay the power supply on its side on a flat surface.

UltraSCSI RAID Enclosure 6–15

Power Distribution

Table 6–4
3.

Installing a Power Supply (Continued)

Insert the power supply
into the guide slots and
push it in until the bezel
mounting tabs engage
the shelf, fully seating
the power supply.

CXO-4314A-MC

Connect the power cord to the power supply.

For a cold swap method, go to step 6 to turn ON the subsystem.

If there are no expansion enclosures, go to step 7.

On the master enclosure,
turn O N all the ac input
boxes by pressing I on each
ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn ON all the ac
input boxes for each
expansion enclosure by
pressing I on each ac power
switch (CALLOUT 1).

CXO5488B

Verify the array controller places the subsystem into an operational
state.

Verify that both power supply status LEDs are ON.

6–16 UltraSCSI RAID Enclosure

Power Distribution

Replacing an AC Input Box
Complete the procedure in Table 6–5 to remove an ac input box.
Complete the procedure in Table 6–6 to install an ac input box.
WARNING!
Only qualified service personnel can replace an ac input
box.
To prevent the possibility of injury or death as a result
of electrical shock:
•

Always disconnect the ac power cord before
removing an ac input box.

•

Never touch the backplane connector or circuit
board.

•

Always install the ac input box before connecting the
ac power cord.

Table 6–5

Removing an AC Input Box

For redundant and optimum (n+4) power configurations, verify that
all eight power supplies are operational and then go to step 3.

For a standard (n+1) power configuration, clear the cache module
and shut down the array controller per the array controller
documentation.

Turn O FF the ac input box
being replaced by pressing
the O on the ac power switch
(CALLOUT 1).
1
1
0

CXO5488B

UltraSCSI RAID Enclosure 6–17

Power Distribution

Table 6–5

Removing an AC Input Box (Continued)

Disconnect the ac power cord (CALLOUT 4) from the ac input box
(CALLOUT 1).

Loosen the two retaining screws (CALLOUT 2) and remove the ac
input box (CALLOUT 1).

1
2
1
0

3
1
0

CXO5186B

Place the ac input box into an electrostatic bag.

Table 6–6

Installing an AC Input Box

On the ac input box being
installed, press O on the ac
power switch (CALLOUT 1).

1
0

Verify that the ac input box
being installed is a model
DS–BA35X–HE.
1

CXO5488B

6–18 UltraSCSI RAID Enclosure

Power Distribution

Table 6–6
3.

Installing an AC Input Box (Continued)

Align the ac input box (CALLOUT 1) with either slot A or slot B in the
lower left or lower right corner of the RAID enclosure, respectively.

1
2
1
0

3
1
0

CXO5186B

Push in on the ac input box to fully seat it.

Tighten the two retaining screws (CALLOUT 2).

Connect the ac power cord (CALLOUT 4) from the ac power source to
the ac input box (CALLOUT 1).

If necessary, repeat steps 1 through 6 to install a second ac input box.

Turn O N each ac input box
by pressing the I on the ac
power switch (CALLOUT 1).
1
1
0

CXO5488B

Verify that all power supply status LEDs turn ON.

UltraSCSI RAID Enclosure 6–19

7
Storage Devices
This chapter describes the procedures for handling, addressing,
identifying, and replacing disk drive SBBs (see Figure 7–1) in an
UltraSCSI RAID enclosure. Error conditions and the associated LED
displays are also described.
Figure 7–1 Typical Disk Drive SBB

CXO-4562A-MC

Caution
The Compaq UltraSCSI RAID enclosure supports 3.5-inch
disk drive SBBs. This enclosure does not support tape
drives, solid state disks, optical disk drives, CD-ROMs, or
5.25-inch devices.
For a list of supported devices, refer to the operating
system-specific release notes.

UltraSCSI RAID Enclosure 7–1

Storage Devices

Identifying a Storage Device
To determine the model number of a storage device (disk drive SBB),
look at the regulatory label (see Figure 7–2) on the side of the SBB or
at the disk drive SBB bezel label.
Figure 7–2 Typical Drive SBB Regulatory Label

CXO5916A

A typical drive SBB bezel label (see Figure 7–3) includes the
following device information:
•

Device type

•

SCSI bus type

•

User-defined information, such as:
— LUN # (logical unit number)
— ID # (device ID)
— CH # (channel number)

7–2 UltraSCSI RAID Enclosure

Storage Devices

Figure 7–3 Typical Drive SBB Bezel Label
1
2

Figure C ALLOUTS

DS-RZ1CB-VW
4.3 GB

LUN #

ID #

CH #

40
F20
W

Device model number

Device storage capacity

User-assigned SCSI LUN

User-assigned SCSI identification number (target ID)

User-assigned array controller channel number—the
array controller port or SCSI bus number

SCSI bus device width:

7
CXO5917A

N — an 8-bit device (narrow)
W — a 16-bit device (wide)
7.

StorageWorks shelf type (wide or narrow)
compatibility:
N — device is 8-bit shelf compatible (narrow)
W — device is 16-bit shelf compatible (wide)
N/W — device is 8- and 16-bit shelf compatible

Bus speed expressed in megabits per second (Mb/s):
S — Slow device. For example: S5 = 5 Mb/s
F — Fast device. For example: F10 = 10 Mb/s or F20
= 20 Mb/s

Maximum rate at which this device transfers
data—expressed in megabytes per second (MB/s)

Disk Drive SBB Device Addressing
Determining the disk drive SBB device address is a function of the
following:
•

PVA SCSI ID switch setting (0, 2, or 3)

•

Shelf number containing the disk drive SBB

UltraSCSI RAID Enclosure 7–3

Storage Devices

Disk Drive SBB Status Reporting
The array controller monitors the status of the disk drive SBBs. If a
fault occurs, the fault and the disk drive SBB device address (SCSI
target ID) are reported to the array controller or host for processing.
The disk drive status LEDs define the status of individual disk drive
SBBs.
Each disk drive SBB has two LED indicators (see Figure 7–4) that
display the SBBs status. These LEDs have three states: ON, OFF, and
FLASHING.
Figure 7–4 Disk Drive Status LEDs
Figure CALLOUTS
1.

Device activity LED

Device fault LED

1
2

CXO4652B

•

The green device activity LED (CALLOUT 1) is ON or FLASHING
when the disk drive SBB is active.

•

The amber device fault LED (CALLOUT 2) indicates an error
condition when it is either ON or FLASHING. The array controller
controls this LED. The device fault LED also FLASHES when the
array controller issues a “locate” command.
Caution
Removing a disk drive SBB when the upper LED is ON or
FLASHING might compromise data integrity.

7–4 UltraSCSI RAID Enclosure

Storage Devices

See Table 7–1 for definitions of the LED displays.
Table 7–1
LED
Display

Disk Drive Status LED Displays
Description and Possible Corrective Actions
The disk drive SBB is operating properly.

LED is ON

LED is
FLASHING

LED is OFF

The disk drive SBB is operating normally. The unit is
inactive and no fault is present.

The disk drive SBB is probably not responding to
control signals.
Compaq recommends replacing this storage SBB.
The disk drive SBB is active and in spin-down status
because of a fault.
Compaq recommends replacing the disk drive SBB after
the device has spun down.
Two conditions are possible:
1.

In response to a fault bus message, the array
controller is spinning down the device.
Compaq recommends replacing the disk drive SBB
after the device has spun down.

A signaling response to a user initiated array
controller “locate” command.

