Brocade, the B-wing symbol, BigIron, DCX, Fabric OS, FastIron, NetIron, SAN Health, ServerIron, and TurboIron are registered
trademarks, and Brocade Assurance, Brocade NET Health, Brocade One, CloudPlex, MLX, VCS, VDX, and When the Mission Is
Critical, the Network Is Brocade are trademarks of Brocade Communications Systems, Inc., in the United States and/or in other
countries. Other brands, products, or service names mentioned are or may be trademarks or service marks of their respective
owners.
Notice: This document is for informational purposes only and does not set forth any warranty, expressed or implied, concerning
any equipment, equipment feature, or service offered or to be offered by Brocade. Brocade reserves the right to make changes to
this document at any time, without notice, and assumes no responsibility for its use. This informational document describes
features that may not be currently available. Contact a Brocade sales office for information on feature and product availability.
Export of technical data contained in this document may require an export license from the United States government.
The product described by this document may contain “open source” software covered by the GNU General Public License or other
open source license agreements. To find out which open source software is included in Brocade products, view the licensing
terms applicable to the open source software, and obtain a copy of the programming source code, please visit
http://www.brocade.com/support/oscd.
Brocade Communications Systems, Incorporated
Corporate and Latin American Headquarters
Brocade Communications Systems, Inc.
130 Holger Way
San Jose, CA 95134
Tel: 1-408-333-8000
Fax: 1-408-333-8101
E-mail: info@brocade.com
European Headquarters
Brocade Communications Switzerland Sàrl
Centre Swissair
Tour B - 4ème étage
29, Route de l'Aéroport
Case Postale 105
CH-1215 Genève 15
Switzerland
Tel: +41 22 799 5640
Fax: +41 22 799 5641
E-mail: emea-info@brocade.com
Asia-Pacific Headquarters
Brocade Communications Systems China HK, Ltd.
No. 1 Guanghua Road
Chao Yang District
Units 2718 and 2818
Beijing 100020, China
Tel: +8610 6588 8888
Fax: +8610 6588 9999
E-mail: china-info@brocade.com
Asia-Pacific Headquarters
Brocade Communications Systems Co., Ltd. (Shenzhen WFOE)
Citic Plaza
No. 233 Tian He Road North
Unit 1308 – 13th Floor
Guangzhou, China
Tel: +8620 3891 2000
Fax: +8620 3891 2111
E-mail: china-info@brocade.com
• Chapter 2, “Administering FICON Fabrics” allows you to set up and get started using FICON
quickly, if you are already familiar with FICON concepts and basic procedures.
• Chapter 3, “Configuring FICON CUP” provides basic conceptual information on FICON CUP and
instructions on how to set CUP up in your FICON environment.
• Chapter 4, “Administering FICON Extension Services” provides basic conceptual information on
emulation products and how to set them up in a FICON environment.
• Chapter 5, “Maintaining and Troubleshooting FICON” discusses maintaining your switch in a
FICON environment, provides symptoms, troubleshooting tips, and possible solutions to known
issues.
• Appendix A, “Platforms supporting FICON” lists Brocade switches and Backbones supported
for FICON for various IBM systems. Also listed are Brocade switch blades supported for FICON
on Backbone platforms
• Appendix B, “Basic Switch Configuration” provides basic steps and commands to quickly
configure a switch for fabric and possible FICON and cascaded FICON operation.
• Appendix C, “Address Binding Examples” provides examples and scripts with appropriate
commands for the following:
-“Sequential address binding”
-“Zero-Based to Port-Based Address Binding”
-“Unbinding Multiple Ports”
FICON Administrator’s Guidevii
53-1002753-01
Page 8
• Appendix D, “Configuration Information Record” provides a form to record your FICON
configuration information.
• Appendix E, “EBCDIC Code Page” provides a table of the EBCDIC Code Page 37.
Supported hardware and software
Although many different software and hardware configurations are tested and supported by
Brocade Communications Systems, Inc. for Fabric OS v7.1,0 documenting all possible
configurations and scenarios is beyond the scope of this document.
For a complete list of platforms supported by FICON and Fabric OS v7.1.0, refer to Appendix A,
“Platforms supporting FICON”.
In cases where procedures or parts of procedures do not apply to all Brocade hardware platforms,
this guide identifies which platforms are supported.
What’s new in this document
The following information was added or changed in this document:
• Throughout manual
-Removed mention and details of FR4-18i blade since this is not supported for FCIP this
release.
• Preface.
-Updated “What’s new in this document” section to include all changes incorporated in this
manual.
• Chapter 1, “Introducing FICON.”
-Added new commands to Table 1, “Fabric OS commands related to FICON.”
-Removed “Platforms supporting FICON” and added “Platforms Supporting FICON”
appendix.
-Added information on Fabric OS routing policies supported by FICON in the “Fabric OS
support for FICON” section.
-Fabric OS support for FICON
• Removed Device connection control (DCC) policy
• Chapter 2, Administering FICON Fabrics
-Added step under “Configuring switched point-to-point FICON” to configure the fill word for
8 Gbps ports.
-Added “Automating CS_CTL Mapping” section.
-In “FICON and FICON CUP in virtual fabrics section,” added the following:
• Details from “Configuring FICON CUP in Virtual Fabrics” in Chapter 3 to this section
since having same subject matter in two different chapters is confusing.
• FMS mode cannot be enabled when a logical switch is using 10-bit addressing mode.
• Port swap limitations section.
viiiFICON Administrator’s Guide
53-1002753-01
Page 9
• Removed restriction that XISLs cannot be configured.
• Added instructions and restrictions on enabling XISL use and referenced Fabric OS
Command Reference for details.
-Added section on FICON Best Practices.
-Added section on FICON link address and FC address.
• Chapter 3, “Configuring FICON CUP.”
-Under “FMS Mode and FICON CUP” section, added the following:
• New section called “Fabric OS command limitations and considerations” to collect
differences in using Fabric OS commands when FMS mode is enabled.
• Noted that FMS mode cannot be enabled when a logical switch is using 10-bit
addressing mode.
• Section on port swapping limitations.
-Moved information that was in “Configuring FICON CUP in Virtual Fabrics” section to
“FICON and FICON CUP in Virtual Fabrics” section of Chapter 2 because having sections in
two different chapters dealing with same subject is confusing. Left heading in the chapter
but referenced section in chapter 2.
-Under “Port and switch naming standards for FMS mode,” added restrictions for using
unique CUP port name.
• Chapter 5, Maintaining and Troubleshooting FICON
-Under “Configuration restoration in a FICON environment” section, fixed ASM bit “on” row
in “Backup and restore in a FICON CUP environment” table, to say that “The IPL file is not
replaced because Active=Saved mode is on.” This did say Active=Saved mode is “off.”
• Added Appendix on Brocade and IBM platform support for FICON, Appendix on basic switch
configuration, and Appendix containing switch binding examples and scripts.
Document conventions
This section describes text formatting conventions and important notice formats used in this
document.
Text formatting
The narrative-text formatting conventions that are used are as follows:
bold textIdentifies command names
Identifies the names of user-manipulated GUI elements
Identifies keywords and operands
Identifies text to enter at the GUI or CLI
italic textProvides emphasis
Identifies variables
Identifies paths and Internet addresses
Identifies document titles
code textIdentifies CLI output
Identifies command syntax examples
FICON Administrator’s Guideix
53-1002753-01
Page 10
For readability, command names in the narrative portions of this guide are presented in mixed
NOTE
ATTENTION
lettercase: for example, switchShow. In actual examples, command lettercase is often all
lowercase. Otherwise, this manual specifically notes those cases in which a command is case
sensitive.
Command syntax conventions
Command syntax in this manual follows these conventions:
commandCommands are printed in boldface.
--option, optionCommand options are printed in bold.
-argument, argArguments.
[ ]Optional element.
variableVariables are printed in italics. In the help pages, values are underlined
enclosed in angled brackets < >.
...Repeat the previous element, for example “member[;member...]”
valueFixed values following arguments are printed in plain font. For example,
--show WWN
|Boolean. Elements are exclusive. Example:
--show -mode egress | ingress
or
Command examples
This guide describes how to perform configuration tasks using the Fabric OS command line
interface, but does not describe the commands in detail. For complete descriptions of all Fabric OS
commands, including syntax, operand descriptions, and sample output, refer to the Brocade Fabric OS Command Reference Manual.
Notes, cautions, and warnings
The following notices and statements are used in this manual. They are listed below in order of
increasing severity of potential hazards.
A note provides a tip, guidance, or advice, emphasizes important information, or provides a
reference to related information.
An Attention statement indicates potential damage to hardware or data.
xFICON Administrator’s Guide
53-1002753-01
Page 11
CAUTION
A Caution statement alerts you to situations that can be potentially hazardous to you.
DANGER
A Danger statement indicates conditions or situations that can be potentially lethal or extremely
hazardous to you. Safety labels are also attached directly to products to warn of these conditions
or situations.
Key terms
For definitions specific to Brocade and Fibre Channel, see the technical glossaries on MyBrocade.
See “Brocade resources,” for instructions on accessing MyBrocade.
For definitions of SAN-specific terms, visit the Storage Networking Industry Association online
dictionary at:
http://www.snia.org/education/dictionary
Notice to the reader
This document may contain references to the trademarks of the following corporations. These
trademarks are the properties of their respective companies and corporations.
These references are made for informational purposes only.
CorporationReferenced Trademarks and Products
International Business Machines Corp.IBM, FICON, S/390, z/OS, zSeries, Series z, Redbooks, z/OS
Teradata CorporationTeradata
Optica Technologies, Inc.Prizm FICON to ESCON converter and ESBT Bus/Tag Interface
Additional information
This section lists additional Brocade and industry-specific documentation that you might find
helpful.
Brocade resources
To get up-to-the-minute information, go to http://my.brocade.com and register at no cost for a user
ID and password.
Global Mirror, System z, zEnterprise
Module for Prizm
FICON Administrator’s Guidexi
53-1002753-01
Page 12
For additional Brocade documentation, visit the Brocade SAN Info Center and click the Resource
Library location:
http://www.brocade.com
Release notes are available on the MyBrocade website and are also bundled with the Fabric OS
firmware.
Other industry resources
White papers, online demonstrations, and data sheets are available through the Brocade
website at http://www.brocade.com/products-solutions/products/index.page
• Best practice guides, white papers, data sheets, and other documentation is available through
the Brocade Partner website.
For additional resource information, visit the Technical Committee T11 website. This website
provides interface standards for high-performance and mass storage applications for Fibre
Channel, storage management, and other applications:
http://www.t11.org
For information about the Fibre Channel industry, visit the Fibre Channel Industry Association
website:
http://www.fibrechannel.org
Getting technical help
Contact your switch support supplier for hardware, firmware, and software support, including
product repairs and part ordering. To expedite your call, have the following information available:
1. General Information
• Switch model
• Switch operating system version
• Error numbers and messages received
• supportSave command output
• Detailed description of the problem, including the switch or fabric behavior immediately
following the problem, and specific questions
• Description of any troubleshooting steps already performed and the results
• Serial console and Telnet session logs
• syslog message logs
2. Switch Serial Number
The switch serial number and corresponding bar code are provided on the serial number label,
as illustrated below:
*FT00X0054E9*
FT00X0054E9
xiiFICON Administrator’s Guide
53-1002753-01
Page 13
The serial number label on FICON-qualified devices is located as follows:
• Brocade 5100, 5300, 7800, and 6510 — On the switch ID pull-out tab located inside the
chassis on the port side on the left.
• Brocade DCX 8510-8 and DCX - On the port side of the chassis, on the lower right side and
directly above the cable management comb.
• Brocade DCX 8510-4 and DCX-4S - On the nonport side of the chassis, on the lower left
side.
3. World Wide Name (WWN)
Use the licenseIDShow command to display the switch WWN.
If you cannot use the licenseIDShow command because the switch is inoperable, you can get
the WWN from the same place as the serial number, except for the Brocade DCX. For the
Brocade DCX, access the numbers on the WWN cards by removing the Brocade logo plate at
the top of the nonport side of the chassis.
Document feedback
Quality is our first concern at Brocade and we have made every effort to ensure the accuracy and
completeness of this document. However, if you find an error or an omission, or you think that a
topic needs further development, we want to hear from you. Forward your feedback to:
documentation@brocade.com
Provide the title and version number of the document and as much detail as possible about your
comment, including the topic heading and page number and your suggestions for improvement.
IBM Fibre Connection (FICON®) is an industry-standard, high-speed input/output (I/O) interface for
mainframe connections to storage devices. This guide discusses support offered by Fabric OS in
intermix mode operations, in which FICON and Fibre Channel technology work together.
For specific information about intermix mode and other aspects of FICON, refer to the IBM
Redbook, FICON® Implementation Guide (SG24-6497-01), and Implementing an IBM/Brocade SAN with 8 Gbps Directors and Switches (SG24-6116-08).
1
In this guide, the term switch is used to refer to a Brocade switch, Backbone, or backbone platform
unless otherwise noted.
Fabric OS support for FICON
The following Fabric OS standard features support FICON fabrics:
• Blade swapping
Allows you to swap a blade of the same type so that you can perform a FRU replacement with
minimal traffic disruption. This feature is available for both FICON
environments. Blade swapping resolves situations in which the hardware has failed and the
channel configurations cannot be changed quickly. In addition, a blade swap minimizes and
eliminates the need to make changes to the I/O sysgen in the hardware configuration
definition (HCD). Blade swapping has minimal or no impact on other switch features.
• Routing policies
FICON does not support exchange based routing, but does support port based and device
based routing. For details on these policies, refer to the “Routing Traffic” chapter in the
Brocade Fabric OS Administrator’s Guide.
• FICON MIB module
Addresses link incident data for FICON hosts and devices connected to a switch. It
supplements other MIBs used to manage switches and should be used in conjunction with
those other MIBs. For more information, refer to the Fabric OS MIB Reference.
and open system
FICON Administrator’s Guide1
53-1002753-01
Page 16
FICON overview
NOTE
NOTE
1
• Insistent (IDID)
Disables the dynamic domain ID feature and only allows the switch to use a pre-set domain ID.
All switches in a fabric must have a unique domain ID. An insistent domain ID is required with
2-byte addressing. IDID is required for cascaded FICON configurations.
• Link incident detection, registration, and reporting
Provides administrative and diagnostic information.
• Swap port area IDs (PIDs) of physical ports
Redirects resources from a failed port to a healthy port without changing the mainframe
hardware configuration definition (HCD) settings. This feature, also called “port swapping,” is
available for both FICON
situations in which the hardware has failed and the channel configurations cannot be changed
quickly. This feature has minimal or no impact on other switch features.
and open system environments. Swapping PIDs on ports resolves
• Switch connection control (SCC) policy
Includes switch binding security methods that prevent unauthorized switches from joining a
fabric. SCC policy is required for cascaded FICON configurations and whenever 2-byte
addressing is used.
• Traffi c I s o l ation (TI) Zones and Enhanced TI Zones
TI Zones are used to direct traffic across links through a specified path. Enhanced TI Zones
allow you to have ports in more than one TI Zone and to program domain controller routes to
destination domains for F-class traffic, ensuring fabric stability.
For more detail on these features and configuration procedures, refer to the Fabric OS Administrator’s Guide.
Brocade management tools provide further support:
• Brocade Network Advisor
Brocade Network Advisor is an optional software program that can be used to manage a fabric
that supports FICON and FCP devices and traffic. This is the recommended GUI management
tool for FICON environments on B-series enterprise-class switches. For more information on
Brocade Network Advisor refer to the manual appropriate for your version requirements:
-Brocade Network Advisor SAN + IP User Manual
-Brocade Network Advisor SAN User Manual
• Fabric OS MIB Reference
The MIB reference provides information and procedures on setting up the SNMP agent and
FICON Management Information Base (MIB) on your switch.
• Web Tools
Web Tools is an embedded GUI management tool that can be used to manage a Brocade
switch or Backbone that supports FICON and Fibre Channel Protocol (FCP) devices and traffic.
For more information on Web Tools, refer to the Web Tools Administrator’s Guide.
Some licenses are installed and activated on the switch at the factory. Use a Brocade management
interface to verify that the required licenses are installed and activated on the switch.
2FICON Administrator’s Guide
53-1002753-01
Page 17
Latency guideline
The maximum supported distance for a FICON channel is 300 Km (1.5 msec of delay).
Synchronous mirroring applications are generally limited to 100 Km (0.5 msec of delay). Greater
distances require that the FICON Acceleration feature be used with FCIP. The FICON Acceleration
feature emulates control unit response to the channel to make the devices appear closer to the
channel than they actually are.
FICON concepts
Figure 1 shows how the traffic in a switched point-to-point configuration flows in a FICON
environment. The logical path of the traffic is defined as frames moving from the channel to the
switch to the control unit. FICON traffic moves from a logical partition (LPAR) and through the
channel, through a Fibre Channel link to the switch through the control unit, and ends at the device.
This is also called a channel path, which is a single interface between a central processor and one
or more control units along which signals and data can be sent to perform I/O requests. The
channel path uses the logical path to traverse the Fibre Channel fabric. The channel path is defined
using an ID, called the channel path ID (CHPID). This information is stored in the Input/Output
Definition File (IODF) and may be dynamically configured using the mainframe feature, zDAC
(Dynamic Auto-Discovery). The IODF is typically built using the hardware configuration definition
(HCD).
FICON concepts
1
FIGURE 1FICON traffic
The traffic on the channel path communicates using channel command words (CCWs) that direct
the device to perform device-specific actions, such as Seek, Read, or Rewind. In a FICON
environment, buffer credits are used at the fibre channel protocol level for flow control between
optically adjacent ports, while information unit (IU) pacing is the flow control mechanism used by
the channel. There are times when there are no more buffer credits to pass back to the other end
of the link and a frame pacing delay occurs. Frame pacing delay is the number of intervals of 2.5
microsecond duration that a frame had to wait to be transmitted due to a lack of available buffer
credits. Frame pacing delay information is reported in the FICON Director Activity Report with the
System z RMF feature.
FICON introduces the following concepts:
FICON Administrator’s Guide3
53-1002753-01
Page 18
FICON concepts
1
• FICON Control Unit Port (CUP)
The internal port in a switch that assumes an FC address such that it is the FC DID used to
direct FICON traffic to the FICON Management Server.
• FICON Manager
Host communication includes control functions such as blocking and unblocking ports, as well
as monitoring and error-reporting functions.
• Hardware Configuration Definition (HCD)
HCD is an IBM interactive interface application that allows you to define the hardware
configuration for both the processor channel subsystem and the operating system running on
the processor.
• Information unit
A unit of FICON data consisting of from one to four Fibre Channel frames.
• Link Incident Record Registration (LIRR)
The LIRR Extended Link Service (ELS) requests that the recipient add the requesting port to its
list of ports that are to receive a Registered Link Incident Report (RLIR).
