Beruflich Dokumente
Kultur Dokumente
Product Guide
REV 03
Copyright © 2001- 2015 EMC Corporation. All rights reserved. Published in the USA.
EMC believes the information in this publication is accurate as of its publication date. The information is subject to change without
notice.
The information in this publication is provided as is. EMC Corporation makes no representations or warranties of any kind with respect
to the information in this publication, and specifically disclaims implied warranties of merchantability or fitness for a particular
purpose. Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.
EMC2, EMC, and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries.
All other trademarks used herein are the property of their respective owners.
For the most up-to-date regulatory document for your product line, go to the technical documentation and advisories section on the
EMC online support website.
Preface
Chapter 2 Operations
Implementing AutoSwap through ConGroup................................................ 18
Defining configuration parameters ........................................................ 18
Starting the ConGroup address space ................................................... 19
Issuing commands ................................................................................ 20
When a swap needs to take place ............................................................... 22
Scenarios for swapping – some implementation choices............................. 23
Unplanned swaps ................................................................................. 23
Planned swaps...................................................................................... 24
Parameter settings for planned and unplanned swaps ......................... 24
Performing IODF configuration changes ....................................................... 26
Unacceptable configuration ACTIVATE ................................................... 26
Acceptable configuration ACTIVATE ....................................................... 27
Processing unacceptable configuration changes ................................... 27
Processing acceptable configuration changes ....................................... 33
Additional considerations for configuration changes ............................. 34
As part of an effort to improve its product lines, EMC periodically releases revisions of its
software and hardware. Therefore, some functions described in this document might not
be supported by all versions of the software or hardware currently in use. The product
release notes provide the most up-to-date information on product features.
Contact your EMC representative if a product does not function properly or does not
function as described in this document.
Note: This document was accurate at publication time. New versions of this document
might be released on the EMC online support website. Check the EMC online support
website to ensure that you are using the latest version of this document.
Purpose
This guide describes EMC AutoSwap from the perspective of the EMC AutoSwap for z/OS
user.
Audience
This guide is intended for the host system administrator, system programmer, or operator
who is evaluating, planning for, managing, or using EMC AutoSwap.
Related documentation
The following EMC publications provide more information:
◆ Mainframe Enablers Release Notes
◆ Mainframe Enablers Installation and Customization Guide
◆ Mainframe Enablers Message Guide
◆ ResourcePak Base for z/OS Version Product Guide
◆ Consistency Groups for z/OS Product Guide
◆ TimeFinder/Clone Mainframe Snap Facility Product Guide
◆ Symmetrix Remote Data Facility (SRDF) for VMAX 40K, VMAX 20K/VMAX, DMX Series
Product Guide
◆ Symmetrix TimeFinder for VMAX 40K, VMAX 20K/VMAX Series Product Guide
CAUTION, used with the safety alert symbol, indicates a hazardous situation which, if not
avoided, could result in minor or moderate injury.
IMPORTANT
An important notice contains information essential to software or hardware operation.
Typographical conventions
EMC uses the following type style conventions in this document:
Normal Used in running (nonprocedural) text for:
• Names of interface elements, such as names of windows, dialog boxes,
buttons, fields, and menus
• Names of resources, attributes, pools, Boolean expressions, buttons,
DQL statements, keywords, clauses, environment variables, functions,
and utilities
• URLs, pathnames, filenames, directory names, computer names, links,
groups, service keys, file systems, and notifications
Bold Used in running (nonprocedural) text for names of commands, daemons,
options, programs, processes, services, applications, utilities, kernels,
notifications, system calls, and man pages
Used in procedures for:
• Names of interface elements, such as names of windows, dialog boxes,
buttons, fields, and menus
• What the user specifically selects, clicks, presses, or types
Italic Used in all text (including procedures) for:
• Full titles of publications referenced in text
• Emphasis, for example, a new term
• Variables
Courier Used for:
• System output, such as an error message or script
• URLs, complete paths, filenames, prompts, and syntax when shown
outside of running text
Courier bold Used for specific user input, such as commands
Courier italic Used in procedures for:
• Variables on the command line
• User input variables
<> Angle brackets enclose parameter or variable values supplied by the user
[] Square brackets enclose optional values
| Vertical bar indicates alternate selections — the bar means “or”
{} Braces enclose content that the user must specify, such as x or y or z
... Ellipses indicate nonessential information omitted from the example
Note: To open a service request through the EMC Online Support site, you must have a
valid support agreement. Contact your EMC sales representative for details about
obtaining a valid support agreement or to answer any questions about your account.
Product information
For documentation, release notes, software updates, or for information about EMC
products, licensing, and service, go to the EMC Online Support site (registration required)
at:
https://support.EMC.com
Technical support
EMC offers a variety of support options.
Support by Product — EMC offers consolidated, product-specific information on the Web
at:
https://support.EMC.com/products
The Support by Product web pages offer quick links to Documentation, White Papers,
Advisories (such as frequently used Knowledgebase articles), and Downloads, as well as
more dynamic content, such as presentations, discussion, relevant Customer Support
Forum entries, and a link to EMC Live Chat.
EMC Live Chat — Open a Chat or instant message session with an EMC Support Engineer.
eLicensing support
To activate your entitlements and obtain your Symmetrix license files, visit the Service
Center on http://support.EMC.com, as directed on your License Authorization Code
(LAC) letter emailed to you.
For help with missing or incorrect entitlements after activation (that is, expected
functionality remains unavailable because it is not licensed), contact your EMC Account
Representative or Authorized Reseller.
For help with any errors applying license files through Solutions Enabler, contact the EMC
Customer Support Center.