The disk drive SBB is inactive and spun down.
Replace the disk drive SBB.

UltraSCSI RAID Enclosure 7–5

Storage Devices

Guidelines for Replacing a Disk Drive SBB
This section describes the SBB connector and items to consider when
replacing a disk drive SBB.

The SBB Connector
The StorageWorks SBB connectors ensure that the disk drive SBBs
operate reliably, even when replaced many times.
The disk drive SBB has a 96-pin receptacle DIN connector that
provides positive mating with the SBB enclosure plug connectors.
This connector is designed to ensure that dc power is present:
1. Before the SCSI bus connection is made.
2. After the SCSI bus connection is broken.
This feature protects the integrity of the SCSI data bus and avoids
introducing noise on the bus that could either distort data or cause the
bus to “hang.”
The SBB connector is a simple, highly reliable mechanism that
supports 200 SBB replacement cycles. A replacement cycle consists
of removing and replacing an SBB. Over the active product life of 5
years, this is the equivalent of 40 replacement cycles per year, or
approximately 1 cycle every 9 days. The replacement cycle limit
applies equally to enclosure connectors and device connectors.
Replacing an SBB more often than the recommended cycle causes the
gold contact coating to wear away and destroy the integrity of the
connection. This product does not support environments that require a
greater number of replacement cycles, because this is considered
improper treatment or use (paragraph 9.4b of U.S. Standard Terms
and Conditions). Products or connectors damaged due to a higher
number of replacement cycles are not eligible for return under
warranty or standard service plans. Compaq markets a specific family
of removable storage elements for higher replacement cycle
environments. Contact your Compaq Account Representative for
more information.

7–6 UltraSCSI RAID Enclosure

Storage Devices

Prerequisites for Replacing a Disk Drive SBB
Replacing a disk drive SBB is a relatively simple procedure.
However, there are several things to consider.
When an array controller initializes a storage device, it establishes the
device’s physical location, the device address, and the LUN.
Physically moving the device causes the bus to become erratic.
Use the disk drive SBB bezel label to identify the device model,
LUN, device address, array controller bus, and other device specific
characteristics.
Caution
To protect this sensitive electronic device from ESD, use the
following precautions:
• Wear an ESD wrist strap.
•

Do not touch the printed circuit board or the backplane
connector.

•

Do not lay the device on a work surface; place it on an
electrostatic mat.

•

Place the device in an electrostatic bag for shipment.

Disk Drive SBB Replacement Methods
There are three methods for replacing disk drive SBBs: hot swap,
warm swap, and cold swap. Before replacing a device, determine the
appropriate replacement method based upon the capabilities of the
array controller.
Caution
The methods for replacing Compaq array controllers and
cache modules are device dependent. Refer to the array
controller documentation or release notes for detailed
instructions.

UltraSCSI RAID Enclosure 7–7

Storage Devices

Hot Swap
When performing a hot swap, power and data are present on the disk
drive SBB backplane connector.
Caution
Although the HSZ70 series array controllers support hot
swap, other controllers do not. Refer to the array controller
documentation to determine which array
controller-supported swap method to use.
If you are not positive that your SCSI controller supports
hot swap, Compaq recommends using warm swap to protect
data integrity.
Use hot swap to remove and replace disk drive SBBs from a system
that is online and active.
Warm Swap
When performing a warm swap, power is present on the disk drive
SBB plug connector and there is no data on the bus. Before
performing a warm swap, quiesce the bus at the array controller and
observe the green activity status LED on the disk drive SBB to ensure
that it is not FLASHING.
Caution
Perform a warm swap only when the green device activity
LED is OFF.

Cold Swap
When performing a cold swap, neither power nor data is present on
the disk drive SBB plug connector. This requires turning OFF the
enclosure power in all the UltraSCSI RAID subsystem enclosures.

7–8 UltraSCSI RAID Enclosure

Storage Devices

Replacing a Disk Drive SBB
This section describes how to properly handle disk drive SBBs and
the procedures for replacing a disk drive SBB.

Handling a Disk Drive SBB
Table 7–2 defines the proper methods for handling a disk drive SBB
to prevent damage.
Table 7–2

SBB Handling Rules
Do . . .

Do NOT . . .

Set an SBB on its side.

Set an SBB on its edge.

CXO6549A

CXO6550A

Set SBBs side by side.

Stack SBBs.

CXO6551A

CXO6552A

Set an SBB down gently.

Drop an SBB.

CXO6553A

CXO6554A

Push an SBB into the shelf
gently, but firmly.

Force the SBB into the shelf.

CXO6555A

CXO6556A

UltraSCSI RAID Enclosure 7–9

Storage Devices

Procedures for Replacing a Disk Drive SBB
Complete the procedure in Table 7–3 to remove a disk drive SBB.
When initially installing disk drive SBBs in an enclosure or installing
a single disk drive SBB, complete the procedure in Table 7–4.
Table 7–3

Removing a Disk Drive SBB

To perform a hot swap, go to step 8.

To perform a warm swap, quiesce the SCSI bus using array controller
documentation, then go to step 8.

To perform a cold swap, shut down the subsystem using array
controller documentation.

CXO6303A

Press and hold the PVA standby power switch (CALLOUT 1) on the
master enclosure until the master EMU initiates a power shut down
on the master enclosure.

7–10 UltraSCSI RAID Enclosure

Storage Devices

Table 7–3
6.

Removing a Disk Drive SBB (Continued)

If there are no expansion enclosures, go to step 7.

Turn O FF all the ac input
boxes for the master
enclosure by pressing O on
each ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn OFF all the
ac input boxes for each
expansion enclosure by
pressing O on each ac power
switch (CALLOUT 1).

CXO5488B

Press the disk drive SBB mounting tabs together to release the SBB.

Use both hands and pull
the disk drive SBB out
of the shelf.

CXO-5338A-MC

10. Place the disk drive SBB into an electrostatic bag.

UltraSCSI RAID Enclosure 7–11

Storage Devices

Table 7–4
1.

Installing a Disk Drive SBB

Verify that the disk drive SBB is the
correct model and fill in bezel label
information for this disk drive.

CXO-4562A-MC

If assembling an UltraSCSI enclosure, go to step 8.

For a hot or warm swap,
insert the disk drive
SBB into the guide slot
and push it in until the
mounting tabs engage
the shelf, fully seating
the disk drive SBB.

After installing the disk
drive SBB during a hot
swap, go to step 14.

CXO-5338A-MC

If performing a warm swap, refer to array controller documentation
to restart the subsystem, then go to step 14.

7–12 UltraSCSI RAID Enclosure

Storage Devices

Table 7–4
6.

Installing a Disk Drive SBB (Continued)

If performing a cold swap, determine if there are expansion
enclosures. If there are no expansion enclosures, go to step 7.

On the master enclosure,
turn O FF all the ac input
boxes by pressing O on each
ac power switch
(CALLOUT 1) and then go to
step 14.
Install all the disk drive SBB
devices, starting in the lower
left corner in shelf 1 (devices 1
through 6) and advancing to the
right.

1
1

If there are expansion
enclosures, turn OFF all the
ac input boxes for each
expansion enclosure by
pressing O on each ac power
switch (CALLOUT 1).

CXO5488B

4A 19 20 21 22 23 24 4B

Shelf 4

3A 13 14 15 16 17 18 3B

Shelf 3

10 11 12 2B

Shelf 2

Shelf 1

CXO5811B

Starting from the left, install all the disk drive SBB devices in shelf 2
(devices 7 through 12).