• Node
A node is an endpoint that contains information. It can be a computer (host), a device
controller, or a peripheral device, such as a disk array or tape drive. A node has a unique 64-bit
identifier known as the Node_Name. The Node_Name is typically used for management
purposes.
• Prohibit Dynamic Connectivity Mask (PDCM) and connectivity attributes
PDCM controls whether or not communication between a pair of ports in the switch is
prohibited. Connectivity attributes control whether all the communication is blocked for a port.
• Read Record Set (RRS)
RRS is an IBM Channel-initiated CCW command. The Brocade FCIP FICON Acceleration License
allows the emulation of command chains that include this CCW command. The command is
used in IBM z/OS Global Mirror configurations to read updates from a volume in an active
mirroring session.
• Registered Link Incident Report (RLIR)
RLIR ELS provides a way for a node port to send an incident record to another node port.
• Request Node Identification Data (RNID)
RNID ELS acquires the associated node’s identification data, which provides configuration
discovery and management purpose information.
• Resource Measurement Facility (RMF)
Performance monitoring software that gathers transaction data from the environment and
generates performance reports. All Level II reports, which include port statistics, require the
FICON Control Unit Port (CUP) and FICON Management Server (FMS).
• Systems Operations (SysOps)
SysOps provides the ability to monitor and control all subsystems in a sysplex from any system
in the sysplex. This includes controlled startup, controlled shutdown, and automated recovery
of software resources.
4FICON Administrator’s Guide
53-1002753-01
Page 19
• Sysplex
In IBM mainframe computers, a Systems Complex, commonly called a sysplex, allows multiple
processors to be joined into a single unit, sharing the same sysplex name and Couple Data
Sets.
FICON configurations
There are two types of FICON configurations that are supported using Brocade Fabric OS: switched
point-to-point and cascaded topologies.
Switched point-to-point
A single-switch configuration is called switched point-to-point and allows the channel to use
single-byte addressing. Cascaded configurations require two-byte addressing, Figure 2 illustrates
an example of the switched point-to-point configuration.
FICON configurations
1
FIGURE 2Switched point-to-point FICON
FICON Administrator’s Guide5
53-1002753-01
Page 20
FICON configurations
Site A
FC
FC
FC SwitchFC Switch
FC Link
FC Link
FC Link
FC Link
FICON CU
FICON CU
Site B
ISL
1
Cascaded FICON
Cascaded FICON refers to an implementation of FICON that uses one or more FICON channel paths
whereby the domain ID of the entry switch is different than the domain of the switch where the
control unit is attached. Therefore, cascading requires a 2-byte link address. Any time a 2-byte link
address is defined on a channel, all link addresses must be 2-byte link addresses.
Switches may be interconnected using the following links:
• Traditional ISLs
• Inter-Chassis Links (ICL)
• Fibre Channel over Internet Protocol (FCIP)
The processor interface is connected to one switch (known as the entry switch), while the storage
interface is connected to the other. This configuration is supported for both disk and tape, with
multiple processors, disk subsystems, and tape subsystems sharing the ISLs or ICLs between the
switches and Backbones. Multiple ISLs between the switches and Backbones are also supported.
Cascading between switches and Backbones is also supported, An example of this would be a
Brocade DCX 8510-8 Backbone enterprise-class platform connected to a Brocade 6510.
A cascaded configuration (Figure 3) requires two-byte addressing. Two byte-addressing requires a
list of authorized switches. This authorization feature, called fabric binding, is available through the
Secure Access Control List feature. The fabric binding policy allows a predefined list of switches
(domains) to exist in the fabric and prevents other switches from joining the fabric. This type of
configuration is described in “User security considerations” on page 17.
FIGURE 3Cascaded FICON
6FICON Administrator’s Guide
There are hardware and software requirements specific to two-byte addressing:
• The FICON switches themselves must be from the same vendor (that is, both should be from
Brocade).
53-1002753-01
Page 21
FICON configurations
Channel
A
Control
Unit
B
Switch
Domain ID = 21
Switch
Domain ID = 22
Channel
A
Control
Unit
C
Control
Unit
D
Switch
Domain ID = 23
Switch
Domain ID = 21
Switch
Domain ID = 22
1
• The mainframes must be zSeries machines or System z processors: z196, z114, z800, 890,
900, 990, z9 BC, z9 EC, z10 BC, EC, and zEC12. Cascaded FICON requires 64-bit architecture
to support the 2-byte addressing scheme. Cascaded FICON is not supported on 9672 G5/G6
mainframes.
• z/OS version 1.4 or later, or z/OS version 1.3 with required PTFs/MCLs to support 2-byte link
addressing (DRV3g and MCL (J11206) or later) is required.
• Two-byte addressing, which has the following requirements:
-E_D_TOV must be the same on all switches in the fabric (typically this is not changed from
the default).
-R_A_TOV must be the same on all switches in the fabric (typically this is not changed from
the default).
-Insistent Domain ID.
-Fabric binding (strict SCC policy).
Qualified FICON cascaded configurations
Not all fibre channel fabrics are qualified for FICON. Cascaded FICON configurations are limited to
well-controlled paths. Only the channel paths described in this section are supported for FICON.
The resulting fabric scenario after ISL failures must not result in an unsupported configuration.
When physical cabling is not practical to enforce these configurations, zoning or traffic isolation
zoning (TI zoning) with failover disabled may be used to ensure unsupported fabrics cannot be
formed. Note that these restrictions apply to logical switches and not the chassis.
Figure 4 and Figure 5 show two cascaded configurations. These configurations require Channel A
to be configured for 2-byte addressing and require IDID and fabric binding. It is recommended that
there be only two domains in a path from a FICON Channel interface to a FICON Control Unit
interface. There are exceptions to the two-domain rule when extended fabric solutions are
deployed.
FIGURE 4Cascaded configuration, two switches
Figure 5 illustrates multiple switches cascaded off of switch 21 As long as there is only one hop
from channel to control unit, the configuration is supported.
Zoning is used to control access in a FICON environment. A zone consists of a group of ports or
WWNs. Connectivity is permitted only between connections to the switch that are in the same zone.
There are three types of zoning: WWN, port, and domain index zoning. A zone configuration
includes at least one zone. In open systems environments and in more complex FICON
deployments, the zone configuration contains multiple zones. Although index zoning is supported,
WWN zoning for QoS is recommended in environments where NPIV is deployed. For more
information on how to implement QoS domain index zoning in your fabric, refer to the QOS zones
section of the Fabric OS Administrator’s Guide for details.
When zoning changes occur, Registered State Change Notification (RSCN) messages are sent out
throughout the zone. RSCNs are part of the low-level Fibre Channel protocol that alerts channels
and devices to changes in the fabric. Using multiple smaller zones instead of one large zone helps
alleviate the need for channels and device interfaces to process RSCNs that are not relevant.
Session-based zoning enforcement is not recommended on a FICON switch. For more information
on session-based zoning enforcement, refer to the Fabric OS Administrator’s Guide.
Cascaded zoning
Figure 6 illustrates multiple sites sharing the same disaster recovery site. Each switch or Backbone
at a remote site—labeled Director 1 and Director 3—can pass traffic to Director 2, but no traffic is
permitted between Zone A and Zone B.
FIGURE 6Simple cascaded zoning
Figure 7 illustrates the multiple zoning concepts that can be used to restrict traffic. In Figure 7, any
host channel at the Backup Site (connected to Director 11 or Director 12) can connect to the
backup tape contained within the same zone. Notice that no more than a single hop is ever allowed
and only Channel Path Identifiers (CHPIDs) 79 and 7A on the Primary Site can connect to the
backup tape. Furthermore, CHPIDs 79 and 7A can only connect to the backup tape at the Backup
Site.
Zoning does not replace the need to set up the connectivity from the host to storage control units in
the HCD or IOCP. For more information on zoning, refer to the Fabric OS Administrator’s Guide.
8FICON Administrator’s Guide
53-1002753-01
Page 23
Access control in FICON
1
FIGURE 7Complex cascaded zoning
Blue Zone A: Any CHPID connected to Director 1, except CHPID 79, can get to any control unit
connected to Director 1. The zone includes all ports in Director 1 except ports 4, 5, and 6.
Orange Zone B: Any CHPID connected to Director 2, except CHPID 7A, can get to any control unit
connected to Director 2. The zone includes all ports in Director 2 except ports 4, 5, and 6.
Green Zone C: Any CHPID connected to Director 11 can get to any control unit connected to Director
11. The zone includes all ports in Director 11 except ports 5 and 6. Adding ports 5 and 6 to the
zone, so that all ports in the switch or backbone are in the same zone, would not affect permitted
connectivity and may be a more practical alternative.
Yellow Zone D: Any CHPID connected to Director 12 can get to any control unit connected to
Director 12. The zone includes all ports in Director 12 except ports 5 and 6, which are used for
ISLs.
Red Zone E: CHPID 79 can talk only to the remote tape connected to ports 7 and 8 on Director 11.
The zone includes port 4 of Director 1 and ports 7 and 8 of Director 11. Either ISL can be used.
Purple Zone F: CHPID 7A can talk only to the remote tape connected to ports 7 and 8 on Director
12. The zone includes port 4 of Director 2 and ports 7 and 8 of Director 12. Either ISL can be used.
FICON Administrator’s Guide9
53-1002753-01
Page 24
Access control in FICON
1
Error reporting
Non-implicit (such as Fabric OS recognized or bit error rate threshold exceeded) and implicit (FRU
failure) link incidents are reported to registered listeners on the local switch. The RMF 74-7 record
(FICON Director Activity Report, which is the same RMF Record containing the average frame
pacing delay information) reports port errors, which in turn are also reported back to the
mainframe host management consoles.
Secure access control
Binding is a method used to prevent devices from attaching to the switch or Backbone. Secure
Access Control List (ACL) provides the following fabric, switch, and port binding features:
• Fabric binding is a security method for restricting switches within a multiple-switch fabric.
Brocade recommends using fabric binding for cascaded FICON. SCC ACL with strict fabric-wide
consistency is required for FICON fabric binding.
• Switch binding is a security method for restricting devices that connect to a particular switch or
Backbone. If the device is another switch, this is handled by the SCC policy. If the device is a
host or storage device, the device connection control (DCC) policy binds those devices to a
particular switch. Policies range from completely restrictive to reasonably flexible, based upon
customer needs. SCC ACL with strict fabric-wide consistency is necessary for FICON switch
binding.
• Port binding is a security method for restricting host or storage devices that connect to
particular switch ports. The DCC policy also binds device ports to switch ports. Policies range
from completely restrictive to reasonably flexible, based on customer needs.
Figure 8 on page 11 demonstrates the three types of binding you can use depending on the
security requirements of your fabric.
10FICON Administrator’s Guide
53-1002753-01
Page 25
Access control in FICON
1
FIGURE 8Three types of binding
FICON Administrator’s Guide11
53-1002753-01
Page 26
FICON commands
NOTE
1
FICON commands
The Fabric OS CLI supports only a subset of the Brocade management features for FICON fabrics.
The full set of FICON CUP administrative procedures is available using the Brocade Network Advisor
and Web Tools software features. You can also use an SNMP agent and the FICON
Information Base (MIB).
Tab le 1 summarizes the Fabric OS CLI commands that can be used for managing FICON fabrics.
For detailed information on these commands, refer to the Fabric OS Command Reference Manual.
TABLE 1 Fabric OS commands related to FICON
CommandDescription
Standard Fabric OS commands
bladeSwapSwaps the area numbers for matching port pairs of two blades.
configureChanges a variety of switch configuration settings, including
configUploadBacks up the current configuration.
firmwareShowDisplays the current version of the firmware.
licenseAddAdds a license to the switch. The license key is case-sensitive and
licenseRemoveRemoves a license from the switch. Note that FMS mode must be
licenseShowDisplays current license keys, along with a list of licensed
licenseSlotCfg Enable and display slot-based licenses for a switch chassis. Note
portAddress Binds the 16-bit address to the lower two bytes of a port 24-bit
portSwapSwaps ports. Note that the portswap --restore command restores
portSwapDisableDisables the portSwap feature. The portSwap command cannot
portSwapEnableEnables the portSwap feature.
portSwapShowDisplays information about swapped ports.
supportShowCfgEnable ficonTurns on logging of FICON information on the switch.
Commands specific to FICON
ficonclear rlirRemoves all RLIR records from the local RLIR database.
Management
setting the domain ID and the insistent mode.
must be entered exactly.
disabled before removing the FICON license.
products enabled by these keys.
that for a switch without blades, such as the 7800 Extension
Switch, slot 0 is used as the slot-based license target. For blades,
slot numbers are based on the switch chassis numbering
scheme. A license key with the specified capacity must be
installed with the licenseAdd command before you can enable a
feature on a specified slot with this command.
Fibre Channel address. Also unbinds the currently bound address
for the specified port.
swapped ports to an unswapped state.
be used after this feature is disabled.The disabled state is
persistent across reboots and power cycles. Enabling or disabling
the PortSwap feature does not affect previously performed
PortSwap operations.
12FICON Administrator’s Guide
53-1002753-01
Page 27
FICON commands
TABLE 1 Fabric OS commands related to FICON (Continued)
CommandDescription
ficonclear rnidRemoves all outdated RNID records from the local RNID
database.
ficoncfg
ficoncfg
ficonhelpDisplays a list of FICON support commands.
ficoncupset fmsmode enable|disableEnables or disables FICON Management Server (FMS) mode.
ficoncupshow LPDisplays the CUP logical path and error reporting path
ficonshow lirr [fabric]Displays registered listeners for link incidents for the local switch
ficonshow rlir [fabric] While all FICON channels register as “conditional” recipients of
ficonshow rnid [fabric]Displays node identification data for all devices registered with
ficonshow rnid [port]Displays node identification data for a specified port.
ficonshow switchrnid [fabric]Displays node identification data for the local switch or for the
--set LIRR <portnumber>Sets the current LIRR device port number persistently.
--reset LIRR <portnumber>Clears the currently configured LIRR port number.
command processed by the CUP.
sampling information for diagnostic information collected by the
CUP.
information.
or for the fabric, if specified.
registered link incident reports (RLIRs) and are added to the
switch LIRR database, only one channel per switch is selected to
forward reports to the host. The command output displays all
channels which have registered and indicates which node on
each switch is selected to generate reports to the host.
the local switch or all devices registered with all switches defined
in the fabric, if specified.
fabric, if specified.
1
For limitations and considerations for using Fabric OS commands with FMS mode enabled, refer to
“Fabric OS command limitations and considerations” on page 40.
FICON Administrator’s Guide13
53-1002753-01
Page 28
Link and FC addressing
1
Link and FC addressing
To understand the addressing mode requirements and restrictions for FICON it is important to
understand the relationship between the link address and the FC address. Understanding this
relationship is also valuable for troubleshooting paths.
Figure 9 represents components of link and FC addresses that are explained in this section.
FIGURE 9Link and FC address components
Domain ID
Although you enter the domain ID in decimal format when configuring a switch, it is represented in
hexadecimal formal in the FC address.
For single-byte addressing the domain area returned from the switch where the channel logs in is
used for the FC address. Therefore the channel and control unit must be in the same logical switch.
With 2-byte link addressing the most significant byte of the link address is used for the domain
area of the FC address. Once 2-byte link addressing is defined for a channel, all link addressing for
that channel must use 2-byte link addressing.
Port area
The port address is the single-byte link address or the least significant byte of a 2-byte link address.
The link address is entered in the HCD in hexadecimal format and is represented in hexadecimal in
the FC address.
ALPA
The Arbitrated Loop Physical Address (ALPA) was originally used in fibre channel for loop devices.
Currently, the ALPA is used for Node Port Identification Virtualization (NPIV), which allows multiple
WWNs to log into the same switch port. The ALPA determines the logical entity to which frames
belong. This is how virtual servers using zLinux or zVM can share the same channel. Since the DCX
and DCX 8510-8 Backbones can have up to 512 ports, the upper two bits of the ALPA are used in
certain addressing modes.
14FICON Administrator’s Guide
53-1002753-01
Page 29
Link and FC addressing
FICON protocol does not use the ALPA byte. However, the ALPA is a required byte in the FC address.
The channel completes the FC address for a control unit link address by acquiring the ALPA that the
switch returned to the channel when the channel logged in. This is why the ALPA must be the same
for both the channel port and the control unit port. This is also the reason why 10-bit addressing
mode cannot be used for FICON (refer to “Addressing modes” on page 27).
To administer FICON, you must have one of the following roles associated with your login name on
the switch:
2
• Admin
• Operator
• SwitchAdmin
• FabricAdmin
The User and BasicSwitchAdmin roles are view-only. The ZoneAdmin and SecurityAdmin roles have
no access.
In an Admin Domain-aware fabric, if you use the FICON commands (ficonShow, ficonClear,
ficonCupShow, and ficonCupSet) for any Admin Domain other than AD0 and AD255, the current
switch must be a member of that Admin Domain. The output is not filtered based on the Admin
Domain. In virtual fabrics, these commands apply to the current logical or specified switch only.
Meeting Query Security Attribute requirements
In a cascaded switch configuration, FICON channels use an Extended Link Services Query Security
Attributes (ELS QSA) function to determine whether they are connected to a high integrity fabric.
When a FICON channel is connected to a fabric that is not high integrity, the channel will go into an
invalid attachment and isolated state (drop light), which then requires you to recover with the CPU
Hardware Management Console (HMC).
To ensure the FICON Channel QSA requirements have been met, be sure to configure the following
features:
To enable Insistent Domain ID, follow these steps for each switch in the fabric:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the configure command and step through the interactive prompts.
Enabling the fabric-wide consistency policy
Enable the fabric-wide consistency policy after all the switches have joined the merged fabric. If
there are fabric-wide data distribution (FDD) conflicts on any of the ISLs, disable the fabric-wide
consistency policy on each switch in the fabric.
Once the fabric has merged successfully (use fabricshow to verify), then enter the following
command.
fddcfg --fabwideset "SCC:S"
Note that the :S enforces strict mode, which is required for FICON.
a. At the “Fabric parameters” prompt, type y.
b. At the “Insistent Domain ID Mode” prompt, type y.
Using other security commands
The following commands are some other security-related commands that you might find useful.
Disabling fabric-wide consistency policy.
To disable the policy, enter the following command.
fddcfg --fabwideset ""
Displaying fabric-wide consistency policy
To display fabric-wide consistency policy information, enter the following command:
fddcfg --showall
Displaying the current security policy
To display the current security policy, enter the following command:
secpolicyshow
Creating the SCC policy
To create the switch connection control (SCC) policy and define switches allowed in the fabric, use
the secpolicycreate command to add all switches in the fabric, if they are connected.
secpolicycreate "SCC_POLICY","*"
You can also use the secpolicyadd command to add one or more members to an existing policy.