If you are missing a LAC letter, or require further instructions on activating your licenses
through the Online Support site, contact EMC's worldwide Licensing team at
licensing@emc.com or call:
◆ North America, Latin America, APJK, Australia, New Zealand: SVC4EMC
(800-782-4362) and follow the voice prompts.
◆ EMEA: +353 (0) 21 4879862 and follow the voice prompts.
Your comments
Your suggestions help us continue to improve the accuracy, organization, and overall
quality of the user publications. Send your opinions of this document to:
techpubcomments@emc.com
Licensing
As of Enginuity level 5875 or higher, Mainframe Enablers introduces support for Electronic
Licensing (eLicensing). With the introduction of eLicensing, Symmetrix licensing is moving
from a host-based model to a Symmetrix-based model, with the majority of licenses now
being stored in a feature registration database on the Symmetrix system.
However, there are still a number of Symmetrix licenses that remain host-based and use
License Feature Codes (LFCs).
For these remaining host-based licenses and for Symmetrix systems running Enginuity
level 5874 or lower, ResourcePak Base (EMCSCF) requires LFCs to enable separately
chargeable features in EMC software. These features require an LFC to be provided during
the installation and customization of EMCSCF. The EMC Mainframe Enablers Installation
and Customization Guide lists the LFCs for the Mainframe Enablers components.
Note: The Consistency Groups for z/OS Product Guide provides more information.
AutoSwap handles automatic workload swaps between Symmetrix systems when the
AutoSwap software detects an unplanned outage or problem.
In both cases, ConGroup or ConGroup with AutoSwap performs its swaps transparently,
without operational interruption. Swaps can take place while application workload
continues.
You can use AutoSwap in both shared and non-shared DASD environments. AutoSwap
uses standard z/OS operating system services to ensure serialization and to affect swaps.
AutoSwap uses the Cross System Communication facility (CSC) of SCF 1 to coordinate
swaps across multiple z/OS images in a shared DASD or parallel sysplex environment.
Swap types
As noted previously, there are two types of swaps that ConGroup and AutoSwap can
perform:
◆ Planned swaps
◆ Unplanned swaps
Planned swaps
Planned swaps are swaps that are staged at the direction of site personnel when certain
planned events must take place. Planned swaps can facilitate many functions such as
non-disruptive building maintenance, power reconfiguration, DASD relocation, and
channel connectivity reorganization.
You generally perform planned swaps, either manually or automatically, through
ConGroup.
1. Symmetrix Control Facility (SCF) delivers a “persistent address space” on the host that facilitates
communication between the host and the Symmetrix system as well as other EMC-delivered and
partner-delivered applications.
Unplanned swaps
Unplanned swaps are automatically initiated when certain situations are detected.
Unplanned swaps can protect systems against outages of various kinds. Examples include
failures occurring to power supply, building infrastructure, air conditioning, channel
connectivity or entire DASD subsystems, operator error or the consequences of intended
or unintended sprinkler discharge.
You set up the conditions under which you want an unplanned swap to occur through the
AutoSwap extension. Then, when such a situation does occur, ConGroup with the
AutoSwap extension executes the swap on your specification.
AutoSwap modes
AutoSwap has two modes of operation.
◆ CAX (ConGroup AutoSwap Extension)
◆ Basic AutoSwap
CAX
CAX is the normal AutoSwap mode. It is an extension of ConGroup. In CAX mode,
AutoSwap functions are invoked:
◆ By you, through commands directed to the ConGroup address space
◆ Automatically, according to rules defined by ConGroup parameters
You must use CAX for automated (unplanned) swaps.
Note: The Consistency Groups for z/OS Product Guide provides more information about
CAX.
Basic AutoSwap
AutoSwap also has a limited “basic” mode that you can invoke without ConGroup,
through the SCF Address Space. Basic AutoSwap only handles manual (planned) swaps.
Basic AutoSwap is used in situations where ConGroup is not, or cannot be, active. (An
example is an R2 to R1 swap.)
Group swaps are always performed with dependent write consistency. This results in a
restartable image, even if single or multiple failures occur during the swap process.
Note: “Consistent swaps” on page 15 and the Consistency Groups for z/OS Product Guide
provide definitions and explanations of dependent write consistency.
1. Modified Indirect Data Address Words. Refer to IBM z/Architecture Principles of Operation,
publication number SA22-7832.
Note: : “When a swap needs to take place” on page 22 provides more information.
During a planned AutoSwap swap the high priority device pairs are swapped prior to the
normal priority devices. During an unplanned swap the high priority and normal priority
devices are swapped concurrently.
High Priority devices have a number of restrictions that limit normal application, non High
Priority, data from or shared data being placed on these devices:
◆ CAX swaps are always consistent, however data on the "streamlined" volumes may
not be consistent with other volumes in the group.
◆ RESERVE transfer processing is not applicable to these device which means that a
RESERVE may be lost if it is held at the time the AutoSwap swap occurs.
◆ To allow these devices to be swapped quickly the normal cross system checkpoint
processing is not performed. High priority R1 devices are set to a NRDY state early in
the swap which prevents any other system updating the device. This might cause a
false unplanned event due to 'intervention required' to be detected on other hosts.
In order to allow RDF reconfiguration processing to complete quickly AutoSwap verifies
during VALIDATE processing that High Priority devices are always in an R1->R2 direction
and that they are part of a CAX group. AutoSwap fails the validation if this is not the case.
In addition, as checkpoint processing is eliminated with High Priority devices, normal
priority devices must be part of the swap group.
Consistent swaps
Swaps performed by ConGroup or AutoSwap are consistent swaps. Page and OPS/MVS
data are treated with a high priority and volumes containing them swap ahead of other
volumes. If regular data are on these volumes, the result may be an inconsistent swap.