10. Starting from the left, install all the disk drive SBB devices in shelf 3
(devices 13 through 18).

UltraSCSI RAID Enclosure 7–13

Storage Devices

Table 7–4

Installing a Disk Drive SBB (Continued)

11. Starting at the left end, install all the disk drive SBB devices in
shelf 4 (devices 19 through 24).
12. If performing a cold swap, determine if there are expansion
enclosures. If there are no expansion enclosures, go to step 13.

13. On the master enclosure,
turn O N all the ac input
boxes by pressing I on each
ac power switch
(CALLOUT 1).

1
1

If there are expansion
enclosures, turn ON all the ac
input boxes for each
expansion enclosure by
pressing I on each ac power
switch (CALLOUT 1).

CXO5488B

14. Verify that the amber device fault LED is OFF.

Note
When the SCSI bus is active and the disk drive SBB has power,
the green device activity LED is either ON, FLASHING, or OFF.
The amber device fault LED is OFF.

7–14 UltraSCSI RAID Enclosure

8
Enclosure Blowers
Ensuring proper UltraSCSI RAID enclosure operation requires
establishing and maintaining the proper operating environment.
Meeting the temperature and humidity ranges specified in
Appendix A involves using air conditioning to establish and maintain
the proper external, that is, ambient, environment. Failure to achieve
these goals might cause components to malfunction and compromise
data integrity.
The eight blower assemblies on the rear of the enclosure (see Figure
8–1) are an integral part of the enclosure and ensure that the
components remain cool. These blowers are arranged in two banks of
four blowers each. Air flow provided by these blowers is sufficient for
proper operation of the UltraSCSI subsystem.

UltraSCSI RAID Enclosure 8–1

Enclosure Blowers

Figure 8–1 Enclosure Blower Assembly Locations
Figure CALLOUTS
Bank A
A1. Blower 1

A2. Blower 2
A1

A3. Blower 3
B6

A4. Blower 4
A2

Bank B
B7

B5. Blower 5
A3

B6. Blower 6
B8

B7. Blower 7
A4

B8. Blower 8

CXO5494B

Enclosure temperature sensors (see Chapter 3) detect overtemperature
conditions or blower malfunctions and alert the user to the problem.
Simultaneously, the EMU causes all operational blowers to operate at
high-speed, increasing air flow through the enclosure to increase
cooling.
•

If an overtemperature condition is caused by either the intake air
temperature or exhaust air temperature (backplane) being too
high, the EMU initiates routines that can cause a power shut
down.

•

Removing a blower significantly changes the air flow pattern
within the enclosure. The EMU initiates a routine that causes a
power shut down to occur after approximately 8 minutes.

•

If two or more nonoperating blowers (that is, not operating or
rotating too slowly) are in the same blower bank (see Figure 8–1),
the EMU initiates a routine that causes a power shut down to
occur after approximately 8 minutes.

8–2 UltraSCSI RAID Enclosure

Enclosure Blowers

•

If a blower error condition is caused by only one defective blower
per blower bank (see Figure 8–1), the EMU does not initiate a
power shut down.

Function and Operation
Cooling enclosure components is accomplished by pulling air in
through the front of each component, over the internal circuitry,
through the backplane, and exhausting it out the rear of the enclosure.
The +12 V dc is required for blower operation and is available on
each of the backplane blower connectors. Blower status and speed
control signals are also on these connectors and are routed to the
EMU and PVA.
A blower guide pin ensures connector alignment when installing a
blower. A combination of blower connector and mounting tabs
provides positive mating of the blower with the blower plenum.
If an error condition involving the blowers, intake air temperature, or
exhaust air temperature exists, the EMU can cause all blowers to
switch into high-speed, increasing air flow through the enclosure.
Conditions that might cause this change in blower speed include:
•

Removing a blower

•

A blower operating too slowly

•

A stalled blower

•

The intake (EMU) air temperature is within 4°C (7°F) of the
sensor_1 setpoint. For example, if the setpoint is 35°C (95°F), the
blowers will switch to high-speed at 31°C (88°F).

•

The exhaust (backplane) air temperature is within 4°C (7°F) of
the sensor_1 or sensor_2 setpoint. For example, if the setpoint is
35°C (95°F), the blowers will switch to high-speed at
31°C (88°F).

UltraSCSI RAID Enclosure 8–3

Enclosure Blowers

Caution
The only StorageWorks dual-speed blower that is
compatible with this UltraSCSI RAID enclosure is model
DS–BA35X–MK. Do not use the BA35X–MD dual-speed
blower as a replacement.

Error Condition Reporting
The EMU LEDs display both blower and overtemperature error
conditions. See Chapter 3 for blower error conditions and
recommended corrective actions.

Replacing a Blower
When there is a malfunctioning blower, the remaining blowers
operate at high-speed to increase air flow through the
components—enough air to permit continued operation without
corrupting or losing data. However, Compaq recommends replacing a
defective blower immediately. If a blower is removed and not
replaced within approximately 8 minutes, the array controllers will
shut down and then the EMU will turn OFF the dc power distribution
to protect enclosure components.
Caution
Some array controllers bypass the standby power mode
functionality from the PVA standby power switch. Refer to
the array controller documentation for availability of this
feature.
Complete the procedures in Table 8–1 and Table 8–2 to replace a
blower. Blowers are hot swap components.

8–4 UltraSCSI RAID Enclosure

Enclosure Blowers

Caution
Operating an UltraSCSI enclosure with a blower removed
significantly changes the air flow pattern and reduces air
flow through the components. If a blower is removed and
not replaced within approximately 8 minutes, the array
controllers will shut down and then the EMU will turn OFF
the dc power distribution.

Table 8–1
1.

Removing a Blower

Use both hands to firmly
press the upper and lower
blower mounting tabs
(CALLOUTS 3 and 4)
together to release the
blower.

CXO5753B

Pull the defective blower straight out.

Place the blower into an electrostatic bag.

UltraSCSI RAID Enclosure 8–5

Enclosure Blowers

Table 8–2

Installing a Blower

Verify that the blower is a model DS–BA35X–MK.

Orient the replacement
blower to align its
connector (CALLOUT 2)
and guide pin (CALLOUT 1)
with the backplane blower
connector.

CXO5753B

Push the blower straight in, until the upper and lower mounting tabs
(CALLOUTS 3 and 4) snap into place.

If dc power is applied to the enclosure, verify that the blower starts
operating.

8–6 UltraSCSI RAID Enclosure

Appendix A
UltraSCSI Enclosure
Specifications
This appendix defines the physical, electrical, and environmental
specifications for a Compaq StorageWorks UltraSCSI enclosure with
24 disk drives and all its major electronic components. See Figure
A–1 and its associated CALLOUT numbers on the following page.
Figure A–1 UltraSCSI RAID Enclosure Major Components
Figure CALLOUTS

9
8x

HA–2450CU enclosure

Blowers (8 each)

I/O modules (6 each)

PVA (1 each)

AC input box (2 each)

Cache module (2 each)

Array controller (2 each)

EMU (1 each)

180 W power supply (8 each)

2 8x

3 6x

7 2x

5 2x

6 2x
CXO5555A

10. ECB (2 each)

UltraSCSI RAID Enclosure A–1

UltraSCSI Enclosure Specifications

Note
The operating and servicing physical specifications as well
as the electrical specifications for the UltraSCSI RAID
enclosure are cabinet- and configuration-specific.