The following command is an example of adding a member using device WWNs.
secpolicyadd "SCC_POLICY","wwn1;wwn2"
18FICON Administrator’s Guide
53-1002753-01
Page 33
Deleting SCC_POLICY
NOTE
Enter the following command if you get messages that the E_Port in a security violation state.
secpolicydelete "SCC_POLICY"
Recovering the E_Port
Enter the following commands for each switch if the E_Port is down.
secpolicyactivate
portenable n
For more detailed information on commands and command output referenced in this section, refer
to the Brocade Fabric OS Command Reference Manual.
Preparing a switch for FICON
Use the following steps to verify and prepare a switch for use in a switched point-to-point FICON
environment. A single-switch configuration does not require insistent domain IDs (IDIDs) or fabric
binding, provided that connected channels are configured for single-byte addressing. However, you
should configure IDID to ensure that domain IDs are maintained.
Preparing a switch for FICON
2
1. Connect to the switch and log in using an account assigned to the admin role.
2. Verify the management IP addresses have been set up.
3. Verify the switches can be seen by your management tool.
4. Verify the switches have the correct version of Fabric OS.
5. Add feature keys, if applicable.
6. Enter the configUpload to save a baseline of the switch configuration.
Cascaded FICON and 2-byte addressing considerations
The following are considerations when installing a switch in a FICON environment where 2-byte
addressing is used. Two-byte addressing is always used in cascaded environments but may be used
in single-switch fabrics as well. Making changes to your Brocade switch or Backbone may require
scheduled downtime.
• All fabric operating parameters, such as timeout values, must be the same. If you have not
made any changes outside the scope of this document, there is nothing additional to consider
regarding these parameters.
• The domain IDs of all switches must be unique and insistent.
• Although not required, it will be easier to configure the security policies if the zoning matches.
Configuration, connecting the ISLs, and ensuring that the switches and Backbones have
merged into a fabric, will also make the process of setting the security attributes much easier.
FICON Administrator’s Guide19
53-1002753-01
Page 34
Configuring switched point-to-point FICON
CAUTION
NOTE
2
Configuring switched point-to-point FICON
This section provides detailed steps and commands to configure a switch for point-to-point FICON
operation. The following steps assume that you have used your hardware reference manual to
perform the initial setup of the switch and have performed all the steps from “Preparing a switch
for FICON” on page 19.
For basic steps and commands in a checklist format to quickly configure a switch for fabric and
possible FICON operation, refer to Appendix B, “Basic Switch Configuration”, then return to this
chapter for detailed FICON configuration procedures.
Use the worksheet in Appendix D, “Configuration Information Record,” to record your configuration
information.
Refer to “FICON commands” on page 12 for a list of FICON-related Fabric OS commands.
Configuring the switch for FICON is a disruptive process. The switch must be disabled to configure
switch parameters.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the switchDisable command. You will need to disable the switch to access all the switch
parameters.
3. Configure the switch and chassis name, if applicable.
4. Set the routing policy to port-based routing (aptPolicy 1) by entering the aptPolicy command.
The recommended best practice is to use port-based routing.
In addition, if FICON Emulation features (IBM z/OS Global Mirror or Tape Pipelining) are
enabled on an FCIP Tunnel in the switch, aptPolicy 1 must also be used.
5. Configure Dynamic Load Sharing.
The recommended best practice is to enable Dynamic Load Sharing (DLS); however, DLS is
only supported when Lossless is enabled.
• To enable Lossless with DLS, use the dlsSet --enable –lossless command.
• If Lossless will not be used, use the dlsReset command.
If Lossless with Lossless DLS is not enabled, the routing policy must be port-based routing
(aptPolicy 1).
6. Set In-Order Delivery using the iodSet command.
20FICON Administrator’s Guide
53-1002753-01
Page 35
Configuring switched point-to-point FICON
7. Configure the switch parameters using the configure command and enter the responses
shown in Tab le 2 when prompted. (Items in italics are top-level parameters.)
TABLE 2FICON switch parameters
ParameterResponseComment
Fabric parameterYesPrompts for the fabric parameters.
DomainThe domain ID is the switch address. The
recommended best practice is to set the switch
address and domain ID to be the same. The domain ID
is entered in decimal, so a switch address of
hexadecimal 0x22 would have a domain ID of decimal
34.
R_A_TOV10000Do not change. The Resource Allocation TimeOut Value
(RA_TOV) is entered in milliseconds. This works with
the variable E_D_TOV to determine switch actions
when presented with an error condition.
Allocated circuit resources with detected errors are not
released until the time value has expired. If the
condition is resolved prior to the timeout, the internal
timeout clock resets and waits for the next error
condition.
E_D_TOV2000The Error Detect TimeOut Value (E_D_TOV) is a timeout
value entered in milliseconds. This timer is used to flag
a potential error condition when an expected response
is not received within the set time limit. If the time for
an expected response exceeds the set value, then an
error condition occurs. Leave this value at 2 seconds
(2000) unless connected to extension equipment. In
some cases, when connecting to extension equipment,
such as the Brocade 7800, it needs to be set to 5
seconds (5000). This field should only be changed
after consulting a Brocade Certified Fabric
Professional.
WAN_TOV0Wide-Area Network TimeOut Value (WAN_TOV) is set in
milliseconds. This is the maximum frame timeout value
for a WAN, if any, interconnecting the Fibre
Channel islands.
MAX HOPS7This parameter should not be confused with the one-
hop FICON cascading limitation. There are
configurations that are more than one hop because
more than two domain IDs are used in the path that fit
within the IBM support limitation of “one hop.”
Data field size2112Do not change. Specifies the largest possible value, in
bytes, for the size of a type 1 (data) frame. The switch
advertises this value to other switches in the fabric
during construction of the fabric and to other devices
when they connect to the fabric. Setting this value
smaller than 2112 might result in decreased
performance.
2
FICON Administrator’s Guide21
53-1002753-01
Page 36
Configuring switched point-to-point FICON
2
TABLE 2FICON switch parameters (Continued)
ParameterResponseComment
Sequence level switching0Do not change. When set to 1, frames of the same
Disable device probing1The recommended best practice is to disable device
Suppress Class F traffic 0Do not change. By default, the switch can send Class F
Per-frame route priority0Do not change. In addition to the virtual channels used
Long-distance fabric0Do not change. The recommended best practice is to
Insistent Domain ID modeYesThe recommended best practice is to set the domain
Virtual Channel parametersNoDo not change. The switch allows fine-tuning of a
F-Port login parametersNoWith Fabric OSv6.1.0 and later, it is not necessary to
Maximum logins per switchChange this only if NPIV is being used and the number
Maximum logins per portChange this only if NPIV is being used and the number
Maximum logins per secondFor Fabric OS v6.1.0b and later – Do not change. Leave
Login stage intervalDo not change. Specifies the stage interval in the
Zoning operation parametersNoDo not change. Disables and enables NodeName Zone
sequence from a particular source are transmitted
together as a group. When set to 0, frames are
interleaved among multiple sequences when
transmitted. Under normal conditions, sequence-level
switching should be disabled for better performance.
probing. When disable device probing is set to 1,
devices that do not register with the Name Server will
not be present in the Name Server database.
frames. When this option is turned on (set to 1),
Class F traffic is converted to Class 2 traffic before
being transmitted to support remote fabrics that
involve ATM gateways.
in frame routing priority, when this value is set to 1,
support is also available for per-frame-based
prioritization.
configure individual ports for long distance when
cascading at extended distances.
ID to be insistent. Setting the insistent domain ID is
required for 2-byte addressing.
specific application through virtual channels
parameters. The default virtual channel settings have
already been optimized for switch performance.
Changing these values can improve switch
performance, but can also degrade performance.
make any changes to F_Port login parameters. The
only other reason to answer yes (y) to this prompt is if
NPIV is being used and there is a need to change the
default number of NPIV logins.
of fabric logins will exceed the default or there is a
need to limit the number of logins.
of port logins will exceed the default or there is a need
to limit the number of logins.
at default.
NOTE: The default value is incorrectly shown as 0.
Arbitrated Loop parametersNoParameters include Send FAN frames, Enable CLOSE
on OPEN received, and Always send RSCN. Refer to the
configure command in the Brocade Fabric OS
Command Reference Manual for more information.
System servicesNoEnables or disables the read link status (RLS) probing
performed by the FCP daemon
Portlog events enableNoDetermines whether or not various types of port events
are logged.
SSL attributesNoEnables or disables Secure Socket Layer attributes.
RPCD attributesNoEnables or disables Remote Procedure Call Daemon
(RPCD) attributes.
cfgload attributesNoConfigures configuration upload and download
parameters. Refer to the configureChassis command
in the Brocade Fabric OS Command Reference Manual
for more information.
Web Tools attributesNo
SystemNo
8. Enter the switchEnable command to set the switch online.
9. Enter the trackChangesSet command to enable or disable the track-changes feature and to
configure the SNMP-TRAP mode.
10. Enter the fabricPrincipal command on the switch you want to be the principal switch.
In cascaded fabrics, only one switch should be the principal. In core-edge designs, the director
class switch should be the principal.
11. Enter the portCfgSpeed command to configure port speeds.
By default, all ports are configured to auto-negotiate. Normally, the only time the port
configuration is changed is when connecting to 1 Gbps ports. Some DWDM ports are known
not to auto-negotiate well. Furthermore, although older 1 Gbps FICON channels log in correctly,
they will generate errors, so the speed should be forced to 1 Gbps with 1 Gbps FICON
channels.
8 Gbps ports will not auto-negotiate to 1 Gbps and cannot be configured for 1 Gbps. 16 Gbps
ports will not auto-negotiate to 2 Gbps or 1 Gbps.
FICON Administrator’s Guide23
53-1002753-01
Page 38
Configuring switched point-to-point FICON
NOTE
2
12. Enter the portCfgLongDistance command for ports with fiber connections exceeding 10 km.
For ports with fiber connections exceeding 10 km, configuring the port for long distance mode
increases the number of buffer-to-buffer (BB) credits available on that port. If you have any
Extended Fabrics links, enable VC translation link initialization to stabilize them. Refer to the
Fabric OS Administrator’s Guide for details on this option of the portCfgLongDistance
command.
13. Configure the fill word for all 8 Gbps ports using the portcfgfillword command. The following
example shows the command with recommended mode 3 and passive operands set:
portcfgfillword slot/port, 3 passive
This command is only supported on 8 Gbps FC ports. For more information, refer to the
portcfgfillword command in the Fabric OS Command Reference.
14. Enter the defZone
15. Configure zoning for your fabric.
Follow standard FICON zoning practices published by IBM. For more information on zoning,
refer to the administrator’s guide for your management tool.
16. Enter the statsClear command to clear port statistics.
17. Ente r t he switchShow command to verify that the switch and devices are online.
18. Enter the ficonShow rnid command to verify that the FICON
switch. Enter this command only if fabric binding is enabled; otherwise, channels will be
attached as invalid. The ficonShow rnid <port> command displays rnid information for a port. T
19. Enter the ficonShow lirr command to verify that the FICON
listen for link incidents. Enter this command only if fabric binding is enabled; otherwise,
channels will be attached as invalid.
20. Enter the configUpload command to save your FICON configuration.
21. Enter the command appropriate for your environment:
--noaccess command to disable the default zone.
devices are registered with the
host channels are registered to
• If you do not have access to an FTP server, use a Telnet session that can log console
output and enter the supportShow command to save a FICON baseline.
• If you have access to an FTP server, enter the supportSave command to save the FICON
baseline of your switch.
22. Take the appropriate action based on your configuration:
• If you have a cascaded configuration, continue configuring your switches using the
instructions in “Configuring cascaded FICON” on page 25.
• If you have and plan to use a CUP license, continue configuring your switches using the
instructions in “Configuring FICON CUP” on page 37.
• If you have a switched point-to-point configuration, you have completed the switch
configuration.
24FICON Administrator’s Guide
53-1002753-01
Page 39
Configuring cascaded FICON
CAUTION
In addition to performing the steps listed in “Configuring switched point-to-point FICON” on
page 20, you also need to perform the following steps to configure cascaded FICON.
Configuring the switch for FICON is a disruptive process. The switch must be disabled to configure
switch parameters. This procedure must be followed prior to channels joining the fabric.
1. Perform the following actions for each switch:
a. Enable the IDID mode using the Fabric OS configure command. For details on enabling the
IDID mode, refer to “Enabling insistent domain ID” on page 18.
b. Set the domain ID. It is highly recommended that the switch ID used in HCD and IOCP be
the same as the switch address in hexadecimal. All switches ship from the factory with a
default domain ID of 1. The best practice is to use something other than 1 so that a new
switch can be added to the fabric without changing the domain ID.
2. Use the secPolicyCreate command to configure the SCC policies on all switches to limit
connectivity to only the switches in the selected fabric.
In the command syntax, member indicates a switch that is permitted to join the fabric. Specify
switches by WWN, domain ID, or switch name. Enter an asterisk (*) to indicate all the switches
in the fabric. To create a policy that includes all the switches in the fabric, enter the following
command:
switch:admin> secPolicyCreate SCC_POLICY ”*”
3. Save or activate the new policy by entering either the secPolicySave or the secPolicyActivate
command. If neither of these commands is entered, the changes are lost when the session is
logged out. To activate the SCC policy, enter the following command:
switch:admin> secPolicyActivate
4. Enter the fddCfg command to enable the ACL fabric wide consistency policy and enforce a
strict SCC policy.
switch:admin> fddcfg --fabwideset ”SCC:S”
5. Connect and enable channel and control unit (CU) devices. The Query for Security Attributes
(QSA) response to the channel indicates that the fabric binding and IDID are enabled.
FICON Administrator’s Guide25
53-1002753-01
Page 40
FICON and FICON CUP in Virtual Fabrics
2
FICON and FICON CUP in Virtual Fabrics
For FICON CUP, FICON Management Server (FMS) mode is enabled on the logical switch and not
the chassis. For example, in a Virtual Fabrics environment on the Brocade 6510 switch and the
Brocade DCX Backbones, enable CUP on each logical switch.
You can configure up to four CUP switches as logical switches in a Virtual Fabrics-enabled platform
for FMS mode. You can find the complete instructions for creating a logical switch using Fabric OS
commands in the Fabric OS Administrator’s Guide or for using Brocade Network Advisor in the
Brocade Network Advisor User Manual. Once the logical switches are configured, install CUP on the
logical switch per the instructions in “Configuring FICON CUP” on page 37.
Tab le 3 specifies the maximum number of logical switches and maximum number of logical
switches with FMS mode enabled for supported Brocade switches and Backbones.
TABLE 3Maximum logical switches for FICON switches and Backbones
Switch or backboneMaximum number of logical switchesMaximum number of logical switches
8510-884
8510-484
65104 (3 if using a base switch)2
78004 (does not support base switch)2
DCX84
DVC-4S84
530042
with FMS (CUP) enabled
When creating a logical switch, consider the following for FICON:
• Set the addressing mode to 1 (zero-based) or 2 (port-based) for FICON CUP. Setting these
modes may be required for specific hardware configurations. Following are examples:
-For 48-port blades in an 8-slot chassis, FICON is supported only on a logical switch with
zero-based (mode 1) configured. This must be in a logical switch other than the default
switch.
-For 48-port blades in a DCX-4S backbone, FICON is supported only on a logical switch
defined for port-based (mode 2) addressing. This must be in a logical switch other than the
default switch.
For more information on addressing modes, refer to “Addressing modes” on page 27.
• You can use the Fabric OS Configure command to Allow XISL Use.
An extended interswitch link (XISL) is an interswitch link (ISL) that connects the logical switch
to a base switch and carries traffic for multiple logical fabrics. The XISL is a logical switch
feature, not a link feature. It is shared by multiple logical switches and can be used for multiple
(Fabric IDs) FIDs. Creating an XISL is disruptive since it provides a core routing function for the
switch. For more information on creating XISLs and restrictions, refer to the Configure
command in the Fabric OS Command Reference.
• A base switch can only be used for XISLs and therefore cannot be managed through a CUP port
and cannot report statistics for RMF.
• Logical switches enabled for FICON CUP are not supported as fabric-capable logical switches,
so these switches will not participate in the use of extended ISLs on the base switch.
26FICON Administrator’s Guide
53-1002753-01
Page 41
• If addresses 0xFE or 0XFF have been assigned to physical ports, those ports must be disabled
before FMS can be enabled.
• For FICON CUP, FMS mode is enabled on the logical switch and not the chassis. For example, in
a Virtual Fabrics environment on the Brocade 6510 switch and the Brocade DCX 8510-4 and
DCX 8510-8 Backbones, you must set FMS mode independently on each logical switch. FMS
mode is currently limited to four logical switches per chassis.
• Each logical switch has its own configuration data, such as the initial program load (IPL) file,
configuration data elements (PDCM, PIB, port address names, mode register, key counter),
and CUP configuration files.
Addressing modes
There are three 256-area addressing modes that provide extended functionality in dynamically
created partitions:
• Mode 0 or 10-bit addressing.
• Mode 1 or zero-based addressing.
• Mode 2 or port-based addressing.
Addressing modes
2
Mode 0 (10-bit addressing)
This mode is called “Disabled” in the Brocade Network Advisor. It is not Supported For FICON, but
can be used for mainframe FCP channels associated with zLinux or zVM partitions.
Mode 0 uses the two upper bits in the ALPA to address up to 1024 ports in a single logical switch.
Since a 64-port blade is the highest port count blade available and eight slots is the highest slot
count available in a Brocade chassis, the practical address limit is 512 ports. Since NPIV also uses
these bits, this address mode leaves just 6 bits for NPIV. Therefore, the maximum NPIV logins is
limited to 64 (63 practical), not 256.
Mode 1 (zero-based addressing)
This is the recommended mode for FICON. In this addressing mode, port numbers are assigned
starting with 00 as the ports are moved into the logical switch. Typically, several ports are moved at
once, resulting in a port address assignment starting with 00 for the bottom port in the left-most
port column on the blade in the lowest slot number. Starting from port 00, addresses increment by
one as ports progress up the blade and to the bottom port of the right port column. Numbers then
increment to the top port of the column, and then on to the next blade. This is different than how
port addresses are assigned in a chassis that is not enabled for virtual fabrics.
The ALPA is always 0x00. Therefore zero-based addressing limits the maximum switch size to 256
ports. Since zero-based addressing does not use any of the ALPA bits and may be preferred for
native FCP with zVM to allow up to 256 (255 practical) NPIV logins.
FICON Administrator’s Guide27
53-1002753-01
Page 42
Addressing modes
NOTE
NOTE
2
Although port addresses are assigned when moved to a logical switch, the address is not bound to
the port. To ensure that the link address for a control unit attached to a port does not change, the
portAddress command must be used to bind the address to the port even if you are not changing
the port address assignments. Refer to “Sequential address binding” on page 89 for a list of
commands that will bind ports on DCX or DCX 8510-8 Backbones fully populated with 32-port
blades.