Consistent swaps ensure that the swap is a unique, atomic operation that maintains
dependent write consistency. Consistent swaps use EMCs consistency technology to
temporarily “freeze” write operations to the entire swap group during the swap.
Consistency technology honors a guaranteed sequence of updates in the I/O stream to a
group of devices that are remotely mirrored. Consistency technology is provided by
ConGroup.
Applications, database systems, and system components (such as VTOC processing)
require and ensure sequence integrity by waiting for successful completion of one update
before requesting the next.
In a remote disk copy environment, data consistency cannot be ensured if a later update
was remotely mirrored when its predecessor was not. This can happen when remote link
failure(s) affect only a subset of the disk controllers that are performing the remote copy
function.
When a transmission to a remote Symmetrix system fails, ConGroup halts subsequent
(possibly dependent) writes to all remote DASD systems containing members of the group
of devices. The synchronous relationship is suspended on all subsystems until recovery
action is taken. ConGroup ensures that all LPARs are aware of these changes. Should the
local system fail, the remote system is in a restartable condition.
System users may notice slight performance degradation during a consistent swap. This
usually lasts no more than a few seconds. The degree of impact depends upon factors
such as:
◆ Number of volumes being swapped
◆ Overall workload
◆ Write content
Note: The Consistency Groups for z/OS Product Guide describes the consistent copy
process.
Consistent swaps 15
EMC AutoSwap Overview
If there is any question as to whether AutoSwap and/or CSC are running under your
MFE/SCF task, look for the following messages in your MFE/SCF task log:
SCF0301I SCF.DAS.ACTIVE=YES
SCFS175I AutoSwap Initialization complete.
SCFS285W AutoSwap waiting for EMCSCF Cross System Communication
SCFS226I
AutoSwap has initialized with EMCSCF Cross System Communication
For more information concerning this INI parameter, refer to the section in the
ResourcePak Base Product Guide entitled Configuration File and Initialization Parameters.
Operations 17
Operations
Note: The Consistency Groups for z/OS Product Guide describes the ConGroup parameters
and the ConGroup started task. Chapter 5 in this guide describes the AutoSwap
commands.
PAGEDEV_ALLOWED allows you to add volumes with page datasets to a consistency group,
or explicitly prevent addition of volumes with page datasets to the consistency group.
Note: These global configuration parameters are described in the Consistency Groups for
z/OS Product Guide and in Chapter 4 of this guide.
Consistency-group-specific parameters
Consistency-group-specific configuration parameters apply to individual consistency
groups.
For example, you use the CONGROUP=name parameter to define a group’s name. After
you define the name, define the devices in the group. You can define the devices through:
◆ The DEVICE_LIST=devices parameter
◆ The SMS_GROUP=smsgroupname parameter
Another consistency group specific configuration parameter, CAXOPTS, is particularly
important for AutoSwap. You define a ConGroup’s AutoSwap behavior with the CAX
parameter, which names the CAXOPTS set to be associated with this ConGroup. This
enables you to create one behavioral definition for multiple consistency groups, and know
that their behaviors are identical.
There are other group-specific parameters not covered here. These consistency group
specific parameters are described in the Consistency Groups for z/OS Product Guide.
If you do not, you may obtain an undesirable result if you run ConGroup under the Job
Entry Subsystem, and ask ConGroup to control JES datasets. In such a case, ConGroup
may request services from JES (such as WTO) at the same time as it suspends I/O to JES
devices. Running as SUB=MSTR resolves this issue and also gives the ConGroup address
space higher priority. This can be advantageous in heavily loaded systems.
The Mainframe Enablers SAMPLIB contains the sample started task procedure to run the
ConGroup task, EMCCGRP. Customize these members for your installation, and install
them in the appropriate PROCLIB.
Note: The Mainframe Enablers Installation and Customization Guide provides more
information.
Issuing commands
AutoSwap has a series of operator commands that:
◆ Define how processing is to proceed
◆ Allow you to view the state of the AutoSwap configuration
Command categories
The major AutoSwap operator command categories are as follows:
◆ DEFINE GROUP
◆ DELETE GROUP
◆ DISPLAY
◆ SET
◆ SETSWAP
◆ SWAP
◆ VALIDATE GROUP
Command tasks
These commands allow you to:
◆ Concurrently swap large numbers of devices
◆ Perform dynamic workload configuration without application downtime
◆ Handle device group operations
◆ Relocate logical volumes
◆ Perform consistent swaps
◆ Implement planned outages of individual devices, of one or multiple subsystems, or
of a whole site
Command syntax
After you define the configuration parameters, you may issue these commands (using
MODIFY) to direct AutoSwap and ConGroup processing to the ConGroup started task
through the ConGroup operator console.
Use the following format:
F EMCCGRP,DAS keyword parameter,...
Where:
EMCCGRP
The ConGroup started task name. Your installation may have chosen a different name.
DAS
Tells ConGroup that an AutoSwap command follows.
keyword
An AutoSwap command.
parameter
An AutoSwap parameter.
,...
Note: If you are going to a new line, begin the line with a continuation character.
Continuations are indicated with a '-' after the last value on a line, for example:
...
DEF GRP PROD INC CUU=312C-312F -
CFW=RES PREVAL PROCCNT=3 RETAIN REPLACE
...
Example
After you have set up swap groups and behaviors as configuration parameters, you may
request a swap by a command such as:
F EMCCGRP,DAS SWAP GROUP groupname
Where:
EMCCGRP
DAS
Note: “SWAP ” on page 84 provides more information about the SWAP command.
GROUP
Tells AutoSwap that it’s a group that is swapped, not an individual device.
groupname
Note: You can enter AutoSwap commands through the CONFIGCA. These commands are
processed at startup.