Physical and Electrical Specifications
The following specifications apply to a single UltraSCSI RAID
enclosure (see Figure A–2) with all of the components installed.
Figure A–2 UltraSCSI Enclosure
Shipping Dimensions
Height
Width
Depth
Weight

Installed Dimensions

1156 mm
(45.5 in)
978 mm
(38.5 in)
737 mm
(29 in)
135 kg
(298 lb)

Height
Width
Depth
Weight

749 mm
(29.5 in)
435 mm
(17.125 in)
432 mm
(17 in)
102 kg
(225 lb)

Heat Dissipation
3070 BTUs/hr
Power
110–240 V ac, 50/60 Hz,
single phase, 12 A/6 A
CXO5797A

A–2 UltraSCSI RAID Enclosure

UltraSCSI Enclosure Specifications

Operating Environments
Table A–1, Table A–2, and Table A–3 provide the environmental
specifications for the controller shelf.
Table A–1 Optimum Operating Environment
Condition

Specification

Temperature

18–24°C (64–75°F) with an average rate of change of 3°C/h maximum and a
step change of 3°C/h or less

Relative humidity

40–60% (noncondensing) with a step change of 10% or less (noncondensing)

Altitude

Up to 2400 m (8000 ft)

Air quality

Not to exceed a maximum of 500,000 particles, 0.5 micron or larger, per cubic
foot of air

Nominal airflow

50 cubic ft/min

Heat dissipation

3070 BTUs/h

Table A–2 Minimum Operating Environment
Condition

Specification

Temperature

10–40°C (50–104°F)
Reduce rating by 1.8°C for each 1000 m altitude (1°F for each 1000 ft of
altitude)

Relative humidity

10–85% at a maximum wet bulb temperature of 32°C (90°F) and a minimum
dew point of 2°C (36°F)

Altitude

Up to 2400 m (8000 ft)

Air quality

Not to exceed a maximum of 500,000 particles, 0.5 micron or larger, per cubic
foot of air

Nominal airflow

40 cubic ft/min

Heat dissipation

3070 BTUs/h

UltraSCSI RAID Enclosure A–3

UltraSCSI Enclosure Specifications

Table A–3 Shipping Environment
Condition

Specification

Temperature

–40°C to +66°C (–40°F to +150°F)

Relative humidity

10–80% noncondensing

Altitude

4900 m (16,000 ft)

A–4 UltraSCSI RAID Enclosure

UltraSCSI Enclosure Specifications

SBB Thermal Stabilization
For proper operation upon initial power application, maintain the
Compaq SBB storage device temperature within the range of
18–29°C (65–85°F). See Table A–4 for condition statements.
Caution
Always thermally stabilize a disk drive SBB in the
operating environment before installing or operating it.
Otherwise, applying power can damage the disk drive.
Table A–4 SBB Thermal Stabilization
Condition

Then You Must…

When there is condensation on the outside of the
SBB…

Thermally stabilize the device and the SBB in
the operating environment for 6 hours or until
the condensation is no longer visible, whichever
is longer. Do not install a storage device until it
stabilizes.

When there is no condensation on the outside of
the SBB…

Thermally stabilize the device for the amount of
time specified in the chart below.

Storage Temperature Range

°C

°F

Stabilize for

°C

°F

Stabilize for

60 to 66

140 to 151

3 hours

0 to 9

32 to 48

1 hour

50 to 59

122 to 138

2 hours

–10 to –1

14 to 30

2 hours

40 to 49

104 to 120

1 hour

–20 to –11

–4 to 12

3 hours

30 to 39

86 to 102

30 minutes

–30 to –21

–22 to –6

4 hours

18 to 29

64 to 84

None

–40 to –31

–40 to –24

5 hours

10 to 17

50 to 63

30 minutes

UltraSCSI RAID Enclosure A–5

Appendix B
Assembling an UltraSCSI
RAID Subsystem
Assembling an UltraSCSI RAID subsystem (see Figure B–1) requires
sequentially installing the components in a DS–BA370 series
rack-mountable enclosure. The UltraSCSI RAID enclosure arrives as
either a departmental server cabinet (a pedestal with a factory
installed enclosure) or an empty enclosure that requires mounting
within a data center cabinet (SW600 series).
Figure B–1 UltraSCSI RAID Subsystem

CXO5797A

UltraSCSI RAID Enclosure B–1

Assembling an UltraSCSI RAID Subsystem

Installation Sequence
The basic installation sequence is the same for all UltraSCSI RAID
subsystems, whether they use one, two, or three DS–BA370
enclosures. To reduce the possibility of multiple interacting problems,
the preferred assembly sequence is as follows:
1. Assemble and test the master enclosure.
Note
Prior to power application and testing, a minimum
compliment of power components require installation, such
as the standard (n+1) configuration (see Chapter 6).
2. Assemble the first expansion enclosure.
3. Connect the first expansion enclosure to the master enclosure and
test the subsystem operation.
4. Assemble the second expansion enclosure.
5. Connect the second expansion enclosure to either the master
enclosure or the first expansion enclosure and test the subsystem
operation.
The assembly procedures for an enclosure are almost identical to
those described elsewhere in this publication. Therefore, rather than
repeating the procedures, the following sections primarily address
special assembly considerations. The recommended sequence for
assembling an UltraSCSI RAID enclosure is as follows:
WARNING!
A full enclosure weighs more than 102 kg (225 lb). To
prevent personal injury:

B–2 UltraSCSI RAID Enclosure

•

Always mount an empty enclosure into a cabinet
prior to installing devices.

•

Use at least two people to lift, align, and insert the
enclosure into a cabinet.

Assembling an UltraSCSI RAID Subsystem

1. Mount an empty UltraSCSI RAID enclosure in the data center
cabinet using cabinet documentation. Departmental server
cabinets arrive prepared for component installation.
2. Install the EMU and the EMU communication cables as required.
3. Install the PVA.
4. Install the disk drive SBBs.
5. Install the I/O modules and UltraSCSI bus cables as required.
6. Install the array controllers and cache modules.
7. Install the ECBs per array controller documentation.
8. Install the power system.
9. Connect subsystem cabling.
10. Configure the subsystem using configuration rules described in
Chapter 2.

Installing the EMU
Each UltraSCSI enclosure requires an EMU for proper operation. In
an expansion configuration, connect an EMU-EMU communication
cable between the master EMU and all expansion EMUs. Complete
the following procedure to install an EMU.
Complete the procedure in Table 3–10, steps 1 through 4, to install an
EMU.

Installing the PVA
Each UltraSCSI enclosure requires a PVA for proper operation.
Complete the procedure in Table 4–7, steps 1 through 4, to install a
PVA.

Installing the Disk Drive SBBs
Install the disk drive SBBs sequentially in the UltraSCSI enclosure
using the procedure in Table 7–4, steps 1, 2, and 8 through 11.

UltraSCSI RAID Enclosure B–3

Assembling an UltraSCSI RAID Subsystem

Installing the I/O Modules and Expansion Cables
Six single-ended I/O modules (DS–BA35X–MN) mount on the lower
rear enclosure shroud (see CALLOUT 1 in Figure B–2). Expansion
cables attach to these I/O modules for an expanded subsystem,
connecting one or two additional enclosures to the master enclosure.
Figure B–2 I/O Module Locations
Figure
CALLOUT
1.