Mode 2 (port-based addressing)
Port based addressing follows a default port numbering scheme that starts with 00 for the port on
the bottom left-most column of the blade in slot 1 and increments for each port going up the blade.
For a DCX and DCX 8510-8 Backbone, the right-most port column on the blade starts with 0x80,
and port numbering increments going up the blade. Subsequent blade slots follow the same port
numbering scheme with an offset of 0x80. The ALPA is always 0x00 except on the 48-port blades.
The 48-port blades use bits in the ALPA for the upper 16 ports on bladed switches that support
more than 256 ports. This is why the 48-port blade in the DCX and DCX 8510-8 Backbones is only
supported with zero-based addressing. The DCX-4S and DCX 8510-4 Backbones do not use the
ALPA bits with the 48-port blades. Port-based addressing is used in a Brocade switch or backbone
not enabled for virtual fabrics.
Refer to“Zero-Based to Port-Based Address Binding” on page 95 for a list of commands that will
allow DCX or DCX 8510-8 Backbones fully populated with 32-port blades that are configured for
zero-based addressing to use port-based addresses.
Changing the addressing mode
Changing the addressing mode can have serious implications. After changing the addressing
mode, the following actions occur:
• Physical ports can receive new area assignments.
• Port addresses that are referenced in an IOCDS gen can be associated with entirely new
physical ports.
To change the addressing mode, use the following procedure.
To access all parameters controlled by the configure command, you must disable the switch. Any
configuration change to a logical switch that requires disabling the switch also requires disabling
FMS mode.
1. Disable FMS mode using the ficonCupSet fmsmode disable command.
2. Disable the switch using the switchDisable command.
3. Change the address mode using the configure command.
The configure command runs in interactive mode and presents you with a series of hierarchical
menus. To change the address mode, select the Fabric parameters menu, then select the
Enable a 256 Area Limit field and enter appropriate option values. For more information on
this command and an example of command output, refer to the Fabric OS Command Reference Manual.
4. Enable the switch using the switchEnable command.
28FICON Administrator’s Guide
53-1002753-01
Page 43
Clearing the FICON management database
NOTE
5. Enable FMS mode using the ficonCupSet fmsmode enable command.
For more information on these Fabric OS commands, refer to the Fabric OS Administrator’s Guide.
2
FICON and blade support for addressing modes
Note the following considerations about FICON and blade support on logical switches in DCX
systems:
• On default logical switches (and non-Virtual-Fabrics switches), FICON is not supported if
address mode 1 (dynamic address mode) is enabled.
• Address mode 1 is not supported if FICON CUP is enabled on the default logical switch.
• On default logical switches with an address mode other than mode 1, any 48-port and 64-port
blades are disabled.
Moving ports to a logical switch
Ports moved into a logical switch will have the following states based on the Active=Saved (ASM) bit
setting:
You cannot issue the portCfgPersistentDisable command if FMS mode is enabled. When FMS mode
is enabled, use the portDisable command.
• If the ASM bit is on and the port is blocked in the initial program load (IPL) file, the port
becomes persistently disabled. The port displays as persistently disabled through the
portCfgPersistentDisable command and remains blocked in the IPL file.
• If the ASM bit is on and the port is unblocked in the IPL file, the port becomes persistently
disabled. The port displays as persistently disabled through the portCfgPersistentDisable
command, but is blocked in the IPL file.
• If the ASM bit is off and the port is blocked in the IPL file, the port will become persistently
disabled. The port displays as persistently disabled through the portCfgPersistentDisable
command and remains blocked in the IPL file.
• If the ASM bit is off and the port is unblocked in the IPL file, the port will become persistently
disabled. The port does not display as persistently disabled through the
portCfgPersistentDisable command and remains unblocked in the IPL file.
Port swap limitations
For limitations for port swapping on logical switches and when FMS mode is enabled, refer to “Port
swap limitations” on page 42.
Clearing the FICON management database
Perform the following steps to clear RLIR and RNID records from the FICON management database.
1. Connect to the switch and log in as admin.
2. Enter ficonClear rlir to remove all the RLIR records from the local RLIR database.
FICON Administrator’s Guide29
53-1002753-01
Page 44
Automating CS_CTL Mapping
NOTE
NOTE
2
3. Enter ficonClear rnid to remove all the RNID records marked “not current” from the local RNID
database.
Automating CS_CTL Mapping
The CS_CTL-based frame prioritization feature allows you to prioritize the frames between a host
and target as having high, medium, or low priority, depending on the value of the CS_CTL field in
the FC frame header. High, medium, and low priority frames are allocated to different virtual
channels (VCs). High priority frames receive more VCs than medium priority frames, which receive
more VCs than low priority frames. The virtual channels are allocated according to the CS_CTL
value.
Check with your host and storage manufacturer to determine if they support Fibre Channel CS_CTL
prioritization on their devices.
The Fabric OS Administrator's Guide provides additional information on this feature and steps to
enable CS_CTL-based frame prioritization on your switch using the portcfgqos command.
The configurechassis command allows you to set chassis-level configuration parameters, such as
cfgload, custom, system, and Fabric OS attributes, to be used by ASIC drivers. The following
procedure uses the Fabric OS attribute, CSCTL QoS Mode, to change CSCTL values mapped to VCs
for priority levels from default to auto mode. While default mode maps CSCTL values 1-8 (low), 9-16
(medium), and 17-24 (high) to specific VCs, auto mode maps CSCTL values 1 (low), 2 (medium),
and 3 (high) to VCs based on the ASIC supported by your platform. For DCX 8510-4, DCX 8510-8,
and 6510 models, a single CSCTL value is mapped to multiple VCs.
To enable auto mode for CSCTL-VC mapping on a FICON switch, use the following steps.
This process is disruptive as you must reboot the chassis to reset the configuration in the ASIC. You
cannot enable CS_CTL mapping on individual ports until this is done. If Auto Mode is selected, you
must select Auto Mode for each chassis in the fabric. Not doing so will mean different virtual
channels (VCs) will be used across the fabric with undesirable results.
1. Connect to the switch and log in to an account that has admin permissions.
2. Launch the configurechassis command.
3. Select no when prompted to modify cfgload attributes, custom attributes, and system
attributes, then select yes to modify Fabric OS attributes as in the following example:
root> configurechassis
Configure...
cfgload attributes (yes, y, no, n): [no]
Custom attributes (yes, y, no, n): [no]
system attributes (yes, y, no, n): [no]
fos attributes (yes, y, no, n): [no] y
4. Select 1 (auto mode) when prompted to configure CSCTL QoS Mode as in the following
example:
• If auto mode is set, the following should display in the command output:
fos.csctlMode:1
• If default mode is set, the following should display in the command output:
fos.csctlMode:0
To set return to default mode, follow the preceding steps, but select 0 (default mode) for CSCTL
QoS Mode when modifying Fabric OS attributes through the configurechassis command.
The configurechassis command does not provide options to enable the CSCTL QoS mode on
individual ports in the chassis. Once you set the mode on the chassis, you must enable individual
ports to use this mode with the portcfgqos command. For details on the configurechassis and
portcfgqos commands, refer to the Brocade Fabric OS Command Reference for details.
FICON best practices
Use the following best practices for configuring FICON on Brocade switches and Backbones.
• Configuring Domain ID
Give all switches a unique domain ID and always set to IDs to “insistent.”
• Switch ID
Make the Switch ID in the IOCP or HCD the same as the hexadecimal equivalent of the domain
ID. Although not required, this will make it easier to correlate switches to the IOCP. For
example, if the domain ID is 32 (0x20), the Switch ID should be 20.
• Firmware
Following are requirements:
-Use only qualified firmware for FICON.
-All versions of firmware for switches that Brocade Network Advisor will be managing
should be in the Brocade Network Advisor repository.
-All switches in a fabric should be at the same code level.
-When upgrading code more than two levels, plan the upgrades so no switch in the fabric is
more than one release apart.
-Always check the version of firmware on a switch before putting the switch into production.
-If a firmware upgrade is required on a new switch installation, upgrade the firmware before
making any configuration changes.
• CHPID and control unit placement
Place CHPID and control unit connections on the same port card when possible to minimize
the impact of a card failure.
FICON Administrator’s Guide31
53-1002753-01
Page 46
FICON best practices
2
• Default switch
Leave the default switch FID at 128.
• Logical switches
Always use a logical switch other than the default switch for FICON and do not disable virtual
fabrics.
• Cascading
When cascading switches, connect ISLs so that a fabric forms before configuring switches for
FICON. Merging switches into the fabric after configuring FICON is possible, but allowing the
fabrics to form before applying the FICON settings will simplify the configuration process.
• Address mode
Always use Mode 1 (zero-based) addressing. You can bind ports to specific addresses and
ensure Mode 1 or Mode 2 (port-based) addressing using the portaddress --bind command.
Use the Fabric OS scripts in“Sequential address binding” on page 89 and “Zero-Based to
Port-Based Address Binding” on page 95 to help simplify configuration.
•Persistently enabling and disabling ports for CUP . . . . . . . . . . . . . . . . . . . . 46
Control Unit Port overview
The Control Unit Port (CUP) provides an in-band management interface defined by IBM that defines
the Channel Command Words (CCWs) that the FICON host can use for managing the switch. The
protocol used is the IBM version of the ANSI FC-SB3 single-byte command code specification,
which defines the protocol used for transporting CCWs to the switch, and for the switch to direct
data and status back. The CUP is an optional licensed feature.
3
The CUP provides an interface for host in-band management and collection of FICON switch
performance metrics using the RMF 74-7 record, more commonly known as the FICON Director
Activity Report. Host-based management programs manage the FICON switches by sending
commands to the switch Control Unit defined in the IOCDS and Hardware Configuration Definition
(HCD). A FICON switch that supports CUP can be controlled by one or more host-based
management programs or switch consoles. Control of the FICON switches can be shared between
these options. CUP commands, or CCWs, monitor and control FICON switch functions.CUP
commands are oriented towards management of a single switch, even though the use of CUP in a
cascaded FICON environment is fully supported.
The CUP port address will always be xxFExx. Port Addresses xxFE and xxFF are unavailable to assign
to physical ports when FMS is enabled. For more information, refer to “Disabling ports 0xFE and
0xFF” on page 38.
FICON Management Server mode (FMS mode) must be enabled on the switch to enable CUP
management features. For more information, refer to “FMS mode and FICON CUP” on page 39.
FICON Administrator’s Guide33
53-1002753-01
Page 48
Control Unit Port overview
FE
FE
CHPIDs
Management
Application Server
FICON Switch
Embedded
port
FE
Embedded
port
Embedded
port
FICON director
FICON director
System Automation
OS390 application
I/0-CPs
3
FIGURE 10FICON CUP
Figure 10 is a simplified representation of a FICON environment and how CUP fits into that
environment. FICON switches with FMS enabled use port address “FE” as the embedded internal
port for host connectivity to the CUP, and port address “FF” is reserved. Therefore, addresses “FE”
and “FF” cannot be used for assignment to physical ports when FMS is enabled. For more
information, refer to “Disabling ports 0xFE and 0xFF” on page 38.
In a FICON environment, only one RMF LPAR should be configured to poll the CUP for port statistical
counter data in order to minimize contention for the CUP device. Host contention for access to the
CUP can lead to missing interrupts, potentially causing the host to box the device. It is a best
practice however to have more than one CHPID defined as a path to the CUP in order to ensure
redundancy for host-to-CUP connectivity.
FICON CUP restrictions
The following restrictions apply to Brocade FICON switches having at least 256 ports when FMS
mode is enabled and when host applications are used for inband monitoring and management of
the switch:
• FICON Management Server mode (FMS mode) must be enabled on the switch to enable CUP
management features. When this mode is enabled, Fabric OS prevents configuration requests
from out-of-band management interfaces from interfering with host in-band configuration
requests by serializing access to switch parameters.
34FICON Administrator’s Guide
53-1002753-01
Page 49
Control Unit Port overview
ATTENTION
Once FMS mode is enabled, you cannot bind FE/FF to any ports using either the portAddress
or wwnAddress commands.
3
• The switch is advertised to the mainframe by the CUP as having a specific number of ports,
which is based on the hardware platform. The maximum supported switch size is 256 ports.
The addresses FE/FF are not components of the Allow/Prohibit Matrixes (as they are called in
the Brocade Network Advisor and Web Tools), because the Prohibit Dynamic Connectivity
Masks (PDCMs) for these two ports are defined architecturally and cannot be modified by the
user.
The Port Address Name for the internal FE port can be read by the host, and corresponds to
the switch name that is set for the switch. The switch name can be modified by the host, with
the limitation that the host can write EBCDIC characters that cannot be converted directly to
ASCII for display in the GUI or CLI management interfaces. Such characters will be converted to
periods for these displays.
• Refer to Appendix A, “Platforms supporting FICON” for details on FICON support for Brocade
blades in various chassis products.
• Addresses 0xFE and 0xFF are not available for external connections. Before enabling FMS
mode, you must manually disable physical ports associated with 0xFE and 0xFF if there is an
active device connected to these ports. If there is an active device connected to either of these
ports, Fabric OS will not enable FMS until either the port is disabled or the device is taken
offline (logs out of the switch). If there is no active device connected, these ports will
automatically be disabled when FMS enables. Because these ports are not available after
enabling FMS mode, you should first move any fiber connected to either of them to another
free port. For more information, refer to “Disabling ports 0xFE and 0xFF” on page 38.
CUP configuration recommendations
The following are recommendations for setting up FICON CUP:
• Device Type: 2032
Use this type as the CUP "UNIT" type for the CNTLUNIT and IODEVICE statements in the IOCDS
or HCD. Although IBM has developed new machine types for switches, all Brocade Backbones
and switches regardless of model or manufacture should be configured as 2032.
• Paths
Define at least two CHPIDs as paths to the CUP in the IOCDS Control Unit statement for the
switch to ensure redundancy for host connectivity.
• RMF 74-7 statistics gathering
So as not to overwhelm the CUP, the recommended best practice in a sysplex environment is to
disable FICON switch statistics (RMF polling) on all but one host system that has inband
access to the CUP. If no FICON statement is specified, the system will gather switch statistics
by default so FICON switch statistics should be disabled on all but the system intended to
gather statistics.
• Missing Interrupt Handler (MIH)
Set the MIHPTO to three minutes (180 seconds).
FICON Administrator’s Guide35
53-1002753-01
Page 50
Port and switch naming standards for FMS mode
NOTE
3
Although 180 seconds should be the default value set on the switch when FMS is enabled, this
value may not be set when older versions of Fabric OS are upgraded. This value must be set on
the host system as well.
FICON CUP zoning and PDCM considerations
The FICON PDCMs control whether or not communication between a pair of ports in the switch is
prohibited or allowed. PDCMs are a per-port construct. Each port on the switch has its own PDCM
that defines whether communication is allowed between that particular port, and each of the other
ports in the switch, including itself. The Allow/Prohibit Matrix presents a matrix that allows you to
set and clear the PDCMs for all external ports. If there are any differences between restrictions set
up with Brocade Advanced Zoning and PDCMs, the most restrictive rules are automatically applied.
FMS must be enabled for the Allow/Prohibit Matrix to be operational, and there must be an active
zoning configuration. Domain,Index (D,I) zoning is recommended for FICON, but is not specifically
required. Hard-port or Hard-WWN zoning enforcement is required however. PDCMs will not be
enforced on ports with session-based enforcement.
If Brocade Advanced Zoning is in use, refer to the Fabric OS Administrator’s Guide for more
information on configuration.
Port and switch naming standards for FMS mode
FMS and the FICON Programming Interface employ 8-bit EBCDIC characters in port address names
and switch names. Fabric OS employs 7-bit characters. Therefore, when FMS mode is enabled, all
characters greater than 0x40 and not equal to 0xFF (EBCIDC code page 037 [0x25]; refer to
Appendix E, “EBCDIC Code Page”) are allowed in port address names, and it is possible for a
channel to set a name with nonprintable characters. If a name contains nonprintable characters,
those characters are displayed as dots (...). The following characters are also displayed as dots:
semicolon (;), comma (,), equal sign (=), and at sign (@).
The port name for the CUP port must be unique within the chassis.
FICON CUP Fabric OS commands
Tab le 4 summarizes the Fabric OS CLI commands that can be used for managing FICON fabrics.
For detailed information on these commands, see the Brocade Fabric OS Command Reference Manual.
TABLE 4 Fabric OS commands related to FICON CUP
CommandDescription
ficonCupset fmsmodeSets FICON Management Server mode on or off for the switch.
ficonCupset crpSets the CUP Asynchronous Event Reporting Path.
ficonCupset MIHPTOSets the Missing Interrupt Handler Primary Timeout value.
ficonCupset moderegSets the mode register bits for the switch.
ficonCupShow lpDisplays the host logical path information for the CUP.
36FICON Administrator’s Guide
53-1002753-01
Page 51
ATTENTION
TABLE 4 Fabric OS commands related to FICON CUP
CommandDescription
ficonCupShow fmsmodeDisplays the FICON Management Server enabled or disabled state for the
ficonCupShow moderegDisplays the mode register bit settings for the switch.
Configuring FICON CUP
To set up FICON CUP, use the following procedure and be sure to perform the steps in the order
indicated.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Verify that the switch or Backbone has been set up according to the instructions in
“Configuring switched point-to-point FICON” on page 20 and, if in a cascaded topology,
“Configuring cascaded FICON” on page 25.
3. For switches with at least 256 ports installed, use the portDisable command to disable ports
0xFE and 0xFF. Refer to “Disabling ports 0xFE and 0xFF” on page 38 for more information.
switch.
Configuring FICON CUP
3
4. Enter the ficonCupShow fmsmode command to verify if FMS mode is disabled.
5. Take the appropriate action based on whether FMS mode is enabled or disabled:
• If FMS mode is disabled, proceed to step 6.
• If FMS mode is enabled, disable it using the ficonCupSet fmsmode disable command.
If FMS mode is already enabled, disabling it might be disruptive to operations because ports
that were previously prohibited from communicating will be allowed to do so because prohibits
are no longer enforced.
You must disable FMS mode to continue setting up CUP.
6. Install a CUP license on the switch. For more information on installing a license, refer to the
Fabric OS Administrator’s Guide.
7.En te r t h e ficonCupSet fmsmode enable command to enable FMS mode on the switch.
8. Enter the ficonCupSet modereg command to set the mode register bits. Refer to “Mode
register bit settings” on page 42 for more information.
9. Verify with the systems programmer that the CUP ports are defined properly.
FICON ports are defined as part of the sysgen process. The z/OS systems programmer can
verify if the CUP ports are defined properly.