1.
2.
Condition RDF devices to disable
access to the source devices and
enable access to the target devices
3.
4.
Switch UCB contents of the source and
target pairs
5.
AutoSwap holds all I/O during the swap process to ensure dependent write
consistency. This step protects data and ensures restartability in the swap group
should failures occur in the infrastructure during the swap event.
Note: It is possible to configure Host Read Only devices and make them available to an
AutoSwap group. In addition, an online target R/O R2 can remain R/O following a
swap on selected LPARs.
Refer to the descriptions of the SCF.DEV.ATTR.HRO.INCLUDE.LIST and
SCF.DEV.ATTR.HRO.EXCLUDE.LIST commands in the Initialization chapter in the
ResourcePak Base Product Guide and the description for
“[NO]AllowOnlineToDevice|[NO]AllOnTo” on page 65 in this guide.
2. Condition the RDF devices to disable access to the source devices and enable access
to the target devices.
During the time I/O to the devices is being queued, AutoSwap reconfigures the SRDF
pair to allow the application I/O stream to be serviced by the target SRDF device.
3. Transfer reserves to the target devices.
4. Switch the contents of the Unit Control Blocks (UCBs) of the source and target pairs.
Because the contents of the UCBs are swapped, the redirection of I/O is transparent to
the applications running. The redirection of I/O persists until the next IPL.
5. Release the halted I/O.
AutoSwap uses the Cross System Communication facility of SCF to coordinate swaps
across multiple z/OS images in a shared DASD or parallel sysplex environment.
Because AutoSwap uses SCF, the AutoSwap consistency environment can span
multiple LPARS, within or outside parallel sysplexes, multiplexes, or combinations of
these.
Unplanned swaps
Unplanned swaps provide continuity of operation in the event of certain failures. For
example, if device(s) become unreachable by the host, AutoSwap attempts to switch
access to devices at the other end of the SRDF/S (synchronous) links. This is always a
group swap covering all devices in the group.
An unplanned swap should not be a total surprise. In fact, most unplanned swaps should
be caused by “anticipated” events. Anticipated events are those you would prefer did not
happen, but for which you must plan. An example would be a power failure for which you
already had procedures in place.
Planned swaps
You can use planned swaps to eliminate the impact of short term unavailability of
Symmetrix DASD at the source side. Such a reason may be as simple as:
◆ The need to move systems as part of a computer room reconfiguration
◆ The need to reconfigure power systems or channel connectivity
In addition to establishing SRDF mirrors of data, EMC recommends that you consider
creating an extra dependent write consistent copy of your data on TimeFinder BCVs for
added protection during this period.
Unplanned swaps
Unplanned swaps handle emergencies. In emergency cases, you expect errors or unusual
conditions to occur. For example, an LPAR may lose connectivity with other LPARS in the
complex. You generally choose parameter settings that force the swap to continue in an
emergency.
You should consider what you want to happen when you restore the original source
system to an operational state. One of the requirements you should keep in mind is the
need to reestablish protection of your data quickly with two copies across the SRDF/S
links. Until this occurs, you have a reduced level of data protection.
While the original source system was unavailable, the z/OS systems were updating the
target Symmetrix systems, creating differences between this copy of the data and the copy
in the original source. The Symmetrix systems can handle these differences. They use this
information to resynchronize the copies differentially.
Planned swaps
Planned swaps have different needs. They can usually be deferred if failures occur during
the swap setup. For example, if an LPAR becomes unavailable during a swap, it is generally
safer to abort the swap and attend to the problem. Swapping may make the situation
worse.
You have two methods to perform return swaps after the planned swap completes and the
original source systems are once again available to participate in the protection of your
data.
1. Reverse the direction of the SRDF relationship. To achieve this, use the personality
swap procedure provided by SRDF Host Component for z/OS.
Note: The SRDF Host Component for z/OS Product Guide provides more information.
2. To resume RDF operations, issue the Host Component command below, and wait until
resynchronization is completed:
#SC VOL,devices,RDF-RSUM
3. Update R1s to R2s in ConGroup definition and issue the ConGroup command:
REFRESH congroupname FORCE
Where:
congroupname
Note: The Consistency Groups for z/OS Product Guide provides more information.
Where:
congroupname
This section describes the effect of dynamic configuration changes on AutoSwap, and
which changes AutoSwap considers acceptable and unacceptable.
A dynamic configuration change is requested by using the z/OS ACTIVATE IODF= operator
command, or through the HCD ISPF interface. The ACTIVATE can be a software-only
change, or both software and hardware change. A software-only change is requested using
the ACTIVATE SOFT keyword, and only affects operating system structures. Where SOFT is
not requested, then both a hardware and software change is performed. The IBM
documented usage of software versus the hardware ACTIVATE is to perform the
software-only activate on LPARs N-1 and then, finally, the software and hardware on LPAR
N. Hardware changes can affect LPARs, including the LPAR where the ACTIVATE is
performed. IBM is explicit on the ordering of the software-only as opposed to a software
and hardware change.
Additionally, a TEST keyword can be specified to exercise (or rehearse) the configuration
change without actually doing any real work. This is to verify that the intended IODF
ACTIVATE is likely to succeed. Its success does not mean that the real ACTIVATE will be
successful, as certain blocking actions may only be present at the real ACTIVATE time.
AutoSwap blocks unacceptable configuration changes that could cause a group to become
invalid or cause the triggering of an unplanned swap.