2
4
6
5

SCSI bus
ports (1–6)
for I/O
modules
(1–6),
respectively

1
CXO5805B

For a single enclosure subsystem, complete the procedures in Table
5–10 to install I/O modules in the enclosure.
Caution
Not all array controllers support more than one enclosure.
Refer to the array controller documentation to verify that it
supports more than one enclosure.
In an expanded subsystem, refer to Table 5–3 for preliminary cabling
procedures. Then, complete the procedures in Table 5–6 to install a
two-enclosure subsystem or Table 5–8 to install a three-enclosure
subsystem.

B–4 UltraSCSI RAID Enclosure

Assembling an UltraSCSI RAID Subsystem

Installing Array Controllers and Cache Modules
The master enclosure contains the array controllers and cache
modules. ECBs cannot be installed within the enclosure. In a
departmental server cabinet, install the ECBs in the top cover. In a
data center cabinet (SW600 series), install the ECBs within an ECB
shelf mounted inside the cabinet.
Caution
Two sets of array controllers and cache modules can be
installed within an UltraSCSI RAID subsystem. By default,
the enclosure that contains these components becomes the
master enclosure and must always use PVA address 0.

Note
For detailed array controller, cache module, and ECB
installation procedures and cabling, refer to the array
controller documentation.
Figure B–3 identifies where the two controllers and two cache
modules are located.
Figure B–3 Array Controller and Cache Module Locations

CXO5798A

UltraSCSI RAID Enclosure B–5

Assembling an UltraSCSI RAID Subsystem

Table B–1 describes the mechanical procedures for installing array
controllers and cache modules. Following installation, configure the
subsystem using procedures contained in the following publications:
•

Array controller documentation

•

Operating system-specific release notes
Note
For single array controller and cache set configurations,
Controller A must be configured with Cache A or
Controller B must be configured with Cache B.

Table B–1 Installing Array Controllers and Cache
Modules
1.

Align the array controller
with the Controller A slot
(CALLOUT 3) in the
enclosure.

4A 19 20 21 22 23 24 4B

Shelf 4

3A 13 14 15 16 17 18 3B

Shelf 3

10 11 12 2B

Shelf 2

Shelf 1

3
A
4

Controller A

Cache A
CXO5813B

B–6 UltraSCSI RAID Enclosure

Assembling an UltraSCSI RAID Subsystem

Table B–1 Installing Array Controllers and Cache Modules
(Continued)
2.

Insert the array controller
into the Controller A slot
and push it in until the
extractor latches engage
the enclosure.

CXO5602A

Simultaneously push in both extractor latches to firmly seat the array
controller.

Align the cache module with the Cache A slot (CALLOUT 4 in step 1)
in the enclosure.

Insert the cache module
into the Cache A slot and
push it in until the
extractor latches engage
the enclosure.

CXO5602A

Simultaneously push in both extractor latches to firmly seat the cache
module.

UltraSCSI RAID Enclosure B–7

Assembling an UltraSCSI RAID Subsystem

Table B–1 Installing Array Controllers and Cache Modules
(Continued)
7.

If a second array controller and cache module set is required, repeat
steps 2 through 7 using Controller B and Cache B slots (CALLOUTS 3
and 4) shown below.
1

4A 19 20 21 22 23 24 4B

Shelf 4

3A 13 14 15 16 17 18 3B

Shelf 3

10 11 12 2B

Shelf 2

Shelf 1

Controller B
Cache B

4
CXO5812B

B–8 UltraSCSI RAID Enclosure

CXO5603A

Assembling an UltraSCSI RAID Subsystem

Installing the Power System
The ac input boxes and shelf power supplies comprise the enclosure
power system (see Figure B–4).
Figure B–4 Enclosure Power System

CXO5800A

Caution
To protect this sensitive electronic device from ESD, use the
following precautions:
• Wear an ESD wrist strap.
•

Do not touch the printed circuit board or the backplane
connector.

•

Do not lay the device on a work surface; place it on an
electrostatic mat.

•

Place the device in an electrostatic bag for shipment.

To prevent ESD damage to a power supply SBB, do not
touch the SBB connector.
The sequence for installing the power system is as follows:
1. Install the shelf power supplies using Table 6–4, steps 1
through 4.
2. Install the ac input boxes using Table 6–6, steps 1 through 7.

UltraSCSI RAID Enclosure B–9

Assembling an UltraSCSI RAID Subsystem

Connecting Subsystem Cables
Cabinet user documentation describes the cable routing, location, and
function. The cables involved might include:
•

Controller-to-controller

•

Controller-to-host

•

ECB-to-cache module

For a detailed description of the cable connection procedures and
cable routing, refer to the following publications:
•

Array controller documentation

•

DIGITAL StorageWorks UltraSCSI RAID Cabinet Subsystem
(DS–SW600 Series) Installation and User’s Guide

•

Pedestal documentation

Configuring the Subsystem
Refer to Chapter 2 for configuring the subsystem. Apply
configuration rules that pertain to the UltraSCSI subsystem being
assembled.
Turn ON the subsystem using Table 2–2.

B–10 UltraSCSI RAID Enclosure

Appendix C
Installing EMU Microcode
The microcode resident in the EMU memory controls all the
UltraSCSI RAID enclosure (DS–BA370 series) EMU operations.
Proper operation of the subsystem requires that all EMUs have the
same revision level microcode. Upon system startup, the master EMU
polls the expansion EMUs to ensure that they have the correct
microcode. When microcodes are a different revision level, the master
EMU automatically downloads the latest microcode revision over the
EMU communications bus.

Microcode Upgrades
Microcode upgrades can be downloaded from a personal computer
(PC) to the master EMU. To download the microcode, the PC is
connected to the center UART connector and a communications
program, such as PROCOMM PLUS®, is used.
Note
If PROCOMM PLUS is not used, the prompts, displays,
and responses might be different from those presented in
this document. Refer to the communications program user
documentation for the proper procedures and results.

UltraSCSI RAID Enclosure C–1

Installing EMU Microcode

Prerequisites
The following items are required to upgrade the EMU microcode
using the EMU UART connector:
•

EMU microcode file (for example, FLSH_011.HEX)

•

Communications program, such as PROCOMM PLUS

•

Compaq adapter cable, part number 17–04697–01, with the
following connectors:
— 9-pin RS232
— 8-pin RJ45

Preliminary Procedure
Complete the procedure in Table C–1 prior to starting the installation
procedure.
Table C–1 Preliminary Procedures
1.

Shut down the software for each controller and eject its personal
computer memory card industry association (PCMCIA) card.

Connect the 9-pin RS232 connector to a PC COM port connector.

Connect the 8-pin RJ45 connector to the master EMU center UART
connector.

Configure the communications software parameters as follows:
Parameter
Baud rate
Data bits
Parity
Stop bits
Protocol
CR/LR conversion
XON/XOFF
Connection
Flow control

C–2 UltraSCSI RAID Enclosure

Value
38,400 Baud
8
None
1
Raw ASCII
None
None
Direct (no modem)
None

Installing EMU Microcode

Table C–1 Preliminary Procedures (Continued)
5.

Use the PROCOMM PLUS Setup Utility to establish the following
parameters for modem and protocol options:
Modem Options
General Options
Auto baud detect
Send init if CD high
Protocol Options
General Options
Abort xfer if CD lost
ASCII Protocol Options
Character pacing (millisec)
Line pacing (1/10 second)
Pace character
Strip 8th bit
CR translation (upload)
LF translation (upload)

OFF
NO

NO
0
0
0
NO
NONE
NONE

Insert the microcode disk into the disk drive. Copy the EMU
microcode file into a PC directory (for example,
C:\FLSH_011.HEX).