FICON Administrator’s Guide37
53-1002753-01
Page 52
Determining physical port assignment
3
Disabling ports 0xFE and 0xFF
Addresses 0xFE and 0xFF are not available for external connections, so they must be disabled to
configure FICON CUP. Note that in logical switches or in cases where port address binding or port
swapping has occurred, addresses 0xFE and 0xFF are not necessarily tied to specific physical
ports, such as 254 and 255. After FMS mode has been successfully enabled, these two ports
remain disabled and cannot be used either as an F_Port or an E_Port. Because these ports are not
available after enabling FMS mode, you should first move any fiber connected to either of them to
another free port.
You must manually disable ports associated with 0xFE and 0xFF to enable FMS mode only if there
is an active device connected to these ports. Otherwise, if physical ports have been assigned these
addresses by default prior to enabling FMS, they will be automatically disabled when FMS is
enabled.
On a Brocade DCX and DCX-4S Backbones, when Virtual Fabrics is not enabled or zero-based
addressing is not used, you must disable physical ports associated with addresses 0xFE and 0xFF
(if ports were assigned by default) before enabling FMS mode. Note that physical ports associated
with 0xFE and 0xFF may be different in logical switches using zero-based addressing. Disable these
addresses manually if there is an active device attached. If an active device is not attached to the
ports, they will be disabled automatically when you enable FMS mode.
You can only disable ports if there is a card present in the slot.
If FMS mode is enabled in a logical switch with port-based addressing, the physical ports with
addresses 0xFE and 0xFF are essentially wasted and may be used better elsewhere. This is the
same for a switch with dynamic addressing (256-area addressing mode 1) where all available
addresses are used, and thus no addresses are available for binding to these ports. On logical
switches with dynamic addressing mode, such ports may be reclaimed for external connections by
binding a new address to the port, if unused addresses are available.
Ports that are disabled for having addresses “FE” or “FF” when FMS is enabled are not eligible for
port swapping.
Configuring FICON CUP in Virtual Fabrics
For details on FICON and FICON CUP in virtual fabrics, refer to “FICON and FICON CUP in Virtual
Fabrics” on page 26
Determining physical port assignment
The following instructions help you locate the physical port from CUP_PortAddress.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Set the context to the appropriate logical switch.
3. Enter the switchShow command.
4. Look for the PID in the Address column.
5. Read across to the Slot and Port columns.
6. Enter the portAddress
38FICON Administrator’s Guide
--findPID command.
53-1002753-01
Page 53
For more information and example output from this command, refer to the Fabric OS Command
NOTE
Reference Manual.
FMS mode and FICON CUP
Enabling FICON Management Server (FMS) mode in a Fabric OS environment allows the switch to
behave within the context of the FICON architecture and effectively creates a FICON director. For
Fabric OS 7.0 and later, a FICON CUP license is required to enable FMS mode. The FMS mode
setting can be changed whether the switch is disabled or enabled. If FMS mode is changed while
the switch is enabled, a device reset is performed for the control device and an RSCN is generated
with PID 0xDDFE00, where 0xDD is the domain ID of the switch.
Changing FMS mode from disabled to enabled triggers the following events:
• Access to switch configuration parameters is serialized.
• The active CUP configuration data, including the following, is initialized according to the IPL file:
-Port and switch names.
-Port “Block” and “Unblock” values.
-PDCM values.
• Brocade Advanced Zoning, if used, continues to be in force. If there are any differences
between the port-to-port connectivity restrictions defined by Brocade Advanced Zoning and
those defined by PDCMs, the most restrictive rules are automatically applied.
• RSCNs are sent to devices if PDCM changes result in changes to connectivity between a set of
ports that are zoned together.
• An RSCN for the CUP is generated to all mainframe FICON and FCP channels in the fabric.
FMS mode and FICON CUP
3
Changing FMS mode from enabled to disabled triggers the following events:
• A device reset is performed on the control device.
• Prohibits are no longer enforced.
• RSCNs might be generated to some devices if PDCM removal results in changes to connectivity
between a set of ports that are zoned together.
• If a given port was set to “Block” or “Unblock,” that port remains disabled or enabled.
• The switch interface for host in-band management is no longer enabled.
• An RSCN for the CUP is generated to all mainframe FICON and FCP channels in the fabric.
FMS mode cannot be enabled in a virtual fabric environment when a logical switch is using 10-bit
addressing mode.
FICON Administrator’s Guide39
53-1002753-01
Page 54
FMS mode and FICON CUP
3
Fabric OS command limitations and considerations
This section describes limitations and considerations for using Fabric OS commands with FMS
mode enabled. For a list of Fabric OS commands related to FICON, refer to FICON commands 12.
• The following Fabric OS commands return a “switch busy” response if they are issued when the
host-based management application is performing a write operation. This serialization
prevents interference from out-of-band management applications when a host-based
management program is being used to administer the switch.
bladeEnableswitchCfgPersistentDisable
portDisableswitchDisable
portEnableswitchEnable
portNameswitchName
portShowswitchShow
portSwap
• You cannot use the portCfgPersistentEnable and portCfgPersistentDisable commands to
persistently enable and disable ports. Refer to the procedure “Persistently enabling and
disabling ports for CUP” on page 46 for instructions.
• Specifying a range of ports for the portname command will fail except for the first port in the
range. FMS does not permit multiple ports with the same name.
Displaying FMS mode
Use the ficonCupShow fmsmode command to display the FMS mode enabled or disabled state for
the switch.
Enabling FMS mode
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter ficonCupSet fmsmode enable.
3. Enter the ficonCupShow fmsmode command to verify if FMS mode is enabled.
Disabling FMS mode
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter ficonCupSet fmsmode disable.
3. Enter the ficonCupShow fmsmode command to verify if FMS mode is disabled.
40FICON Administrator’s Guide
53-1002753-01
Page 55
FMS mode and FICON CUP
NOTE
3
FMS mode and FICON OxFE or OxFF ports
OxFE and OxFF are assigned to the CUP and cannot be associated with any physical port with FMS
mode enabled. Therefore, note the following when using FMS mode:
• You must manually disable ports associated with 0xFE and 0xFF using the portDisable
command to before you can enable FMS mode, but only if there is an active device connected
to these ports. Otherwise, these ports will disable automatically when FMS is enabled. Refer to
“Disabling ports 0xFE and 0xFF” on page 38 for more information
• Enabling FMS mode disables all ports with OxFE and OxFF addresses.
• If you enable FMS mode, you will get port-disable messages for physical ports with OxFE or
OxFF FICON addresses.
• You cannot enable FMS mode when physical ports have OxFE or OxFF FICON addresses with
live devices connected.
• Port swapping and blade swapping cannot be performed on ports where FMS mode is enabled
and ports have either OxFF or OxFE FICON addresses.
Upgrade considerations
Take the CUP offline before performing an upgrade or downgrade operation. The upgrade can
disrupt CUP exchanges, especially for active RMF sessions.
For Fabric OS 7.0 and later, a FICON CUP license is required to enable FMS mode. Consider the
following when using FMS mode and upgrading from Fabric OS v6.4 to v7.0 or later:
• If using Fabric OS v6.4 and FMS mode is enabled, but a FICON CUP license was never
installed:
-You can manage the port connectivity configuration through management applications
such as the Brocade Network Advisor and Web Tools.
-The Inband Management feature is not enabled.
-After a nondisruptive upgrade from Fabric OS v6.4.0 to v7.0.0 or later:
You can manage the port connectivity configuration using management programs
such as the Brocade Network Advisor and Web Tools. The Web Tools mode register is
accessible.
To enable Inband Management, you must install a FICON CUP license and disable,
then re-enable FMS mode.
The preceding information does not apply to the Brocade DCX-8510 models as a
nondisruptive upgrade is not possible.
• If using Fabric OS v6.4 and FMS mode is enabled, but a FICON CUP license was installed, then
removed:
-You can manage port connectivity through management applications such as the Brocade
Network Advisor and Web Tools. The FICON CUP Mode Register is accessible through Web
Tool s.
-Inband Management is enabled, providing you do not disable FMS mode, disable, then
re-enable the switch, or power-cycle the switch.
FICON Administrator’s Guide41
53-1002753-01
Page 56
Mode register bit settings
NOTE
3
After a nondisruptive upgrade from Fabric OS v6.4.0 to v7.0.0 or later, the following events
occurs:
These only occur on platforms released prior to Fabric OS v7.0.0.
-You can manage port connectivity using the Allow/Prohibit Matrix in Brocade Network
Advisor and Web Tools. Note that these applications are not available unless you install a
FICON CUP license. The Mode Register is accessible via Web Tools and the Fabric OS
commands.
-Inband Management is enabled, provided you do not disable FMS mode, disable then
enable the switch, or power-cycle the switch.
-If you disable FMS, a FICON CUP license is required to re-enable the feature.
Port swap limitations
Following are limitations for swapping ports or blades using CLI commands or Brocade Network
Advisor options. Note that some of these constraints apply to logical switches in general, and are
not necessarily limited to switches enabled for FMS mode.
• A port may be swapped with only one other port at a time.
• Ports cannot be swapped if one of the ports has a value of 0xFE or 0xFF in the Fibre Channel
Port ID (FC ID) field.
• The logical switch addressing mode can’t be changed if the logical switch contains swapped
ports.
• LSCFG operations cannot operate on swapped ports.
• You cannot swap ports that are in two different logical switches.
• Swapped Ports cannot be moved out of a logical switch.
• The portaddress - -bind command is not allowed on swapped ports.
• The wwnaddress - -bind command is not allowed on swapped ports.
Mode register bit settings
A mode register controls the behavior of the switch with respect to CUP itself, and with respect to
the behavior of other management interfaces. The mode register bits are described in Table 5.
TABLE 5FICON CUP mode register bits
ACPAlternate control prohibited. Because the Fabric OS CLI, Web Tools, and Brocade Network
Advisor are considered to be switch consoles, this bit has no effect on their operation. Attempts
to set CUP parameters through SNMP are denied when this bit is set on. The default setting is 1
(on).
ASMActive=Saved mode. When this bit is set on, all CUP configuration parameters are persistent,
meaning that they will be saved in nonvolatile storage in the IPL file that is applied upon a cold
reboot or a power cycle. The default setting is 1 (on).
DCAMSwitch clock alert mode. When this bit is set on, a warning is issued when the date,
tsClockServer, or tsTimeZone commands are entered to set the time and date on the switch. The
default setting is 0 (off).
42FICON Administrator’s Guide
53-1002753-01
Page 57
Mode register bit settings
3
TABLE 5FICON CUP mode register bits (Continued)
HCPHost control prohibited. When this bit is set on, the host is not allowed to set CUP parameters.
The default setting is 0 (off).
POSCProgrammed offline state control. If the bit is set (=1), the command from the host will set the
switch of fline. If the bit is off (=0), the command from the host will be rejected and the switch will
not go offline. The default setting is 1 (on).
UAMUser alert mode. When this bit is set on, a warning is issued when an action is attempted that
will write CUP parameters on the switch. The default setting is 0 (off).
The ficonCupShow modereg command displays the mode register bit settings for the switch. A
display of 0 indicates that the mode register bit is set to off; 1 indicates that the bit is set to on.
When the POSC bit=1, the FICON host is allowed to set the switch offline. The bit is set to 1 by
default. If you do not want to allow the host to have permission to set the switch offline, the bit
should be turned off. The state of this bit cannot be changed by the FICON host; it can only be
changed through the command line.
For more information on this command and an example of command output, refer to the Fabric OS Command Reference Manual.
To display the mode register bit HCP for the switch, enter the ficonCupShow command as shown in
the following example:
switch:admin> ficoncupshow modereg HCP
FICON file access facility
The FICON file access facility (FAF) stores configuration files. This includes the initial program load
(IPL) and other configuration files. The Fabric OS saves the IPL and all other configuration files on
the switch. A maximum of 16 configuration files, including the IPL file, are supported.
You can upload the configuration files saved on the switch to a management workstation using the
configUpload command. Data uploaded using the configUpload command also contains the IPL file
that was current at the time the command was executed.
Refer to the Fabric OS Administrator’s Guide for more information on the configUpload process.
Configuration files uploaded
When you execute the configUpload command, all the files saved in the file access facility are
uploaded to a management workstation (there is a section in the uploaded configuration file
labeled [FICU SAVED FILES] where the files are stored in an encoded format).
Configuration files downloaded with Active=Saved mode enabled
The contents of existing files saved on the switch, which are also present in the “FICON_CUP”
section, are overwritten.
The files in the [FICU SAVED FILES] section of the configuration file, which are not currently on the
switch, are saved on the switch.
The IPL will not be replaced because Active=Saved mode is enabled. A warning message is
displayed in the event log to warn users that the IPL will not be overwritten.
FICON Administrator’s Guide43
53-1002753-01
Page 58
Mode register bit settings
3
Configuration files downloaded with Active=Saved mode disabled
The contents of existing files saved on the switch, which are also present in the [FICU SAVED FILES]
section, are overwritten.
The files in the [FICU SAVED FILES] section of the configuration file, which are not currently on the
switch, are saved on the switch.
The IPL is replaced because Active=Saved mode is disabled.
Sample IOCP configuration file
The channel subsystem controls communication between a configured channel, the control unit,
and the device. The I/O Configuration Data Set (IOCDS) defines the channels, control units, and
devices to the designated logical partitions (LPARs) within the server; this is generated using the
Input/Output Configuration Program (IOCP). The IOCP statements are typically built using the
hardware configuration definition (HCD). The interactive dialog is used to generate your
Input/Output Definition File (IODF), invoke the IOCP program, and subsequently build your
production IOCDS.
Each FICON switch in a fabric must have a unique domain ID and a unique switch ID. The switch ID
used in the IOCP definitions can be any value between x’00’ to x’FF’. The domain ID range for
switches is hex x'01' to x'EF' or decimal 1 to 239. When defining the switch IDs in the IOCP
definitions, ensure that you use values within the FICON switch’s range.
The switch ID must be assigned by the user and must be unique within the scope of the definitions
(IOCP and HCD).
The domain ID is assigned by the manufacturer and can be customized to a different value. It must
be unique within the fabric.
Brocade recommends that the switch ID (in IOCP or HCD) be set to the same value as the domain
ID of the FICON switch, which is defined to the switch at installation time. This simplifies the
configuration and reduces confusion by having a common designation across all definitions.
For more information on switch numbering, refer to the IBM publication, FICON® Implementation Guide (SG24-6497-00).
In the following sample IOCP configuration file, the UNIT value for FICON CUP definitions is 2032 for
any FICON switch regardless of vendor or platform. All Brocade switches require UNIT=2032 for the
CUP definition. All Domain IDs are specified in hex values in the IOCP (and not in decimal values).
Consider the following when changing mode register bits:
• The UAM bit can only be set by host programming.
• All mode register bits except UAM are saved across power on/off cycles; the UAM bit is reset to
0 following a power-on.
• Mode register bits can be changed when the switch is offline or online. If the ACP or HCP bits
are changed when the switch is online, they will take effect any time between the completion of
the current command and the end of the CCW command chain (or the next alternate manager
operation).
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the ficonCupSet modereg command to set the FICON CUP mode register bits for the local
switch.
Example of setting the mode register bit HCP to off:
switch:admin> ficoncupset modereg HCP 0
Example of setting the mode register bit ACP to on:
switch:admin> ficoncupset modereg ACP 1
Setting the MIHPTO value
The missing interrupt handler primary timeout (MIHPTO) value determines how long the channel
will wait before timing out an expected response from the CUP. This value is part of the data
returned to the host when it reads the configuration data, but the host is under no obligation to
adhere to this value. The actual value that the host will use before detecting a timeout can be set
by the system programmer (overriding the value returned by the switch).
Use the ficonCupSet MIHPTO command to increase, decrease, or display the MIHPTO value that the
CUP returns to the host when configuration data is read. Changing the MIHPTO value requires that
the CUP control device be varied offline from the host, as the host will not automatically be notified
when the value is changed. The host will normally read the new value, using the Read Configuration
Data command, when the CUP is varied back online.
The MIHPTO setting will persist across reboots, POR, and failovers. Setting this value to the upper
end of the time range will allow the CUP more time to process requests during intervals of peak
usage, without causing the host to detect timeouts. It is recommended that the value be set to 180
seconds (default).
Complete the following steps to set the MIHPTO value.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the ficonCupSet MIHPTO command and set the value between 15 and 600.
FICON Administrator’s Guide45
53-1002753-01
Page 60
Persistently enabling and disabling ports for CUP
3
Persistently enabling and disabling ports for CUP
When FMS mode is enabled, you cannot use the portCfgPersistentEnable and
portCfgPersistentDisable commands to persistently enable and disable ports. Instead, use the
following procedure.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the ficonCupShow modereg command to display the mode register bit settings.
3. Verify that the ASM bit is set on (1).
4. If the ASM bit is set off (0), enter the ficonCupSet modereg asm 1 command to set the bit to
on.
For more information on this command and an example of command output, refer to the Fabric
OS Command Reference Manual.
5. Use the portEnable and portDisable commands to enable and disable ports as necessary.
The ports remain enabled or disabled after a switch reboot.
FICON extension over IP is supported between Brocade 7800 extension switches and FX8-24
blades.
• The Brocade 7800 switch has 16 physical Fibre Channel ports and six physical GbE ports.
• The FX8-24 blade has 12 physical Fibre Channel ports, 10 GbE ports, and two physical 10GbE
ports.
4
FICON emulation overview
FICON emulation supports FICON traffic over IP WANs using FCIP as the underlying protocol. FICON
emulation can be extended to support performance enhancements for specific applications
through use of the following licensed features:
• IBM z/OS Global Mirror (formerly eXtended Remote Copy or XRC)
• FICON Tape Emulation (tape read and write pipelining)
The 8 Gbps platforms (Brocade 7800 or FX8-24) use the following licenses:
• Advanced FICON Acceleration (FTR_AFA)
Allows interoperability for the following features and products:
-Prizm FICON to ESCON converter and ESBT Bus/Tag Interface Module for Prizm from
Optica Technologies, Inc. These products allow emulation and extended distance support
for bus and tag tapes and printers, such as 3203, 3211, 3800, 3900, 4248, and
intelligent printer data stream (IPDS) models.
-Write and read tape pipelining
-Tera d a t a emulation
-IBM z/OS Global Mirror emulation.
FICON Administrator’s Guide47
53-1002753-01
Page 62
FICON emulation overview
ATTENTION
4
The Advanced FICON Acceleration license contains the following options:
-Capacity 1
-Consumed 1
-Configured Blade Slots 1
This is a slot-based license for the Brocade FX8-24 and 7800.
• Advanced Extension (FTR_AE)
Required for multiple-circuit tunnels, FCIP Trunking, Adaptive Rate Limiting, and other FCIP
features. This is not required for FICON Acceleration features, but multiple circuits would be
used in a FICON emulation environment. This is a slot-based license for the Brocade FX8-24
and 7800.