As of version 7.6, AutoSwap interfaces with dynamic configuration processing in order to
prevent unacceptable configuration changes. This is through the z/OS programmatic ENF
31/32 (CONFCHG CHGREQ/CHGCOMPL) interfaces that are documented for this usage:
◆ ENF 31 is equivalent to CONFCHG CHGREQ to notify changes that will be made by the
ACTIVATE. ENF31 is called a second time if the configuration change is rejected due to
affected device(s) and/or path(s) remaining pinned.
◆ ENF 32 is equivalent to CONFCHG COMPL to notify the completion (success) of the
ACTIVATE.
The blocking of a configuration change by AutoSwap does not mean that the configuration
change cannot be performed, but does mean that additional actions may be required to
allow the configuration change to proceed.
Unacceptable and acceptable software configuration changes are detected by AutoSwap,
and processed as described in the following sections.
In this example, the ACTIVATE is the AutoSwap ’TO’ controller access device.
To clarify the reason for the rejection, AutoSwap displays message CGRS646I. It also lists
the devices that the ACTIVATE is attempting to delete and the reason why the delete is
unacceptable.
The ’TO’ and ’FROM’ error device ranges (those blocking the configuration change)
precede the ’TO’ and ’FROM’ informational and warning device ranges. For an
unacceptable change, one or more reason lines describe why the change is not acceptable
and is being rejected.
All unacceptable changes can be overridden using the AutoSwap SETSWAP GROUP
DISABLE command. Unacceptable changes do not mean that the configuration change is
not possible. AutoSwap indicates those circumstances where a SETSWAP DISABLE is a
valid action with the following reason.
Reason: SETSWAP DISABLE required prior to ACTIVATE
SETSWAP DISABLE temporarily disables the AutoSwap group from performing SWAP
processing. This includes unplanned and planned SWAPs. On the subsequent SETSWAP
ENABLE, AutoSwap performs a local host validate to ensure the configuration is still
healthy and valid for SWAP processing. If it validates successfully, then all devices once
again have their unplanned triggers installed.
However, SETSWAP DISABLE can also be used to override AutoSwap, blocking the change
where SETSWAP is NOT a valid action. Performing such an action does not cause an
unplanned SWAP on the subsequent SETSWAP ENABLE. However, the group might be
marked invalid when the SETSWAP ENABLE command is issued.
For example, on a hardware change, the following ‘Reason’ lines indicate why the change
is rejected.
ACTIVATE IODF=0E,FORCE
1.3980-1.3981,1.1E00-1.1E0F,0.3980-0.3981,0.1E00-0.1E0F
COMPID=SC1XL
REASON=0151,CAN NOT DELETE DEVICE 3980
DESCTEXT=DEVICE PINNED
0Dynamic Configuration change blocked by AutoSwap
Each ‘Reason’ must be carefully examined and considered, as a SETSWAP DISABLE allows
an ACTIVATE to be performed for situations where it is not valid to do so. This is discussed
in the next section.
The SETSWAP disable can be for a specific group, or by using the generic ‘*’. While
SETSWAP DISABLE is in effect, all groups DISABLED indicate this at 30-second intervals
with the CGRS599W message. This is a reminder message to indicate during this period of
time that AutoSwap is unavailable for SWAP processing:
F CONGROUP,DAS,SETSWAP GRP * DISABLE
CGRS599W (00165) Group SSA3 has been SWAP DISABLED for 30 seconds.
And now retry the ACTIVATE. This time AutoSwap permits the ACTIVATE ‘Note’
informational messages are displayed to indicate the SETSWAP DISABLE had been
requested:
ACTIVATE IODF=0E,FORCE
CGRS599W (00165) Group SSA3 has been SWAP DISABLED for 60 seconds.
CGRS599W (00165) Group SSA3 has been SWAP DISABLED for 90 seconds.
CGRS599W (00165) Group SSA3 has been SWAP DISABLED for 120 seconds.
Now that the ACTIVATE is complete, the SETSWAP ENABLE can be entered. This results in
AutoSwap performing a local system validate to ensure that the configuration still looks
healthy. Any deleted devices are messaged by the CGRS644W and CGRS643W messages.
CGRS599W (00165) Group SSA3 has been SWAP DISABLED for 150 seconds.
CGRS599W (00165) Group SSA3 has been SWAP DISABLED for 180 seconds.
CGRS595I (00165) Group SSA3 SWAP processing ENABLED from host X006
(010E01F7281800D0).
CGRS644W (00165)(PID 00005) ‘TO’ device 03980 UCB has been deleted.
CGRS598I (00171) SETSWAP ENABLE completed: 193
Group SSA3 now ENABLED
Total groups processed 0: 1
Successful 00: 1
Failed 0 : 0
CGRS644W (00165)(PID 00006) ‘TO’ device 03981 UCB has been deleted.
CGRS643W (00165)(PID 00001) ‘FROM’ device 01E00 UCB has been deleted.
CGRS643W (00165)(PID 00002) ‘FROM’ device 01E01 UCB has been deleted.
CGRS643W (00165)(PID 00003) ‘FROM’ device 01E02 UCB has been deleted.
CGRS614W (00165) Alternate SS1 has 444 devices not in group SSA3
CGRS292I (00165) Group SSA3 216
Total Devices : 6 Highest PID : 6
Valid : 6 Invalid : 0
Auto Swappable: 3 Auto Pending : 0
Swapped : 0 Failed Swap : 0
Offline : 3 Not Defined : 5
Alternate SS : 4
This is an error, as deleting the ’TO’ device in this circumstance doesn’t allow AutoSwap to
SWAP to a device.