Start the communications program.

UltraSCSI RAID Enclosure C–3

Installing EMU Microcode

Preparing the EMU
Complete the procedure in Table C–2 to prepare the EMU for loading
the microcode.
Table C–2 Preparing the EMU
1.

On the master enclosure, press and hold-in the alarm control switch
while removing the EMU from the enclosure.
The PVA audible alarm sounds, indicating that there is no EMU in
the enclosure, and continues to sound until the EMU is reinserted.

While still holding-in alarm control switch, insert the EMU into the
enclosure and firmly seat it.
The PVA audible alarm turns OFF. The temperature (amber), power
(green), and eight blower (amber) LEDs will be ON.

Continue holding in the alarm control switch until:
•

The blower LEDs are OFF.

•

The PC monitor displays the following message:
Waiting for receiver ready

Release the alarm control switch.

Loading the Microcode
From the PC, initiate the upload function by completing the
procedure in Table C–3.
Table C–3 Loading the Microcode
1.

When using PROCOMM PLUS, press the Send File Key.

Select Raw ASCII from the protocol list.

Enter the microcode path and filename (for example,
C:\FLSH_011.HEX).

When the EMU microcode download starts, all blower LEDs are ON
and the PC displays a series of messages similar to those shown in
Table C–4.

C–4 UltraSCSI RAID Enclosure

Installing EMU Microcode

Table C–4 Typical PC Monitor Display
Display Text

Comments

EMU successfully completed
boot.

+++ Attempting to load FLASH image from UART +++
Download Parameters
Raw ASCII, 38400 Baud, 8 data bits, no stop
bits

Communications program
parameters.

Waiting for receiver ready

All the EMU blower LEDs are
ON. Ready to start transfer.

No visible activity for 30 to 60 seconds
+++ Loaded image copied to Upper FLASH +++

Communications program has
uploaded microcode to EMU.

+++ Copying Upper FLASH image to Lower FLASH +++

EMU transferring new
microcode.

+++ Lower FLASH is valid +++
+++ Upper FLASH is valid +++

New microcode verification
complete.

EMU initializing.

This is CAB 000
Bus 0 or 1 I/O mod changed, old 0000, new 0002
Bus 2 or 3 I/O mod changed, old 0000, new 0002
Bus 4 or 5 I/O mod changed, old 0000, new 0002
Power OK changed, old 0000, new 0002
Error sum changed, old 0000, new 0002
SubStat 0 changed, old 0000, new 0002
Supplies present change, old 0000, new 0002
All B fans present
All A fans present
Fan OK changed, old 0000, new 0002
Err sum changed, old 0000, new 0002
Power OK changed, old 0000, new 0002
Power OK changed, old 0000, new 0002

Sample EMU status report.

UltraSCSI RAID Enclosure C–5

Installing EMU Microcode

Restarting the Controller
Complete the procedure in Table C–5 to restart the controller after the
microcode upgrade is loaded.
Table C–5 Restarting the Controller
1.

Remove the 8-pin RJ45 connector from the EMU UART connector.

On each controller, press and hold-in the reset button and reseat its
PCMCIA card.

Release the reset button.

Verify that each array controller reinitializes.

C–6 UltraSCSI RAID Enclosure

Glossary
This glossary defines terms pertaining to the DS-BA370 series
UltraSCSI RAID enclosure. It is not a comprehensive glossary for all
StorageWorks products.
ac distribution
The method of distributing ac power within a StorageWorks shelf,
enclosure, or cabinet.
ac input box
A device that receives ac power from the PDU and distributes the ac
power to all shelf power supplies connected to the ac input box power
bus.
ac power controller
See ac input box.
active
The disk drive SBB is spinning up, or down, or transferring data to or
from the array controller.
adapter
See SCSI bus signal converter.

UltraSCSI RAID Enclosure Glossary–1

Glossary

ambient air temperature
The temperature of the air surrounding the cabinet, enclosure, shelf,
or SBB that dissipates the heat generated by the device.
See also intake air temperature.
array controller
See controller.
backplane
The electronic printed circuit board mounted in the rear of the shelf.
This board contains the disk drive SBB, power supply, and
terminator connectors.
blowers
An airflow device mounted in a StorageWorks shelf.
bus expander
Devices that couple bus segments together without any impact on the
SCSI protocol, or the software. These devices include both
single-ended to differential and single-ended to single-ended bus
extenders. The term expander is a general term that includes
“extender,” “repeater,” and “isolator.”
bus extender
See bus expander.
bus segment
A SCSI bus segment consists of all the conductors and connectors
required to attain signal line continuity between every driver, receiver,
and two terminators for each signal. It is not necessary that a SCSI
bus segment contain any initiators or targets, but it must have at least
two devices attached. (Drivers and receivers can be part of extenders
as well as part initiators and targets.)
Bus segments can be either single-ended or differential. The
terminator properties determine the bus segment type. Devices that do
not have the same transceiver type as the terminators cannot operate
in the segment defined by the terminators.

Glossary–2 UltraSCSI RAID Enclosure

Glossary
Caution
The device transceiver type (single-ended or differential)
must match the bus segment type. Devices that do not meet
this condition cannot operate.
The allowed length of a bus segment depends on the electrical
loading, transmission media type, and data transfer rate. In many
cases, heavier loading, smaller wires, and higher speeds demand
shorter lengths. Increasing the number of devices on a given length of
the bus, by using longer stubs or higher capacitance devices, produces
loading.
cabinet
See data center cabinet and departmental server cabinet.
cache module
A fast storage buffer.
Callout n
A textual reference to a numbered callout in a figure.
CE–Mark
A European Economic Community (EEC) certification label that
identifies electronic devices authorized for sale within member
nations.
CE–Mark Class A
Similar to, but more stringent than the FCC Class A certification, this
certification label appears on electronic devices that are for use in a
commercial environment. A CE–Mark certified device can also be
used in the United States.
CE–Mark Class B
Similar to, but more stringent than the FCC Class B certification, this
certification label appears on electronic devices that are for use in
either a home or a commercial environment. A CE–Mark certified
device can also be used in the United States.

UltraSCSI RAID Enclosure Glossary–3

Glossary

certified device
A storage device tested in a specific configuration and found to be in
compliance with either an FCC or a CE certification standard.
Compaq certifies these devices to operate in a specific shelf,
enclosure, or cabinet.
channel
Another term for a SCSI bus.
CLI
Command line interpreter. The operator interface to HSx series array
controller software.
cold swap
A method of device replacement that requires that power be removed
from one or more shelves in a cabinet thereby affecting other devices.
Use this method during initial installation or StorageWorks subsystem
upgrades, or when conditions preclude using either the “warm swap”
or “hot swap” method.
See also warm swap and hot swap.
command line interpreter
See CLI.
connector
Any connector that is physically part of a cable assembly attached to
backplanes or other nondevice connectors.
controller
(1) A hardware-software device that manages communications on
behalf of host systems over the SCSI bus to devices. Controllers
typically differ by the type of interface to the host and provide
functions beyond those the devices support. (2) A standalone device
that connects a host adapter to the storage SCSI bus. This device
provides RAID functionality, typically has multiple SCSI bus ports,
performs the lower layers of the SCSI protocol, and normally
operates in the initiator role.
See also array controller.
Glossary–4 UltraSCSI RAID Enclosure

Glossary
converter
See SCSI bus signal converter.
data center cabinet
The largest of the StorageWorks cabinets, such as the SW800 series
that can contain as many as 24 drive SBB shelves. These cabinets
include either a 50 Hz or 60 Hz cable distribution unit (CDU),
internal ac power cords, and cooling fans.
departmental server cabinet
The series of smaller StorageWorks cabinets for mounting shelves,
controllers, storage devices, and power control devices to form
subsystems.
device
The targets, initiators, hubs, converters, and bus expanders, and
similar devices interconnected to form a SCSI bus. Connectors,
expanders, and hubs do not use a SCSI bus ID.
See also nodes.
device connector
Any connector physically part of a SCSI device.
differential SCSI bus
A bus in which the voltage potential difference between two wires
determines the signal level.
disk drive
A storage device supporting random access to fixed size blocks of
data.
dual redundant configuration
An array controller configuration consisting of controller A and
controller B mounted in the same controller shelf. When controller A
fails, controller B assumes control over the devices.