-Capacity 1
-Consumed 1
-Configured Blade Slots 1
Use the licenseShow command to verify the licenses are installed on the hardware at both ends of
the FCIP tunnel. For slot-based licenses, use the licenseslotcfg --show command to display all
slot-based licenses in a chassis.
IBM z/OS Global Mirror emulation
The IBM z/OS Global Mirror emulation (formerly eXtended Remote Copy or XRC) application is a
direct-attached storage device (DASD) application that implements asynchronous disk mirroring, as
supported by the disk hardware architecture and a host software component called System Data
Mover (SDM). Within this application, a specific application channel program invokes a
well-constrained type of channel program called a Read Record Set (RRS) channel program. This
channel program is used to read only updates to disk volumes (record sets) after they have been
brought into synchronization with the aim of writing only record updates to a mirrored volume. The
RRS channel program accesses primary volumes from a remote host for the purpose of reading
these record sets (updates) supported by IBM z/OS Global Mirror emulation. The emulation feature
thus allows the primary volume to be located at a distance from its mirrored secondary without
encountering performance degradation associated with IU pacing.
48FICON Administrator’s Guide
53-1002753-01
Page 63
FICON emulation overview
44
Brocade
FICON
Directors
Brocade
extension switch
with emulation
for FICON
Brocade
extension switch
with emulation
for FICON
Brocade
FICON
Directors
System z
hosts
IBM Z/OS global
mirror system
data mover
Primary disk
volumes
Secondary disk
volumes
IP network
IP network
4
FIGURE 11IBM z/OS Global Mirror emulation
Figure 11 shows how the primary volume and the secondary mirrored volume may be
geographically distant across an IP WAN. Updates to the primary disk volumes are completed by
the production applications and then staged in a control unit cache to be read by the SDM. SDM
then writes the updates to the secondary volumes. The latency introduced by greater distance
creates delays in anticipated responses to certain commands. The FICON IU pacing mechanism
may interpret delays as an indication of a large data transfer that could monopolize a shared
resource and react by throttling the I/O. IBM z/OS Global Mirror emulation provides local responses
to remote hosts, eliminating distance-related delays. You can use the Brocade 7800 or an FX8-24
blade with FICON emulation.
For information on configuring IBM z/OS Global Mirror emulation or XRC emulation, refer to
“Configuring FICON emulation” on page 54. For information on displaying IBM z/OS Global Mirror
emulation status and statistics, refer to “Displaying FICON emulation performance statistics” on
page 55.
Tape emulation
Tape emulation (also called tape pipelining) refers to the concept of maintaining a series of I/O
operations across a host-WAN-device environment and should not be confused with the normal
FICON streaming of CCWs and data in a single command chain. Normally, tape access methods
can be expected to read data sequentially until they reach the end-of-file delimiters (tape marks) or
to write data sequentially until either the data set is closed or an end-of-tape condition occurs
(multi-volume file). The emulation design strategy attempts to optimize performance for sequential
reads and writes, while accommodating any other non-conforming conditions in a lower
performance non-emulating frame shuttle. Because write operations can be expected to comprise
the larger percentage of I/O operations for tape devices (for archival purposes), they are addressed
first.
FICON Administrator’s Guide49
53-1002753-01
Page 64
FICON emulation overview
Brocade
extension blade
with emulation
for FICON
FICON
Director(s)
Brocade
extension switch
with emulation
for FICON
Virtual Tape
Controller
Tape
Library
System z
hosts
Local acknowledgements
Write sequences are pre-acknowledged
to host and data is continuously sent
to the network
IP network
1
2
3
Final ending status is
presented to host
4a
Write data is streamed
to remote device
1a
2a
3a
Final ending status is
returned to local device
4
Local acknowledgement
Write commands
Write channel programs
are presented to
the tape device
1b
2b
3b
4
For information on configuring tape emulation, refer to “Configuring FICON emulation” on page 54.
For information on displaying tape emulation status and statistics, refer to “Displaying FICON
emulation performance statistics” on page 55.
Tape Write Pipelining
FICON Tape Write Pipelining (refer to Figure 12)improves performance for a variety of applications
when writing to tape over extended distances. FICON Tape Write Pipelining locally acknowledges
write data records, enabling the host to generate more records while previous records are in transit
across the IP WAN. If an exception status is received from the device, the writing of data and
emulation is terminated.
FIGURE 12Tape Write Pipelining
Tape Read Pipelining
50FICON Administrator’s Guide
FICON Tape Read Pipelining (Figure 13 on page 51) improves performance for certain applications
when reading from FICON tape over extended distances. FICON Tape Read Pipelining reads data on
the tape directly from the tape device. Reading of the tape continues until a threshold is reached.
The buffered data is forwarded to the host in response to requests from the host. When the host
sends the status accept frame indicating that the data was delivered, the read processing on the
device side credits the pipeline and requests more data from the tape. If an exception status is
received from the device, the reading of data and emulation is terminated.
53-1002753-01
Page 65
FICON emulation overview
Brocade
extension blade
with emulation
for FICON
FICON
Director(s)
Brocade
extension switch
with emulation
for FICON
Virtual Tape
Controller
Tape
Library
System z
hosts
Local host held off with command retry
awaiting read data to be delivered
Read emulation indication
sent to remote
Read command
IP network
1
3b
4b
5b
6b
Read commands are satisfied locally
2
3a
4a
5a
Read data is streamed to host
6a
Local response
Read command
Read channel programs
are presented locally
to the tape device
3
4
5
6
4
FIGURE 13Tape Read Pipelining
FICON and ESCON conversion support
The Brocade 7800 switch and FX8-24 blade can operate with the Prizm FICON to ESCON converter
and ESBT Bus/Tag Interface Module for Prizm from Optica Technologies, Inc. This allows FICON
tape emulation when connecting to ESCON and to bus and tag devices including 3480, 3490, and
3590 controllers. Discovery of these devices is automatic when FICON emulation is enabled on an
FCIP tunnel.
Printer emulation
The Brocade 7800 switch and FX8-24 blade can operate with a Prizm FICON to ESCON converter
and an ESBT Bus/Tag Interface Module for Prizm from Optica Technologies, Inc. This allows FICON
printer emulation when connecting to printers such as the 1043, 3203, 3211, 3800, 4245, 4248,
6262, Xeros Printer Access Facility (XPAF), and Intelligent Printer Data Stream (IPDS) models.
Discovery of these devices is automatic when FICON emulation is enabled on an FCIP tunnel.
For printer emulation, basic write streams are emulated to help resolve link latency resulting from
channel programming and associated FICON protocols. Because most supported printers will be
connected through the Prizm FICON to ESCON converter or ESBT Bus/Tag Interface Module,
resolving latency on the WAN is not required. The bottleneck will be the remote channel connection
where it is more important to have outstanding commands queued. Basic read commands known
to be part of the write command chain, typically those that read printer page counters, will be
handled locally without waiting for the actual read data from the device.
This emulation provides the following benefits:
• Optimized performance for FICON printers over distance when printers are extended through
an FCIP tunnel with FICON emulation.
FICON Administrator’s Guide51
53-1002753-01
Page 66
FCIP configuration requirements for FICON extension
4
• Intermediate ending status as appropriate to prevent FlCON channel protocol timeout (PTOV)
expirations.
• Acceptance of device status as required to prevent FICON control unit PTOV expirations.
• Prevention of “clutching” during print jobs.
For information on configuring printer emulation, refer to “Configuring FICON emulation” on
page 54. For information on displaying printer emulation status and statistics, refer to “Displaying
FICON emulation performance statistics” on page 55.
Teradata emulation
Teradata emulation reduces latency on links to Teradata warehouse systems caused by WAN
propagation delays and bandwidth restrictions. It accomplishes this by processing selected FICON
commands for associated control, data, and status responses. FICON Teradata emulation is only
supported between FICON channels and FICON Teradata controllers.
For write commands, control and status frames are generated for the host side of the WAN in order
to pipeline write commands over the same or multiple exchanges. Such pipelined write commands
and their data are queued at the device side of the WAN for asynchronous transfer to the device.
For read operations received by the device side of the WAN, a number of anticipatory read
commands are autonomously generated and transferred to the device. The data and status
associated with such commands are sent to the host side of the WAN and queued in anticipation of
host generated read commands.
FICON emulation for Teradata sequences over an FCIP tunnel is controlled by using FICON
emulation feature flags in the portcfg fciptunnel command associated with the FCIP tunnel.
For information on configuring Teradata emulation or tape emulation, refer to “Configuring FICON
emulation” on page 54. For information on displaying Teradata emulation status and statistics,
refer to “Displaying FICON emulation performance statistics” on page 55.
FCIP configuration requirements for FICON extension
FICON extension uses FCIP for transport. FCIP interfaces and tunnels used for FICON extension
must be defined prior to configuring FICON emulation. Ports should remain persistently disabled
until after FICON emulation is configured. Refer to the FCIP Administrator’s Guide for information
about configuring FCIP interfaces, tunnels, and circuits.
Configuration requirements for switches and Backbones
Consider these configuration issues when an extension switch or blade is connected to a switch in
a FICON configuration:
• If you are creating a cascaded configuration (connecting two switches or Backbones with
different domain IDs), be aware of IBM requirements for high integrity fabrics.
• In configurations with ISLs, ensure that ISL paths are properly configured to meet FICON
emulation requirements for a determinate path for FICON commands and responses.
• FICON networks with FCIP emulating and nonemulating tunnels do not support Dynamic Path
Selection (aptpolicy 3) configurations.
52FICON Administrator’s Guide
53-1002753-01
Page 67
Configuration requirements for switches and Backbones
4
• When running FICON emulation features over an FCIP tunnel, it is highly recommended that
both switches providing the FCIP tunnel are using the same Fabric OS release level.
High integrity fabric requirements for cascaded configurations
You can use extended FICON connections between or through Brocade 7800 Extension Switches or
FX8-24 blade to create a cascaded FICON switched configuration. The fabric must be set up to
provide a secure, or high integrity, fabric to enable this configuration to operate in an IBM
environment. See Chapter 2, “Administering FICON Fabrics” for an explanation of these
requirements.
FICON emulation requirement for a determinate path
FICON emulation processing creates FICON commands and responses on extended CHPIDs, and
intercepts all exchanges between a channel and a CU. For FICON Emulation processing to function
correctly, all the exchanges between a channel and CU must take a single tunnel path.
There are two ways to ensure a determinate path for FICON commands and responses:
• Define only one FCIP tunnel between sites.
• Use a Traffic Isolation zone (TI Zone) to assign a specific tunnel to Channel and control unit
ports. In the cases where multiple tunnels are required between a pair of FICON switches (or
Logical Switches), use TIZ to define ports that should use specific tunnels.
FCIP tunnel between sites
The Brocade 7800 and FX8-24 uses FCIP Trunking features to overcome the limitation of one
Ethernet interface, one IP address and one FCIP tunnel. In Fabric OS v6.3 and later an FCIP Tunnel
is created with multiple FCIP circuits over different IP interfaces to provide WAN load balancing and
failover recovery in the event of a limited WAN outage. This provides a highly redundant WAN
configuration for all FICON or FCP emulation technologies with Fabric OS. Figure 14 shows that a
tunnel can have up to 4 circuits per GbE port.
FIGURE 14FCIP tunnel and circuits
FICON Administrator’s Guide53
53-1002753-01
Page 68
Configuring FICON emulation
ATTENTION
NOTE
4
Configuring FICON emulation
Before you configure FICON emulation, you need to understand the available options, and whether
those options are to be implemented in your installation. If FICON emulation is new to you, refer to
“FICON emulation overview” on page 47.
Make sure both ends of the tunnel match before bringing it up or the tunnel will not form.
1. Verify that the following ports are opened in the IP network for FCIP:
• TCP 3225—FC Class F frames
• TCP 3226—FC Class 2 and 3 frames
• TCP 3227—Brocade proprietary IP path performance utility (ipPerf)
• IP protocol 6—Transmission Control Protocol
• IP protocol 108—IP Payload Compression Protocol (IPComp)
The 4 Gbps platform uses TCP port 3227 for ipPerf and IP protocol 108 for compression.
The 8 Gbps platform runs over the FCIP tunnel, not TCP/IP directly like ipPerf. No specific TCP
port is needed. Also the 8 Gbps platform compresses at layer 4 and no protocol is used.
2. Verify that the firmware is version v6.1.0e or later using the firmwareShow command.
3. Verify the correct licenses are installed using the licenseShow command. You can verify all
slot-based licenses in a chassis using the licenseslotcfg --show command.
4. Verify the switches have been set up according to the instructions in “Configuring switched
point-to-point FICON” on page 20 and “Configuring cascaded FICON” on page 25.
5. Configure the FCIP tunnel according to the FCIP Administrator’s Guide.
6. Enable compression on the FCIP tunnel.
7.Configure FICON emulation on the tunnel that is solution-specific. Refer to “Configuring
emulation on the 7800 switch and FX8-24 blade” on page 54.
8. If you have more than one tunnel being configured, take the following actions:
• If there is more then one tunnel to the same destination and their tunnel commit rates are
different, configure the link cost values the same.
• Configure TI Zones ensuring that if there are multiple E_Ports of different link costs, from
one switch to another switch within TI Zones, configure the link cost values to be the same.
Configuring emulation on the 7800 switch and FX8-24 blade
You can configure FICON emulation with such features as Teradata read, Teradata write, printer
emulation, tape write pipelining, tape read pipelining, TIN/TIR, device level ACK, and IBM z/OS
Global Mirror (formerly XRC) emulation enabled. Following are examples for enabling IBM z/OS
Global Mirror, Teradata emulation, tape read and write pipelining, and printer emulation.
Note that the preceding commands assume that an FCIP tunnel was a enabled for FICON operation
by using the --ficon option. Following are two examples:
4
• To create a empty tunnel as a FICON FCIP tunnel, use the following command:
portcfg fciptunnel 1/17 create -F -c 1
• To modify a tunnel to be FICON enabled and with hardware compression enabled, use the
following command:
portcfg fciptunnel 1/17 modify -c 1 -F 1
Displaying FICON emulation configuration values
For a Brocade 7800 or FX8-24 blade, use the portShow fciptunnel command to display FICON
configuration values. Command output shows which emulation features are enabled or disabled
and displays related emulation parameters for each tunnel.
For more information, refer to “Displaying FICON emulation performance statistics” on page 55.
Modifying FICON emulation
FICON emulation may be enabled on a per FCIP tunnel basis. For the Brocade 7800 switch and
FX8-24 blade, use the portCfg fciptunnel command to make modifications to the feature.
For details on command usage, refer to the Fabric OS Command Reference Manual.
You must bring the FCIP tunnel down in order to make changes. T his pro cess is disrup tive to all traf fic
on the tunnel.
1. Connect to the switches at both ends of the FCIP tunnel and log in using an account assigned
to the admin role.
2. Disable the FCIP tunnel at both ends where the FICON emulation feature is enabled.
3. Use the portCfg fciptunnel modify command for the Brocade 7800 switch or FX8-24 blade.
4. Make matching FICON emulation changes to both ends of the tunnel.
5. Enable the FCIP tunnel.
Displaying FICON emulation performance statistics
You can use the portShow command to view the performance statistics and monitor the behavior
of FICON emulation. You do not need to disable any feature to display statistics.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Use the portShow xtun command to display the statistics for the 7800 switch or FX8-24 blade.
FICON Administrator’s Guide55
53-1002753-01
Page 70
Displaying FICON emulation performance statistics
4
Refer to “FICON emulation monitoring” on page 56,
“+------------+----------------+-+-----+----+----+----+----+-----------+---+------+------+------+” on page 58, and the
Fabric OS Command Reference Manual for information on command formats that you can use
to display various FICON statistics.
FICON emulation monitoring
Use examples in this section to display emulation monitoring for extension blades and switches.
Use the -emul option for the portShow xtun command to display FICON emulation statistics.
Following is an example of using this command:
The -emul option displays statistics and status for printer, Teradata, tape, and IBM z/OS Global
Mirror Emulation (XRC) emulation.
IBM z/OS Global Mirror emulation (formerly eXtended Remote Copy or XRC) and tape statistics are
presented differently in output formats for the -emul option. The following elements are common to
both Tape Emulation and IBM z/OS Global Mirror Emulation outputs (refer to “Emulation statistics
example” on page 57 for an illustration of actual tape and XRC emulation statistics output.)
FDCB ptrA pointer to the FICON Device Control Block. Support personnel may use this
pointer.
PathThe device path, in the format VE-HD-HP-DD-DP-LP-CU-DV. where:
• VE is the internal VE_Port number.
• HD is the hex value for the Host Domain (the entry domain for this host
port into the fabric).
• HP is the hex value for the Host Port (the entry port of this host
connection into the fabric).
• DD is the hex value for the Device Domain (the entry domain for this
device into the fabric).
• DP is the hex value for the Device Port (the entry port for this device
connection into the fabric).
• LP is the Logical Partition (LPAR) value for the host accessing the specific
device.
• CU is the Control Unit number (CUADDR) for the specific FICON
connection.
• DV is the Device Number of this FICON connected device.
For example, the value 2463016406050001 breaks down as follows.
VEHDHPDDDPLPCUDV
2463016406050001
StateA value of zero indicates emulation is idle. Any non-zero value indicates
emulation is active.
56FICON Administrator’s Guide
53-1002753-01
Page 71
Displaying FICON emulation performance statistics
4
Emulation statistics example
Following is a tape and XRC statistics output example for the 7800 switch and FX8-24 blade.
Options for displaying statistics on 7800 switches and FX8-24 blades
Use options for the portshow xtun <slot/>ve_port -ficon -stats command to view statistics and
monitor the behavior of FICON emulation on 7800 switches and FX8-24 blades. You can use the
following options display status, performance statistics, and other statistics for tape, Teradata, IBM
z/OS Global Mirror (formerly XRC), and printer emulation. Refer to Tab le 6 for a description of
command formats for the various outputs.
TABLE 6FICON emulation statistics commands
CommandOutput display
portshow xtun [slot/]ve_port -ficon -emulStatistics and status for printer, Teradata, tape, and
IBM z/OS Global Mirror Emulation (XRC) emulation
portshow xtun [slot/]ve_port -ficon -emulxrcFICON XRC emulation statistics and status.
portshow xtun [slot/]ve_port -ficon -emultapeFICON tape emulation statistics and status.
portshow xtun [slot/]ve_port -ficon -emulprintFICON printer emulation statistics and status.
portshow xtun [slot/]ve_port -ficon -emulteraFICON Teradata emulation statistics and status.
portshow xtun [slot/]ve_port -ficon -statsGeneral statistics, including those for Teradata, XRC,
You can use either the CLI or Brocade Network Advisor to upgrade or downgrade your Fabric OS. Go
to http://www.brocade.com and follow the link under the Services and Support tab for Drivers and
Downloads to access firmware packages. Registered users can log in and follow the links for
downloads. New users can easily register and download release notes.