In the second instance, the group access device is being deleted:
Reason: 'TO' controller ACCESS device
Deleting this device might cause an issue for AutoSwap in both successfully completing a
SWAP and also in the usage of EMCSCF CSC:
ACTIVATE IODF=0E,FORCE
CGRS646I AutoSwap ACTIVATE results
ACTIVATE rejected by AutoSwap
'TO' AutoSwap device(s) being deleted:
Reason: 'FROM' partner device is ONLINE and delete
will result in an invalid AutoSwap group
Reason: Device(s) are blocking the configuration change
03980
'TO' AutoSwap device(s) being deleted:
Reason: 'TO' controller ACCESS device
Reason: Device(s) are blocking the configuration change
03981
'FROM' AutoSwap device(s) being deleted:
Note: UNPLANNED is now DISABLED
01E00-01E02
Devices Affected : 5
Blocking : 2
Ranges displayed : 3
Not displayed : 0
As previously discussed, all unacceptable changes can be overridden using the AutoSwap
SETSWAP GROUP DISABLE command.
However, SETSWAP DISABLE can also be used to override AutoSwap blocking the change
which may subsequently result in the group being marked invalid when the SETSWAP
ENABLE command is issued.
SETSWAP DISABLE can be used to override AutoSwap, blocking the change which may
subsequently result in the group being marked invalid when the SETSWAP ENABLE
command is issued.
For example, following the previous activate rejection, an attempt is made to invalidly use
the SETSWAP DISABLE command. As soon as the DELETE actions are performed, CSC
looses access to its gatekeeper, since it is the AutoSwap access device. In this case, swap
processing cannot be validated or swapped successfully. The ’FROM’ device that is online
is now issuing a warning that the group will be marked invalid when the SETSWAP ENABLE
is performed.
CGRS595I (00158) Group SSA4 SWAP processing DISABLED from host X006
(010E01F7281800D0).
CGRS598I (00163) SETSWAP DISABLE completed: 116
Group SSA4 now DISABLED
Total groups processed : 1
Successful : 1
Failed : 0
And now the SETSWAP ENABLE is performed, which results in the group now being marked
as invalid:
F CONGROUP,SETSWAP G * ENABLE
CGRS595I (00158) Group SSA4 SWAP processing ENABLED from host X006
(010E01F7281800D0).
CGRS644W (00158)(PID 00005) ‘TO’ device 03980 UCB has been deleted.
To correctly process this configuration change, the following planning actions must be
performed:
◆ The CSC gatekeeper must be changed in the EMCSCF SCFINI parameter file to a device
not being affected by the change and an INI,REFRESH performed. This automatically
results in AutoSwap unpinning and repining the new access device.
◆ In the case of the ’TO’ device being deleted where the ’FROM’ device was online, the
delete must be reconsidered, the ’FROM’ device varied offline, or the ’FROM’/’TO’
device ranges must be deleted from the CONGROUP CAX group using the CONGROUP
dynamic delete commands.
CGRS644W (00165)(PID 00005) ‘TO’ device 03980 UCB has been deleted.
CGRS644W (00165)(PID 00006) ‘TO’ device 03981 UCB has been deleted.
CGRS643W (00165)(PID 00001) ‘FROM’ device 01E00 UCB has been deleted.
CGRS643W (00165)(PID 00002) ‘FROM’ device 01E01 UCB has been deleted.
CGRS643W (00165)(PID 00003) ‘FROM’ device 01E02 UCB has been deleted.
CGRS614W (00165) Alternate SS1 has 444 devices not in group SSA3
CGRS292I (00165) Group SSA3 378
• It is possible that a third party incorrectly serializes this ENQ as EXCL. If AutoSwap
determines that an IODF ACTIVATE is not really in progress, then processing
continues after a minimal wait period. The GRS operator command D
GRS,RES=(SYSZIOS,DYNAMIC) can be used to show the current holders of this
ENQ.
This chapter describes the steps you must take to resume AutoSwap protected host
operations after a ConGroup AutoSwap event.
◆ Introduction ............................................................................................................ 38
◆ AutoSwap scenarios................................................................................................ 38
◆ Defining complement device groups ....................................................................... 39
Note: This chapter describes procedures that involve two other components: EMC
Consistency Groups for z/OS (ConGroup) and SRDF Host Component for z/OS. You can find
more information about these products in the Consistency Groups for z/OS Product Guide
and the SRDF Host Component for z/OS Product Guide.
Introduction
This section discusses options for “return swaps.” The discussion is not exhaustive.
Effective protection of your information requires more than just acquiring the best tools.
Use of SRDF, ConGroup, AutoSwap and TimeFinder requires careful planning. You must
have rigorously designed and documented procedures.
If you need assistance in developing such procedures, EMC recommends that you engage
EMC Business Continuity Services.
AutoSwap scenarios
There are two basic scenarios for resuming operation after a swap:
◆ Resuming operations with dynamic SRDF devices
◆ Resuming operations without dynamic SRDF devices
These scenarios are discussed in the following sections.
4. Resynchronize SRDF so that I/O can resume on the restored primary R1 devices.
You can restart ConGroup. After starting, ConGroup monitors invalid tracks. When this
count is zero, ConGroup resumes protection.
Where:
complementgroup
The name for the complement group. This name has the same requirements as other
consistency groups: no more than eight characters.
originalgroup
The name for the original group that was part of the AutoSwap event.
options
The optional arguments defined for the complement group. The options specified in
this statement apply to the complement group, not the original group.
The arguments for the original group and its complement do not have to be the same.
For example:
DEFINE GROUP returngroup COMPLEMENT prodgroup RETAIN SWAPCONTROL=BYDEVICE
Where:
prodgroup
The monitored group.
returngroup
The complement group.
Note: You can only define the complement group after you have defined the original
consistency group.
When defined before a ConGroup AutoSwap event, the complement group is created, but
is not populated with devices.
The “swap-back” device definition for the complement group occurs automatically during
the ConGroup AutoSwap event, so that the swap back definition defines only the devices
that were auto swapped.