UltraSCSI RAID Enclosure Glossary–5

Glossary

ECB
External cache battery. A device that provides backup power to a
cache module for sustaining memory retention.
electromagnetic interference
See EMI.
electrostatic discharge
See ESD.
EMC
Electromagnetic compatibility.
EMI
Electromagnetic interference. The impairment of a signal by an
electromagnetic disturbance.
EMU
Environmental monitoring unit. The device that monitors the status of
the UltraSCSI RAID enclosure to include power, intake air
temperature, blower status, and so forth, detects error and fault
conditions, displays these conditions, reports the conditions to the
user and the array controller, and, in some cases, implements
corrective actions.
enclosure connector
Any connector that is physically part of an enclosure (for example,
pedestal, deskside enclosure, cabinet, and so forth).
environmental monitoring unit
See EMU.
ESD
Electrostatic discharge. The discharge of a potentially harmful static
electric voltage as a result of improper grounding.

Glossary–6 UltraSCSI RAID Enclosure

Glossary
expansion enclosure
An UltraSCSI RAID enclosure connected to the master enclosure
with UltraSCSI bus cables. Each subsystem can have a maximum of
two expansion enclosures. Set the expansion PVA SCSI bus address
switch to either:
2 — Establishing SCSI bus device addresses 8, 9, 10, and 11
3 — Establishing SCSI bus device addresses 12, 13, 14, and 15
external cache battery
See ECB.
Fast 10
An improvement in SCSI technology from SCSI-1 to SCSI-2
(Fast 10). Fast 10 improvements made over SCSI-1 include the
following:
•

Maximum transfer rate increase from 5 Mb/s to 10 Mb/s

•

A maximum wide transfer rate of 20 MB/s

Fast 20
See UltraSCSI.
FCC
Federal Communications Commission. The federal agency
responsible for establishing standards and approving electronic
devices within the United States.
FCC Class A
This certification label appears on electronic devices for use only in a
commercial environment within the United States. A CE–Mark
certified device can be used in the United States in the same
environment as the equivalent FCC certification.
FCC Class B
This certification label appears on electronics for use in either a home
or a commercial environment within the United States. A CE–Mark
certified device can be used in the United States in the same
environment as the equivalent FCC certification.

UltraSCSI RAID Enclosure Glossary–7

Glossary

Federal Communications Commission
See FCC.
filler panel
A panel used to cover open unused areas in cabinets or shelves for
either EMI suppression, air flow control, or cosmetic purposes.
host
The primary or controlling computer (in a multiple computer
network) to which storage is attached.
host adapter
A device that connects the host system I/O bus (for example, a PCI
bus) to the storage SCSI bus. A host adapter performs the lower layers
of the SCSI protocol and normally operates in the initiator role.
host computer
See host.
host controller
A device that connects the host system I/O bus (for example, a PCI
bus) to the storage SCSI bus. A host controller provides RAID
functionality, typically has multiple SCSI bus ports, performs the
lower layers of the SCSI protocol, and normally operates in the
initiator role.
host port adapter
A host controller device that adapts the host system I/O bus to the
array controller SCSI bus configuration.
hot swap
A method of device replacement whereby the complete system
remains online and active during device removal or insertion. The
device being removed or inserted is the only device that cannot
perform operations during this process.
See also cold swap and warm swap.

Glossary–8 UltraSCSI RAID Enclosure

Glossary
HSOF
Hierarchical Storage Operating Firmware. HS-family controller
software contained on a removable ROM card (PCMCIA).
ID
Identifier—as used in SCSI bus ID, target ID or device ID.
initiator
A SCSI device (usually a host system) that requests another SCSI
device (a target) to complete an operation.
I/O module
Input/output module. A SCSI enclosure or drive SBB shelf device
that integrates a single-ended SCSI bus with either an 8-bit
single-ended, 16-bit single-ended, or 16-bit differential SCSI bus.
input/output module
See I/O module.
intake air temperature
The temperature of the air entering the cabinet, enclosure, shelf, or
SBB.
See also ambient air temperature.
LED
Light emitting diode. A source of light on an indicator panel.
logical bus
A single-ended, physical bus connected to a differential physical bus
by a SCSI bus signal converter.
maintenance terminal
The operator terminal used to identify an HSOF family controller, to
enable its host paths, to define its subsystem configuration, and to
check its status. The HS-family maintenance terminal accepts any
terminal conforming to EIA–423. Use the maintenance terminal to
configure a storage subsystem.

UltraSCSI RAID Enclosure Glossary–9

Glossary

master enclosure
The UltraSCSI RAID enclosure that includes the HSZ series
controllers and cache modules. There is only one master enclosure in
each UltraSCSI RAID subsystem. The SCSI bus address switch
setting is always 0. Using address 0 automatically assigns SCSI bus
device addresses 0, 1, 2, and 3 to the devices.
Mb/s
Megabits per second. The bus speed at which the number of bits are
transferred.
MB/s
Megabytes per second. The bus width (8- or 16-bit), the number of
bytes per word (1 or 2, respectively), and the bus clock frequency
determines the transfer rate in MB/s.
megabits per second
See Mb/s.
megabytes per second
See MB/s.
node
A SCSI bus target or initiator that uses a SCSI bus ID. For example,
disk drives, tape drives, array controllers, and adapters all have a
SCSI bus ID and are nodes. Hubs, expanders, and converters are
devices and do not have a SCSI bus ID. An UltraSCSI bus can have a
maximum of 16 nodes (SCSI bus addresses 0 through 15). A narrow
bus can have a maximum of eight nodes (SCSI bus addresses 0
through 7).
overtemperature
A condition where a temperature is above a specified limit and can
cause equipment failure if not corrected.
pedestal
The common name for a deskside expansion unit.
See departmental server cabinet.

Glossary–10 UltraSCSI RAID Enclosure

Glossary
physical bus
Two SCSI terminators separated by cables, connectors, and backplane
circuitry.
port
(1) A logical route for data in and out of an array controller. A port
can contain one or more channels, all of which contain the same type
data. (2) The hardware and software that connects a host controller to
a Computer Interconnect (CI™), SCSI, or DIGITAL® standard
system interconnect (DSSI™) bus.
power verification and addressing assembly
See PVA.
PVA
Power verification and addressing assembly. The module whose
primary functions are (1) to allow the user to select the enclosure
UltraSCSI bus ID, (2) to enable the user to place the subsystem in a
standby power mode and return it to an operational status, (3) in
conjunction with the associated EMU, ensures that critical power
functions are monitored.
quiesce
To make a bus inactive or dormant. For example, the user must
quiesce SCSI bus operations when warm swapping a disk drive SBB.
radio frequency interference
See RFI.
RAID
Redundant array of independent disks. A set of storage techniques
devised to increase the performance and availability of a storage
subsystem.