Firmware is downloaded in a .gz or .zip file. The Brocade Network Advisor has intelligence built-in
that unbundles the necessary files, making it unnecessary to unbundle the package. With both
Web Tools or the CLI, it will be necessary to unzip or ungunzip the package first.
For more information on the firmware download process, refer to the Fabric OS Administrator’s Guide.
Upgrade and downgrade considerations
Concurrent upgrades are generally supported from one major release level to the next, for example,
v6.4.0 to v7.0.0. As a precaution, always read the appropriate section in the Brocade Fabric OS
Release Notes before upgrading or downgrading firmware.
Firmware may need to be upgraded in multiple steps. Consult the Release Notes for details.
When downgrading firmware, turn off and disable all features not supported in the final release.
For more information on the firmware download process, refer to the Fabric OS Administrator’s Guide.
FICON Administrator’s Guide59
53-1002753-01
Page 74
Firmware management in a FICON environment
NOTE
5
Firmware upgrade disruption
Loading new firmware is a non-disruptive process except for the following cases:
• During the firmware download process a failover to the backup processor occurs. Any
outstanding CUP commands are lost and result in a timeout. To avoid this, CUP should be
varied offline before starting a firmware download. The CUP may be varied back online after a
successful firmware download.
• All firmware downloads will cause an I/O disruption when the new code is activated on the
FX8-24 blade and 7800 switch.
If upgrading multiple switches or blades, upgrade one at a time. This is not a requirement for DCX
or DCX-4S platforms.
Non-disruptive firmware upload and download
All systems maintain two partitions of nonvolatile storage areas, a primary and a secondary.
Following is an overview of the firmware download process:
1. The firmware is downloaded to the secondary partition.
2. The current secondary and primary partitions are swapped.
3. The CP then reboots. In dual CP systems (Brocade 48000 and Brocade DCX enterprise-class
platforms) the backup CP is rebooted with the new firmware and the HA failover is
automatically invoked.
4. Once rebooted, the primary code image is copied to the secondary.
Considerations for fixed port switches
Unlike director-class products, fixed port switches have only one control processor (CP) so they
behave differently during Fabric OS hot code upgrades. This can impact FICON operations if best
practice is not followed.
During the code upload process, there is a period of time when the CP is busy rebooting the
operating system and re-initializing Fabric OS, and it is not available to respond to CPU-targeted
queries/frames. For dual-CP products, this “busy” time is a couple of seconds (the time it takes for
mastership to change), and is non-disruptive. For single-CP fixed port switches, however, the “busy”
period can last a minute or two. During this time, data traffic will continue to flow uninterrupted,
unless a response from the CP is required. If a response from the CP is required, for path and
device validation for instance, Interface Control Checks (IFCCs) are possible. Any IFCCs should be
recovered automatically by the System z host, with no required user action.
This does not mean that switches are not appropriate for FICON. Fully certified FICON switch
deployments will continue to be sold and supported. However it is important to follow best
practices of performing the firmware upgrade during scheduled maintenance windows where
traffic is minimized, IFCCs can be tolerated, and CUP is turned off.
60FICON Administrator’s Guide
53-1002753-01
Page 75
Configuration restoration in a FICON environment
NOTE
Configuration restoration in a FICON environment
If the switch is operating in a FICON CUP environment, and the ASM (Active=Saved) bit is set on,
then the switch ignores the IPL file downloaded when you restore a configuration. Table 7
describes this behavior in more detail.
TABLE 7Backup and restore in a FICON CUP environment
ASM bitCommandDescription
on or offconfigUploadAll the files saved in the file access facility are uploaded to the
management workstation. A section in the uploaded configuration file
labeled FICON_CUP is in an encoded format.
onconfigDownloadFiles saved on the switch that are also present in the FICON_CUP
section of the configuration file are overwritten.
Files in the FICON section of the configuration file that are not
currently present on the switch are saved.
The IPL file is not replaced, because Active=Saved mode is on.
offconfigDownloadFiles saved on the switch that are also present in the FICON_CUP
section of the configuration file are overwritten.
Files in the FICON section of the configuration file that are not
currently present on the switch are saved.
The IPL file is replaced, because Active=Saved mode is off. A warning
message is displayed in the syslog to warn that the IPL file is not being
overwritten.
5
If FMS mode is enabled in a configuration file, but is disabled on the switch, the configDownload
command fails and displays an error message. This prevents undesirable conditions that could
result from enabling fmsmode on a switch that does not require it.
Traffic Isolation Zoning
Traffic Isolation (TI) Zoning allows data paths to be specified. In a FICON fabric, you need Fabric OS
v6.2.0 or later. TI Zoning has the following benefits:
• Separates direct attached storage device (DASD) and tape traffic.
• Selects traffic for diverse ISL routes (typical of dual fiber networks).
• In conjunction with long-distance channel extension equipment (FICON Acceleration),
guarantee bandwidth for certain mission-critical data.
Creating logical switches is preferred over creating TI Zones for FICON emulation configurations.
Logical switches can be created easily and they naturally separate disk and tape traffic to different
tunnels and network bandwidth.
For more information on Traffic Isolation Zoning, refer to the Fabric OS Administrator’s Guide.
Session-based zoning enforcement is not recommended on a FICON switch. For more information
on session-based zoning enforcement, refer to the Fabric OS Administrator’s Guide.
FICON Administrator’s Guide61
53-1002753-01
Page 76
Traffic Isolation Zoning
5
Determining ports for the TI Zone
To find the device ports, use the MVS display command. The following example will display the
paths to device 1002 and where it is connected:
D M=DEV(1002)
Response is:
DEVICE 1002 STATUS=OFFLINE
CHP AC AD AE AF
ENTRY LINK ADDRESS .. 6715 6808 6679
DEST LINK ADDRESS 66FE 66FE 66FE 66FE //CUP has link address = DDFE,
where DD= domain ID of switch//
PATH ONLINE Y Y Y Y
CHP PHYSICALLY ONLINE Y Y Y Y
PATH OPERATIONAL Y Y Y Y
MANAGED N N N N
CU NUMBER 1002 1002 1002 1002
MAXIMUM MANAGED CHPID(S) ALLOWED: 0
DESTINATION CU LOGICAL ADDRESS = 00
SCP CU ND = SLKWRM.DCX.BRD.CA.4FT00X39C19A.0001
SCP TOKEN NED = SLKWRM.DCX.BRD.CA.4FT00X39C19A.0000
SCP DEVICE NED = SLKWRM.DCX.BRD.CA.4FT00X39C19A.0000
Enhanced TI Zoning
Enhanced Traffic Isolation Zones allow you to specify a preferred and exclusive path through a
cascaded fabric for a particular flow by defining a set of devices or ports to appear in more than
one Enhanced TI Zone. A preferred path is one that has failover enabled and an exclusive path has failover disabled. An exclusive path in the following topologies is beneficial as it will limit the
number of hops between switches:
• Triangular
• System Data Mover
• Emulation
Enhanced TI Zoning for the topologies listed above is available only on the following platforms
running Fabric OS v6.4.0 or later:
• Brocade 5100
• Brocade 5300
• Brocade 7800
• Brocade DCX
• Brocade DCX-4S
Triangular topology
The triangular topology shown in Figure 15 is a common FICON deployment for managing data
center recovery functions. In this topology, a switch is located in each of three sites and connected
to the other two forming a triangle. In most environments, the purpose of this configuration is to
give the host application the ability to connect to storage media at the other two sites, which allow
it to recover from an outage at either site. Meanwhile, the storage systems constantly synchronize
between the two storage sites.
62FICON Administrator’s Guide
53-1002753-01
Page 77
Channel
Traffic Isolation Zoning
5
Flow A
ISL 1ISL 2
ISL 3
Flow B
Flow C
CU BCU A
FIGURE 15Typical triangular setup
During normal operation, the routes from any device on one switch are a one-hop path to either of
the other two switches. However, if the path between any pair of switches fails then a two-hop path
is available. Multi-hop paths are not certified for FICON configurations.
To get around the multiple-hop issue, use Enhanced TI Zoning to prohibit the connection between
E_Ports on the switches. For this setup you must define a device in two zones with failover
disabled. Figure 16 shows that the setup allows a source device to reach two separate destinations
over exclusive, one-hop paths, while preventing the source device from taking the two-hop path in
the event of a failure.
FICON Administrator’s Guide63
53-1002753-01
Page 78
Traffic Isolation Zoning
5
Channel
Flow A
ISL 1ISL 2
ISL 3
Flow B
Flow C
CU BCU A
FIGURE 16Triangular topology with Enhanced TI Zones
Figure 16 shows the following TI Zone definitions for the triangular topology:
• Blue Zone includes the channel, ISL 1, and CU A.
• Green Zone includes the channel, ISL 2, and CU B.
• Red Zone includes the CU A, ISL 3, and CU B.
Using these definitions with failover disabled, the topology achieves the desired behavior. The
channel can reach CU A only over ISL 1 and CU B only over ISL 2. If either ISL should fail,
connectivity to the destination with the failed ISL is prevented because the exclusive path has
failed.
64FICON Administrator’s Guide
53-1002753-01
Page 79
Traffic Isolation Zoning
5
System Data Mover topology
Figure 17 shows a cascaded topology which can be used to support an operations recovery site. In
this environment, host and storage systems exist at both sites, which are connected by two
switches cascaded together using the System Data Mover (SDM) application. The storage differs
between the sites. In the production site, the storage is DASD to support fast access for production
applications. In the backup site, the storage is tape to support archival operations for system
restoration.
FIGURE 17Typical System Data Mover topology
The host in the production site operates the SDM application that reads data from the DASD and
writes it to the remote tape. The path to the remote site is required to be isolated to support the
bandwidth requirements needed to write the data to the remote tape. The host in the backup site
needs access to the DASD in the production site to support a hot recovery scenario in the event the
production host goes down. Supporting these two functions requires the ability to define
overlapping zones that isolate the tape traffic from the DASD traffic and allow local access to the
DASD ports.
FICON Administrator’s Guide65
53-1002753-01
Page 80
Traffic Isolation Zoning
5
FIGURE 18System Data Mover topology using Enhanced TI Zones
Figure 18 on page 66 shows the following Enhanced TI Zones definitions for the System Data
Mover topology:
• Green Zone includes Host A, ISL 1, and CU B.
• Orange Zone includes CU A, ISL 2, and Host B.
• Red Zone includes Host A and CU A.
• Blue Zone includes Host B and CU B.
Using these definitions with failover disabled achieves the desired behavior. The production
channels can access the local DASD and use an exclusive path to write data to the remote tape.
The remote channels can access the remote tape and use an exclusive path to write data to the
production DASD. In both cases, the production tape traffic is excluded from intruding on the
backup DASD traffic and vice versa.
66FICON Administrator’s Guide
53-1002753-01
Page 81
Traffic Isolation Zoning
5
Emulation topology
FICON device emulation topologies have unique requirements due to the characteristics of the
emulation devices. Each device must track the emulated sessions and maintain the state of the
emulated devices in order to satisfy both ends of the connection. Since this is done over very long
distances, care must be taken when configuring the ISL connections to insure that the sessions
remain along exclusive paths. By maintaining an exclusive path between the emulation units, the
recovery scenarios become manageable and allow system level mechanisms to take precedence,
that is the device level recovery at the channel or control unit.
Generally, channel to control unit access is deployed as one-to-many topology; where, one channel
is defined to access many control units. FICON environments, however, are deployed in multiple
redundant configurations that also define access from one control unit to many channels. When
emulation is involved these configurations can result in overlapping channel and control unit
connections that make emulation isolation difficult. By using the Enhanced TI Zoning feature to
support device membership in multiple zones, these types of emulation configurations are
addressed.
FICON Administrator’s Guide67
53-1002753-01
Page 82
Traffic Isolation Zoning
5
CPU1
0
4041
1
TI Zone 1
TI Zone 2
81
80
1E1E0F1E
Grid 1
82
83
1B1A
Grid 4
FIGURE 19Emulation topology using Enhanced TI Zones
Figure 19 on page 68 shows the following Enhanced TI Zoning definitions for the emulation
topology:
• Green TI Zone includes channel 0, ISL 80, ISL 81, Grid 1, and Grid 4.
• Blue TI Zone includes channel 1, ISL 82, ISL 83, and CU Grid-4.
The unique characteristic of these zones is that each channel is defined with two control unit
device ports and one of the control unit device ports, CU Grid-4, is in both zones. In a traditional TI
Zone definition, these zones would have to be combined to provide the appropriate connectivity,
but then in the event of an ISL failure it would lead to a multi-hop, multi-emulation path that would
severely affect performance. Multi-hop paths are not certified for FICON configurations.
68FICON Administrator’s Guide
53-1002753-01
Page 83
Port fencing
NOTE
Port fencing
Occasionally, bad optics and cables can cause errors to occur at a rapid rate that error processing
and sending and processing RSCNs can cause fabric performance problems. Port fencing allows
the user to limit the number of errors a port can receive by forcing a port offline when certain error
thresholds are met.
The port fencing feature is configured through Fabric Watch. For more information on configuring
Port Fencing, refer to the Fabric Watch Administrator’s Guide.
5
Defining port fencing
Following is an overview of steps needed to define port fencing.
1. (Optional) Clear all alarms.
2. Define threshold levels.
3. Define alarm action.
4. Activate alarming.
Establish a Telnet session with a tool such as Putty that allows the columns to be increased. This is
because some of the displays use more than the standard 80 columns that programs such as
HyperTerminal support. Recommended number of columns is 120.
Settings for FICON environments
For typical FICON environments, port fencing is usually only set for CRC errors and Invalid Words.
The default of 1,000 errors per minute is a little high for CRC errors and Invalid Words. A more
common setting is 50 errors per minute. This is high enough to ignore occasional errors and
transient errors due to re-cabling but low enough to stop problematic optics from causing fabric
issues.
By default, the alarms are set to fence the port, log an alert, send an e-mail, and set an SNMP trap.
In most FICON environments, only fencing the port and logging the alert are desired.
The following are the default thresholds:
• Low 0
• High 1,000 errors per minute
FICON information
You can display link incidents, registered listeners, node identification data, and FRU failures, as
described in the following sections.
Link incidents
The registered link incident record (RLIR) ELS contains the link incident information sent to a
listener N_Port.
FICON Administrator’s Guide69
53-1002753-01
Page 84
FICON information
5
To display link incidents, connect to the switch, log in as user, and enter one of the following
commands:
• For the local switch: ficonShow rlir
• For all switches defined in the fabric: ficonShow rlir fabric
Identifying ports
The ficonShow rlir command displays, among other information, a tag field for the switch port. You
can use this tag to identify the port on which a FICON link incident occurred. The tag field is a
concatenation of the switch domain ID and port number, in hexadecimal format. The following
shows the top portion of command which shows a link incident for the switch port at domain ID
120, port 93 (785d in hex):
switch:admin> ficonshow rlir
{
{Fmt Type PID Port Incident Count TS Format Time Stamp
0x18 F 785d00 93 1 Time server Thu Apr 22 09:13:32 2004
Port Status: Link not operational
Link Failure Type: Loss of signal or synchronization
Registered Port WWN Registered Node WWN Flag Node Parameters
For complete examples of output from this command, refer to the ficonShow command section of
the Fabric OS Command Reference Manual
Registered listeners
To display registered listeners for link incidents, connect to the switch, log in as user, and enter one
of the following commands:
• For the local switch: ficonShow lirr
• For all switches defined in the fabric: ficonShow lirr fabric
For an examples of local LIRR database output, refer to the ficonShow command section of the
Fabric OS Command Reference Manual.
When you configure an LIRR entry and it is current, it is marked with -C. If the Management Server
marks an entry as current based on order, it is marked with -C. If the Management Server marks
an entry as secondary, it is marked with -S.
If the LIRR is configured then the <XX> indicates a valid port number. Otherwise, the line will
display the following:
Current LIRR device port number is not configured.
Node identification data
To display node-identification data, connect to the switch, log in as user, and enter any of the
following commands:
• For the local switch: ficonShow switchrnid
70FICON Administrator’s Guide
53-1002753-01
Page 85
FICON information
5
• For all switches defined in the fabric: ficonShow switchrnid fabric
• For all devices registered with the local switch: ficonShow rnid
• For all devices registered with all switches defined in the fabric: ficonShow rnid fabric
FRU error reporting
FRU failure information is no longer stored on the switch. FRU error information, such as FRU
failures and fan and power supply removals and insertions, are stored in error reports as sense
data. All FRUs in a chassis have an assigned number that is included in this data. Each CUP
instance retains a database of ten FRU error reports. All error reports are sent to host LPARs that
have appropriate Current Reporting Paths established with the CUP.
Error reports are added to the database as they occur, then they are removed when sent. If the
database fills up, new reports are discarded and not delivered.
Error data
Following are some examples of error data available from the host. The number identifying the
failed Brocade FRU (refer to “FRU numbers” on page 72) is derived from a single bit from byte 29
as the high order bit and 4 bits from byte 30 or 31 as the low order bits (refer to the data in bold in
the second lines of the following examples). The FRU numbers are decoded for display as the FRU
names, which display in the third lines of the following examples.
Following are the Brocade FRU numbers provided for error reports and associated FRU names that
display in error reports. Note that the FRU number does not actually display in the error reports, but
its associated FRU name does display.
• 1 = SLOT1
• 2 = SLOT2
• 3 = SLOT3
• 4 = SLOT4
• 5 = SLOT5
• 6 = SLOT6
• 7 = SLOT7
• 8 = SLOT8
• 9 = SLOT9
• 10 = SLOT10
• 11 = SLOT11
• 12 = SLOT12
• 17 = PWR1
• 18 = PWR2
• 19 = PWR3
• 20 = PWR4
• 24 = FAN1
• 25 = FAN2
• 26 = FAN3
• 30 = CHASSIS
• 31 = WWN-CRD
Swapping port area IDs
If a port malfunctions, or if you want to connect to different devices without having to rewire your
infrastructure, you can move a port’s traffic to another port (swap port area ID) without changing
the I/O Configuration Data Set (IOCDS) on the mainframe computer.
To swap port area IDs, refer to the “Swapping port area IDs” procedure in the Fabric OS Administrator’s Guide.
For more detail on the following commands that you will use to swap port area IDs, refer to the
Fabric OS Command ReferenceManual:
• portSwapEnable
• portDisable
• portSwap
• portSwapShow
• PortSwapDisable
72FICON Administrator’s Guide
53-1002753-01
Page 87
Important notes
NOTE
Consider the following when swapping port area IDs:
• Ports that have been swapped cannot be moved to another logical switch.