Using CONFIGCA
You can use a complement group definition created through the COMPLEMENT argument
only if the ConGroup task that executed the AutoSwap event is still running and the base
group has already been defined.
For those instances where the ConGroup task is not running, you need to create your own
complement definition in an external file. Then, specify this file in the ConGroup start up
JCL, as the CONFIGCA DD statement.
However, using the CONFIGCA DD statement in the start up JCL may generate an error if
ConGroup has not been started and the base group has not been defined prior to
executing the CONFIGCA DD statement.
In this event, you need to issue the following statement to identify the group definition:
F CG,DAS,T PARMS
Note: You can also specify the user-defined complement-definition file in the AutoSwap
JCL CONFIGCA DD statement.
Where:
returngroup
Resynchronizing SRDF
The swap back operation using DAS SWAP or SWAP performs the necessary steps of SRDF
resynchronization, so that I/O may begin on the primary R1 devices. This operation is
documented in the SRDF Host Component Product Guide.
This chapter describes ConGroup parameters that are relevant to you, the AutoSwap user.
This chapter also provides guidance for using these ConGroup parameters.
◆ Overview................................................................................................................. 44
◆ CAX......................................................................................................................... 45
◆ CAXOPTS................................................................................................................. 45
◆ COUPLEDS_ALLOWED.............................................................................................. 54
◆ GLOBAL................................................................................................................... 55
◆ PAGEDEV_ALLOWED ................................................................................................ 55
Note: Refer to the Mainframe Enablers Message Guide for information about the error and
event messages that AutoSwap can produce.
Overview
ConGroup has five configuration parameters that are relevant for AutoSwap:
◆ CAX
◆ CAXOPTS
◆ COUPLEDS_ALLOWED
◆ GLOBAL
◆ PAGEDEV_ALLOWED
You specify these configuration parameters through the ConGroup configuration file.
CAXOPTS, COUPLEDS_ALLOWED, GLOBAL, and PAGEDEV_ALLOWED are global
configuration parameters. CAX is a consistency-group specific configuration parameter.
You can only use these parameters when defining a ConGroup AutoSwap group. You
cannot use it in the CONFIGCA DD or in the operator modify command set.
You can include a comment on a parameter line as follows:
keyword=value /* comment text */
Note: ConGroup configuration parameters are described in the Consistency Groups for
z/OS Product Guide.
Syntax Conventions
The parameter syntax shown in this chapter follow these conventions:
◆ italics = An argument.
◆ [ ] = Enclosed item is optional.
◆ item1|item2 = Item1 and item2 are alternative values. Either item1 or item2 can be
entered.
◆ Keywords appear in uppercase (for example, FROM). They must be spelled exactly as
shown.
◆ Variables appear in lowercase and italics (for example, column-name). They represent
user-supplied names or values in the syntax.
◆ Default values are indicated with an underline. For example, (YES|NO) indicates NO
is the default value.
◆ If punctuation marks, parentheses, arithmetic operators, or other such symbols are
shown, you must enter them as part of the syntax.
CAX
Purpose
CAX specifies that you want to activate AutoSwap for a device swap group and specifies
the CAX options set you want to use for that group.
You can only use these parameters when defining a ConGroup AutoSwap group. You
cannot use it in the CONFIGCA DD or in the operator MODIFY command set.
CAX requires that the ConGroup AutoSwap Extension feature is available.
Before you use CAX, you need to define a CAX option set by using CAXOPTS.
Note: “CAXOPTS” on page 45 provides more information. The Consistency Groups for z/OS
Product Guide provides an extended description of CAX.
Syntax
CAX=(CAXOPTS=options_set)
Argument
options_set
The name of an options set created with CAXOPTS. Multiple CAX statements may refer
to the same CAXOPTS statement. The option set name can be one through eight
characters in length.
Comments
To activate AutoSwap, you must have installed the EMC AutoSwap for z/OS Licensed
Feature Code (LFC). The Mainframe Enablers Installation and Customization Guide and the
ResourcePak Base for z/OS Product Guide discuss Licensed Feature Codes.
Example
CAX=(CAXOPTS=CASOPT1)
CAXOPTS
Purpose
CAXOPTS defines a set of ConGroup AutoSwap options and assigns a name to the set. You
can then associate that options set with a consistency group through CAX.
Note: The Consistency Groups for z/OS Product Guide describes CAX.
Syntax
CAXOPTS options_set=(cax_option[,cax_option[,...]])
CAX 45
Relevant ConGroup Parameters
Arguments
options_set
The name given to the set of CAXOPTS statements to use for this
consistency group.
The option set name can be one through eight characters in length.
cax_option
The options that make up the option set. If you use multiple
cax_option arguments, enclose them all in parentheses and
separate them with commas and no intervening spaces.
CAX options
You can use the following as CAX options (default values are underlined).
Page
Parameter and range of valid values link
[NO]AllowSystemsCountMismatch|[NO]ASCM 47
CFW=OFFVAL|NO|ALLOW 47
CROSSSYSTEMTIMEOUT=300|time_interval 48
LOSTOwnerpolicy 48
ONSWAP=[OPERATOR|HOLDIO|BACKOUT|SYSRESET(wait_code)]
[No]ROUTEMESSAGEtoowner [ALL|Warn|Error] 49
QUIESCETimeout=MIH|time_interval 49
UNPLANNEDCONDitions=INTERVENTIONREQUIRED|NOPATHS|SYNCLINKFAILURE 50
UNPLANNEDOPTIONS=FBAUSRNRDY 51
CSD=NRDY|USRNRDY 51
[NO]AllowOnlineToDevice|[NO]AllOnTo 52
[NO]AllowOnlineUndefinedDevice|[NO]ALLONUNDEFDEV 52
Descriptions
[NO]AllowSystemsCountMismatch|[NO]ASCM
During validation processing, the number of participating LPARS and path groups is
determined. Then at swap time, the number of participating LPARs and path groups is
determined again.