UltraSCSI RAID Enclosure Glossary–11

Glossary

redundant power supply configuration
A capability of StorageWorks cabinet and shelves to ensure that there
is no single point of power failure. (1) For a cabinet, two ac power
sources and two distribution units or ac input boxes distribute primary
and redundant ac power to shelf power supplies. (2) For a shelf, the
primary and redundant shelf power supplies ensure that the dc power
is available even when there is a failure of one supply, one ac source,
or one distribution unit or ac input box.
RFI
Radio frequency interference. The impairment of a signal by an
unwanted radio signal or radio disturbance.
SBB
StorageWorks building block. The basic building block of the product
line. Any device conforming to shelf mechanical and electrical
standards installed in either a 3.5-inch or 5.25-inch carrier is
considered to be an SBB, whether it be a storage device, a power
supply, or other device.
SCSI
Small computer system interface. This ANSI interface defines the
physical and electrical parameters of a parallel I/O bus used to
connect computers and a maximum of seven devices.
SCSI bus ID
The bit-significant representation of the SCSI addressing referring to
one of the signal lines numbered 0 through 7 for an 8-bit bus or 0
through 15 for a 16-bit bus. Also known as SCSI device ID and target
ID.
SCSI bus signal converter
(1) A connecting device that permits the attachment of accessories or
provides the capability to mount or link units. (2) The device that
connects a differential SCSI bus to a single-ended SCSI bus. (3) The
device that extends the length of a differential or single-ended SCSI
bus.

Glossary–12 UltraSCSI RAID Enclosure

Glossary
SCSI device
A host computer adapter, a peripheral array controller or an intelligent
peripheral that can be attached to the SCSI bus.
SCSI domain
A SCSI domain is a logical bus consisting of at least one bus
segment, at least one initiator, and at least one target. Domains with
multiple bus segments are enabled through the use of bus extenders.
Domains are limited by device addressability. Domains are limited to
a maximum of 16 initiators and targets without the use of LUN
bridges.
SCSI ID
See SCSI bus ID.
SCSI port
(1) Software. The channel that controls communications to and from
a specific SCSI bus in the system. (2) Hardware. The name of the
logical socket at the back of the system unit to which a SCSI device is
connected.
shelf power supply
A power supply that provides +5 V dc and +12 V dc to the
StorageWorks shelves. These power supplies convert either an ac
input (120 V ac) or dc (48 V dc) voltage into the required output
voltages.
single-ended SCSI bus
A bus in which the voltage of a single wire in relation to ground
determines each signal’s logic level.
small computer system interface
See SCSI.
spin-down
A process that begins when power is removed from a storage device
and data transfer halts, and ends when the device is stopped and can
be moved. In the case of disk drives, the heads are retracted and the
media is stopped.

UltraSCSI RAID Enclosure Glossary–13

Glossary

See also spin-up.
spin-up
A process that begins when power is applied to a storage device, and
ends when the device is determined to be operational and ready for
data transfer operations.
See also spin-down.
standard power configuration
A power configuration in which there is only one ac power source and
cabinet distribution unit, power distribution unit, or ac input box to
condition the input ac power.
STERMPOWER
Terminator power on a single-ended SCSI bus.
See TERMPOWER.
storage array
An integrated set of storage devices.
See also storage sets.
storage sets
A grouping of disk drives that make up a new distinct container.
See also storage array.
storage subsystem
The array controllers, storage devices, shelves, cables, and power
supplies that form a mass storage subsystem.
StorageWorks
The set of enclosure products that allows customers to design and
configure their storage subsystem. Components include power,
packaging, and interconnections in a StorageWorks shelf. SBBs and
array controllers are integrated therein to form storage subsystems.
System-level enclosures to house the shelves and standard mounting
devices for SBBs are also included.

Glossary–14 UltraSCSI RAID Enclosure

Glossary
StorageWorks building block
See SBB.
supported device
(1) A device that has been fully evaluated in an “approved”
StorageWorks configuration (that is, shelf, cabinet, power supply,
cabling, and so forth) and is in complete compliance with
country-specific standards (for example, FCC, CE–Mark, CSA, TÜV,
VDE, and so forth) and with all Compaq standards. (2) A device
supported by an array controller or host operating system.
target
A SCSI device that performs an operation requested by an initiator.
target ID
See SCSI bus ID.
terminators
The interconnect components that form the ends of the transmission
lines in bus segments. A SCSI domain must have at least one
segment and therefore at least two terminators. The terminators
ensure that inactive SCSI bus signals are in a known state.
There are two basic types of terminators—active and passive:
•

Single-ended bus segments use active, linear terminators

•

Differential bus segments use passive (linear totem pole)
terminators except for special cases where the electrical
transmission lines are very short and only one termination or
pull-up is required.

terminator connector
Any connector physically part of a terminator. It is not uncommon
for terminators to have both stub and bus path connectors.
TERMPOWER
Termination power. The electrical current power required for SCSI
bus terminators. This power can be supplied by an external SCSI bus,
the shelf power supply, or an ac-to-dc power converter.

UltraSCSI RAID Enclosure Glossary–15

Glossary

See also STERMPOWER.
topology
The physical arrangement and relationship of interconnected nodes
and SCSI buses in a network. A legal topology must satisfy all the
requirements of the associated SCSI bus (Fast 10, UltraSCSI, and so
forth).
trilink connector
A 3-way SCSI connector.
UltraSCSI
An improvement in SCSI technology invented in 1993 by the Digital
Equipment Corporation (now Compaq Computer Corporation)
StorageWorks Engineering Group. Subsequently, the ANSI SCSI
standards committee issued standard X3T10 for UltraSCSI.
The UltraSCSI improvements over Fast SCSI include the following:
•

Maximum transfer rate increases from 10 Mb/s to 20 Mb/s

•

Maximum wide bus bandwidth increases from 20 MB/s to 40
MB/s

•

VHDCI cables and connectors are significantly thinner and
smaller

See also Fast 20.
UltraSCSI domain
A single, logical UltraSCSI bus composed of multiple diverse bus
segments.
See also bus segment.
UltraSCSI RAID enclosure
A 24-drive SBB RAID shelf, such as the DS–BA370 series.

Glossary–16 UltraSCSI RAID Enclosure

Glossary
UltraSCSI RAID subsystem
One or more UltraSCSI RAID enclosures mounted in either a
pedestal or a cabinet. Each subsystem requires a master enclosure.
Adding one or two expansion enclosures to any single-enclosure
subsystem creates an expanded subsystem.
VHDCI
Very high density cable interconnect. A 68-pin interface cable with
connectors on 0.8 mm centers. Required for UltraSCSI operation.
warm swap
A method of device replacement whereby the complete system
remains online during device removal or insertion. During device
insertion or removal, the bus might halt for a brief period of time.
System booting or code loading cannot occur until insertion of the
replacement device. There is no noticeable impact on user
applications that are not dependent upon the devices on the affected
SCSI bus.
See also cold swap and hot swap.

UltraSCSI RAID Enclosure Glossary–17