• Ports with area OxFE or OxFF addresses cannot be swapped when FMS mode is enabled.
• You can use the portSwapShow command to display information about swapped ports in a
switch.
• You can use the portSwap command to disable the port-swapping feature. You cannot use the
portSwap command after this feature is disabled. The enabled state of the port-swapping
feature is persistent across reboots and power cycles. Enabling and disabling the
port-swapping feature does not affect previously executed port-swapping operations.
Blade swapping
Blade swapping allows you to swap one blade with another of the same type; in this way, you can
perform a FRU replacement with minimal traffic disruption. The entire operation is accomplished
when the bladeSwap command runs on the Fabric OS. The Fabric OS then validates each
command before actually implementing the command on the enterprise-class platform. If an error
is encountered then blade swap quits without disrupting traffic flowing through the blades. If an
unforeseen error does occur during the bladeSwap command, an entry will be made into the
RASlog and all ports that have been swapped as part of the blade swap operation will be swapped
back. On successful completion of the command, the source and destination blades are left in a
disabled state allowing you to complete the cable move.
Blade swapping
5
Blade swapping is based on port swapping and has the same restrictions:
• Shared area ports cannot be swapped.
• Ports or blades with area OxFE or OxFF addresses cannot be swapped when FMS mode is
enabled.
• Ports that are part of a trunk group cannot be swapped.
• GbE ports cannot be swapped.
• Swapping ports between different logical switches is not supported. The ports on the source
and destination blades need to be in the same logical switch.
• Undetermined board types cannot be swapped. For example, a blade swap will fail if the blade
type cannot be identified.
• Blade swapping is not supported when swapping to a different model of blade or a different
port count. For example, you cannot swap an FC8-32 blade with an FC8-48 port blade.
• Blade swapping is not supported on application blades.
Only the FC ports will be swapped on an FX8-24 blade. GE and VE ports will not be swapped.
For specific limitations for port swapping, refer to “Port swap limitations” on page 42.
For procedures and more information on blade swapping, refer to the Fabric OS Administrator’s Guide.
FICON Administrator’s Guide73
53-1002753-01
Page 88
Common FICON issues
5
Common FICON issues
Any information you need to verify that FICON has been set up correctly can be found in the Chapter
2, “Administering FICON Fabrics” or Chapter 3, “Configuring FICON CUP”.
SymptomCascading: No path – Entry Link is “..”
Probable cause and recommended action
This typically occurs when one or more of the required high integrity features for cascading have
not been met. Check the following:
• Review each step in “Configuring switched point-to-point FICON” on page 20 and “Configuring
cascaded FICON” on page 25 and validate that all features are configured properly.
• Validate that CUP is online for each Backbone or switch in the fabric with the data center’s
system programmer.
SymptomPath to a device does not come online
Probable cause and recommended action
Make sure binding or zoning does not prevent the connection. Validate the following with the data
center's system programmer:
• Correct switch addresses (not switch IDs) are used in Link statements. Remember that the
switch address is the domain ID converted to hex and may have an offset.
• The Switch ID, used in the CHPID statements, was defined with the correct switch address.
• All addressing (LINK statements) uses 2-byte addressing on the channel if 2-byte addressing
has been previously used on that channel. This is necessary even when the specified path is
not cascaded.
SymptomPackets are being dropped between two FICON units.
Probable cause and recommended action
When planning cable needs, the following criteria must be considered:
• Distance considerations
• Fiber Optics Sub Assembly (FOSA) type (SW or LW)
• Cable specifications (SM or MM)
• Patch Panel Connections between FOSA ports (link loss .3-5 dB per)
• Maximum allowable link budget (dB) loss
From a cabling point of view, the most important factor of a Fibre Channel link is the selection of
the Fibre Optical Sub Assembly (FOSA) and matching cable type, to support the required distance.
Both ends of the optical link must have the matching FOSA (SFP) types.
74FICON Administrator’s Guide
53-1002753-01
Page 89
SymptomChannel path with 2-byte addressing does not come online
Probable cause and recommended action
Ask the operator what CHPID he is attempting to bring online. In FICON environments, only the
egress ports are defined in the IOCDS. The CHPID number of the channel appears in the node list,
but only if the channel path has logged in. If the channel path will not come online, it may not have
completed a Fibre Channel fabric login. If this is the case, you will have to rely on customer
documentation to determine what switch port the channel is connected to.
Next, ask the operator what type of error message they are getting. An IOS error message is
reported when there is an error external to the host; an IEE error message indicates something is
wrong internal to the host.
Fabric Not Configured Properly
• Make sure that the insistent domain ID is set and the domain IDs are unique for all switches in
the fabric.
• Make sure that the same binding method is used on all switches in the fabric.
• Scroll through the node list and make sure there are no invalid attachments.
• Make sure that the re-routing delay is turned off.
Cabling
Common FICON issues
5
Validate that cables are connected where intended.
When using multimode fiber, make sure that all fiber is either all 50u or all 62.5u. A common
mistake is to mix 50u and 62.5u fiber cables, which causes errors at patch panels.
Example of a Boxed Channel
In this example, the customer complained that the CHPIDs with 2-byte addressing did not come
online. He was instructed to configure the CHPID offline and then try to configure the CHPID back
online.
MVS console commands and responses (lines beginning with “IEE” are responses and all other
lines are entries.):
CF CHP(60),OFFLINE
IEE503I CHP(60),OFFLINE
IEE712I CONFIG PROCESSING COMPLETE
CF CHP(60),ONLINE
IEE754I NOT ALL PATHS BROUGHT ONLINE WITH CHP(60)
IEE502I CHP(60),ONLINE
IEE712I CONFIG PROCESSING COMPLETE
The line below displays the status of the CHPID with the “D M =” command. All other lines are
responses:
D M=CHP(60)
IEE174I 03.29.45 DISPLAY M 021
CHPID 60: TYPE=1D, DESC=FICON INCOMPLETE, ONLINE
DEVICE STATUS FOR CHANNEL PATH 60
0 1 2 3 4 5 6 7 8 9 A B C D E F
006 . $@ . . . . . . . . . . . . . .
SWITCH DEVICE NUMBER = NONE
ATTACHED ND = NOT KNOWN
PHYSICAL CHANNEL ID = 01F0
************************ SYMBOL EXPLANATIONS ************************
FICON Administrator’s Guide75
53-1002753-01
Page 90
Troubleshooting FICON
5
+ ONLINE @ PATH NOT VALIDATED - OFFLINE . DOES NOT EXIST
* PHYSICALLY ONLINE $ PATH NOT OPERATIONAL
The responses “NOT ALL PATHS BROUGHT ONLINE” after attempting to configure the CHPID online
and “FICON INCOMPLETE” after the display command indicate that the channel did not come
online. This is determined because the IEE error types are shown and combined with validating that
there was no attempt to do a basic FLOGI (no light displayed on the port view (LED) and no Fibre
Channel frames were received).
SymptomThe Control Unit Port cannot access the switch.
Probable cause and recommended action
A 2-byte CHPID (channel path identifier) link is defined using a domain and port ID that must
remain consistent. Any change in the physical link, such as a change in a domain or port ID, will
prevent storage control unit access.
Use the configure command to verify and set the insistent domain ID (IDID) parameter.
FICON:admin> configure
For more information and examples of using this command to set the IDID parameter, refer to the
Configure command in the Fabric OS Command Reference Manual.
SymptomUnable to “vary online” FICON CUP port on the switch.
Probable cause and recommended action
Use the haFailover command on an enterprise class chassis. This the only known fix as there is no
known firmware solution.
Troubleshooting FICON
This section provides information gathering and troubleshooting techniques necessary to fix your
problem.
General information to gather for all cases
The following information needs to be gathered for all FICON setups:
• The output from the standard support commands (portLogDump, supportSave, supportShow)
the Fabric Manager Event Log, EFCM or Brocade Network Advisor logs.
76FICON Administrator’s Guide
53-1002753-01
Page 91
Troubleshooting FICON
By default, the FICON group in the supportShow output is disabled. To enable the capture of
FICON data in the supportShow output, enter the supportShowCfgEnable ficon command.
After you get confirmation that the configuration has been updated, the following will be
collected and appear in the output for the supportShow command:
5
-ficonCupShow fmsmode
-ficonCupShow modereg
-ficonDbg dump rnid
-ficonDbg log
-ficonShow lirr
-ficonShow rlir
-ficonShow rnid
-ficonShow switchrnid
-ficuCmd dump -A
• Type of mainframe involved. Need make, model, and driver levels in use.
• Type of actual storage array installed. Many arrays will emulate a certain type of IBM array and
we need to know the exact make, model, and firmware of the array in use.
• Other detailed information for protocol-specific problems:
-Port data structures, displayed using the ptDataShow command.
-Port registers, displayed using the ptRegShow command.
• Read Brocade Release Notes for specific version information regarding the Fabric OS installed.
The following needs to be performed to troubleshoot all FICON setups:
• Perform a configUpload to gather configuration files. Provide the IOCDS mainframe file.
This will define how all mainframe ports are configured.
• Verify that Dynamic Load Sharing (DLS) has been disabled with the dlsReset command.
If DLS is enabled, traffic on existing ISL ports might be affected when one or more new ISLs is
added between the same two switches. Specifically, adding the new ISL might result in
dropped frames as routes are adjusted to take advantage of the bandwidth provided. By
disabling DLS, you ensure that there will be no dropped frames. In a supportshow, search for
route.stickyRoutes and check for a value of 1 or enter the dlsShow command.
• Verify that IOD is enabled using the iodSet command to ensure in-order delivery.
In the output from the supportShow or supportSave, search for the route.delayReroute and
check for a value of 1 as this indicates that the feature is turned on. Alternately, you can enter
the iodShow command.
Switched point-to-point topology checklist
This checklists lists items you should perform in your FICON environment to ensure proper
functionality of the feature:
• Does your Brocade switch or Backbone have Fabric OS v6.1.0 or later release installed?
• Are you using an appropriate management tool such as EFCM v9.6x, DCFM, or Brocade
Network Advisor?
• Have you verified that no additional license is required to enable your FICON configuration?
FICON Administrator’s Guide77
53-1002753-01
Page 92
Troubleshooting FICON
NOTE
ATTENTION
5
• Have you verified that your switch does not have a special mode setting turned on for FICON?
There is no requirement to have a secure fabric in a switched point-to-point topology.
Brocade Advanced features software package (Trunking, Fabric Watch, Extended Fabric) license
activation is required.
Cascaded topology checklist
This checklist lists items you should perform in your FICON environment to ensure proper
functionality of the feature:
• Verify that your Brocade switch or directory have Fabric OS v6.1.0 or later release installed?
• Verify that you are using an appropriate management tool such as EFCM v9.6x, DCFM, or
Brocade Network Advisor?
• Verify that no additional license is required to enable your FICON configuration?
• Verify that your switch does not have a special mode setting turned on for FICON?
In addition to port-based routing, lossless, and in-order frame delivery (IOD), it is
recommended that you enable the dynamic load-sharing feature.
• Verify that if you have a fabric for intermix mode of operations, separate zones for FICON and
FCP devices are set?
• Verify the mainframe channel device connectivity rule of maximum one hop is applied to both
FCP and FICON devices?
• Verify the Insistent domain ID flag is set to keep the domain ID of a fabric switch persistent?
• Verify the CHPID link path is defined using the 2-byte address format?
• Verify FICON channel connectivity to storage CU port has not exceed one hop?
• Verify that the security policies are enabled? If not, refer to “Configuring cascaded FICON” on
page 25.
The Switch Connection Control (SCC) security policy must be active.
Brocade Advanced features software package (Trunking, Fabric Watch, Extended fabric) license
activation is required.
Gathering additional information
In addition to the information you would collect for your support provider mentioned in the Fabric
OS Troubleshooting and Diagnostics Guide, also gather the following information which is specific
to FICON:
• Is this case logged during an initial install or has this environment been working prior to the
initial problem?
• What was changed immediately prior to the issue occurring?
• Is the switch properly configured for a FICON environment?
78FICON Administrator’s Guide
53-1002753-01
Page 93
Also refer to the most recent version of the Fabric OS Release Notes for notes on FICON setup
NOTE
and configuration.
• Is this a switched point-to-point or cascaded environment?
• Is the FICON group enabled for supportshow?
Check at the top of the supportshow. If not, use supportShowCfgEnable ficon and re-run the
test that was failing.
If this setting is not set to port-based routing on the Brocade 4100, 4900, or 5000 switches and the
Brocade 48000 switch in a FICON fabric, you will experience excessive interface control checks
(IFCCs) on the mainframe whenever a blade or CP is hot-plugged or unplugged.
Troubleshooting FICON CUP
In addition to the information you would collect for your support provider mentioned
“Troubleshooting FICON” on page 76, also gather the following information which is specific to
FICON CUP:
• Capture all data from the “General information to gather for all cases” on page 76.
• Verify the FICON CUP license is installed.
• Check the state of the CUP port by running the ficonCupShow fmsmode command. If it is
disabled, type the ficonCupSet fmsmode enable command to enable it. If this is on a directory,
enter the haFailover command to ensure both CPs are set correctly.
• Verify that the switch has a Fabric OS v6.1.0 or later.
• Ensure no device is plugged into port 254 on the Brocade 48000 or Brocade DCX
enterprise-class platform.
• Verify with the systems programmer that the CUP ports are defined properly.
FICON ports are defined as part of the sysgen process. The z/OS systems programmer can
verify if the CUP ports are defined properly.
Troubleshooting FICON CUP
5
• Verify the ports 254 and 255 display the Disabled (FMS Mode) by entering the switchShow
command. If not, enter the portDisable command on the appropriate models for the related
ports:
-On a Brocade 48000, disable 10/30 and 10/31.
-On a Brocade DCX Backbone, disable 12/30 and 12/31.
Troubleshooting NPIV
You should capture all pertinent data from the “General information to gather for all cases” on
page 76 and “Gathering additional information” on page 78.
NPIV licenses must be installed on Fabric OS v5.0.x. There is no license requirement for Fabric OS
v5.1.0 and later.
FICON Administrator’s Guide79
53-1002753-01
Page 94
Troubleshooting NPIV
5
80FICON Administrator’s Guide
53-1002753-01
Page 95
Appendix
Platforms supporting FICON
This appendix provides the following support details on the following:
• Brocade products with end of support announcements supported for FICON for different IBM
systems and drivers (Table 8).
• Brocade products currently supported for FICON for different IBM systems and drivers (Tab le 9
on page 82).
• Brocade blades supported for FICON in various Brocade Switch and Data Center Backbone
products (Tab le 10 on page 82).
Due to the nature of environments where mainframes are used, product qualification for FICON is
more rigorous than with other products. As a result, not all product combinations and versions of
Fabric OS are qualified for FICON.
Supported platforms with end of support
All Brocade products in Table 8 have end of support announcements. However, end-of-support
dates vary for each product, and your service provider may have end-of-support dates that differ
from Brocade dates. Contact your service provider for details.
A
Firmware listed in the following table is the final version supported and the recommended target
path
TABLE 8Supported platforms with end of support announcements
1.The IP ports of an FX8-24 extension blade can operate with another FX8-24 or 7800. The maximum number of FX8-24 blades
supported in a chassis it four. An FX8-24 can operate with an FX8-24E; however, IPSEC and compression is only supported on
one 10G IP port on the FX8-24.
2.The 64 port card is not supported for FICON but may be in the same chassis used for FICON. It must be in a logical switch that
does not have any FICON traffic.
3.For the 48 port card in an 8-slot chassis, FICON is supported in a logical switch configured for zero-based addressing mode.
For this card in a DCX-4S backbone, FICON is supported only on a logical switch defined for port-based (mode 2) addressing.
4.FC10-6 10Gbps port blade for ISL connections.
Unsupported blades
The following port blades are not supported in a FICON environment:
• FCOE10-24
• FS8-18
• FC4-48
• FC8-32E
• FC8-48E
A
FICON Administrator’s Guide83
53-1002753-01
Page 98
Supported Brocade blades
A
84FICON Administrator’s Guide
53-1002753-01
Page 99
Appendix
NOTE
Basic Switch Configuration
This appendix provides basic steps and commands to quickly configure a switch for fabric and
possible FICON and cascaded FICON operation.
For detailed concepts, procedures, and additional Fabric OS command options to configure a
switch switch point-to-point and cascaded FICON operation, refer to Chapter 2, “Administering
FICON Fabrics”.
The following tables are included in this appendix:
• Tab le 11 on page 86 provides the basic steps and commands in a checklist format to quickly
configure a switch for fabric and possible FICON operation.
• Tab le 12 on page 88 provides additional steps and commands for FICON cascading (2-byte
addressing).
Refer to the Brocade Fabric OS Command Reference or enter help followed by the command in a
Telnet session for additional information on commands used in this appendix.
B
FICON Administrator’s Guide85
53-1002753-01
Page 100
Basic Switch Configuration
B
TABLE 11Switch configuration example and commands
FeatureRelevant Commands
Upgrade code (if necessary)firmwaredownload
firmwareshow
Feature keys (add licenses)licenseadd
licenseshow
Create all logical switches and add all intended ports to each logical switch.lscfg
Set switch offlineswitchdisable
Repeat the following procedure to configure switch parameters for each
logical switch.
1Enter the configure command. You will be asked several questions. Only
answer questions listed in the Relevant Commands column. Use
defaults for other settings unless instructed by a Brocade Certified
Architect for FICON. Note that the BB_Credit setting here should not be
changed. BB_Credits will be set elsewhere in this checklist.
2Change fabric parameter settings by answering Yes. After answering
Yes , you will be asked several questions. Change only the following
parameters. Leave all other parameters at the default unless instructed
by a Brocade Certified Architect for FICON. After completing all of the
Fabric Parameter settings, answer No for all other parameters.
configure
• Domain
• Allow XISL Use
When applicable.
• Enable a 256 Area Limit
-0 - not supported for FICON
-1 - zero-based addressing. Recommended for FICON
-2 - Port based addressing. Not supported for 48 port cards in 8
port slot chassis.
• Disable Device Probing
Set to 1
• Insistent Domain ID Mode
Answer Yes.
Set Interop mode
0 - No other interop mode is supported for FICON
Port Based Routing.
Set to 1.
Set in order delivery with lossless DLS enabled.iodset --enable -losslessDLS
Enable dynamic load sharing only if lossless DLS is set.dlsset
Set Switch Nameswitchname
Set chassis namechassisname
Set Switch Onlineswitchenable
Set the principal switch. This can only be one switch in fabric. In mixed
M-Type and B-Type fabrics, the M-Type must be the principal switch.
interopmode 0
To set the routing policy to 1:
aptpolicy 1
To display the routing policy:
aptpolicy
fabricprincipal 1
86FICON Administrator’s Guide
53-1002753-01
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.