If you use the default, AllowSystemsCountMismatch (ASCM), the LPAR count at swap
time does not have to match the number of participating systems for the device
established at validation time.1 The swap can continue.
If you specify NOAllowSystemsCountMismatch (NOASCM), the LPAR counts at swap
time must be equal to or greater than the number of path groups for the device
established at validation time. If the LPAR count is less than the number of
established path groups for the device, the swap cannot proceed.
EMC recommends using the ASCM default for both planned and unplanned swaps
(especially where you have specified SWAPCONTROL=BYRANGE or BYGROUP) and
validate the group before performing swap processing.
If you specify NOASCM, the devices must be individually validated across all hosts. A
message is output for each device mismatch with an associated CGRS19I message.
Note: The Mainframe Enablers Message Guide provides details about messages you
can receive.
CFW=OFFVAL|NO|ALLOW
Controls cache fast write (CFW).2 The values can be:
OFFVAL
1. If LPARs and path groups for LPARs cannot be found at validation time, the message CGRP195E is
issued. The Mainframe Enablers Message Guide describes CGRP195E.
2. When you use CFW, additional integrity checks may be involved. In some cases, the subsystem ID
is checked. If the device has been swapped, the SSID is different and the job is terminated. For
this reason, EMC recommends against the use of CFW on devices that may be swapped.
CAXOPTS 47
Relevant ConGroup Parameters
NO
If CFW is active on any devices in the consistency group, then validation of the
consistency group fails. ConGroup AutoSwap processing is disabled.
ALLOW
CROSSSYSTEMTIMEOUT=300|time_interval
Specifies the cross-host timeout value. This is the time ConGroup AutoSwap waits for
acknowledgement from other ConGroup AutoSwap instances on other hosts. The
time_interval value may be in the range of 60 seconds through 9999 seconds.
The default value is 300.
LOSTOwnerpolicy ONSWAP=[OPERATOR|HOLDIO|BACKOUT|SYSRESET(wait_code)]
Specifies the action for ConGroup AutoSwap if all communication with the owner or
controlling system is lost during the swap process. Loss occurs when the owner
crashes or when SRDF and channel paths to the consistency group devices become
disconnected. LOSTOwnerpolicy affects the system behavior only during a swap
process. LOSTOwnerpolicy can stop different hosts from accessing different views of
the same data.
Note: The ConGroup GLOBAL configuration parameter controls which smfid is the
owner. The Consistency Groups for z/OS Product Guide provides more information.
In such a situation, ensure that the owner host is really dead and not just isolated
before you take any action, such as issuing a ConGroup TAKEOVERasowner command.
In some circumstances, the owner could back out of the swap without the owner being
able to contact the non-owners or the non-owners being able to contact the owner.
You can abbreviate LOSTOwnerpolicy as LOP.
OPERATOR
Note: Refer to the Mainframe Enablers Message Guide for error message details.
HOLDIO
Specifies that I/O is to be halted on the system that has lost connectivity with the
owner system.
BACKOUT
Specifies that the swap operation is undone on the system that has lost
connectivity with the owner system.
SYSRESET[(wait_code)]
Specifies that a non-restartable wait state of wait_code occur on the system that
has lost connectivity with the owner system. Valid values for wait_code can range
from X’FF0’ to X’FFE’. The default wait_code is 0FF0.
[No]ROUTEMESSAGEtoowner [ALL|Warn|Error]
Allows autoswap messages to be consolidated in a single place (the owner system
SYSLOG). This enables you to better locate issues in AutoSwap processing. (This alias
is ROUTEMSG.)
The keywords are:
ALL All nonowner AutoSwap messages are routed
to the owner system SYSLOG.
Warn Only W and E (warning and error) messages
are routed to the owner system SYSLOG.
ERROR Only E (error) messages are routed to the
owner system SYSLOG.
QUIESCETimeout=MIH|time_interval
Specifies the quiesce time out. During IO halt processing this is the time that
AutoSwap waits for any outstanding IO to complete on a device. If IO is still active on
the device following this timeout then a swap backout occurs.
The time interval can be between 0 and 300 seconds.
If the resolved timeout value is less than 15 seconds, then 15 seconds is used.
MIH, the default and recommended value, tells AutoSwap to use the z/OS Missing
Interrupt Handler timeout interval. A value of zero (0) is equivalent to specifying MIH.
CAXOPTS 49
Relevant ConGroup Parameters
UNPLANNEDCONDitions=INTERVENTIONREQUIRED|NOPATHS|SYNCLINKFAILURE
Details ConGroup AutoSwap action to be taken when the specified condition occurs
for a device in the ConGroup. This is a required argument. There is no default value.
Note that you can specify all UNPLANNEDCONDITIONS values or no
UNPLANNEDCONDITIONS value:
UNPLANCOND=INTERVENTIONREQUIRED
UNPLANCOND=NOPATHS
UNPLANCOND=SYNCLINKFAILURE
UNPLANCOND=INTERVENTIONREQUIRED NOPATHS SYNCLINKFAILURE
IMPORTANT
The UNPLANNEDCONDITIONS option replaces and performs the same function as the
AUTOSWAPCONDITIONS option used in previous releases of ConGroup. For backward
compatibility, however, ConGroup still recognizes AUTOSWAPCONDITIONS.
In this case, all possible autoswap conditions are disabled. There are no unplanned
swaps if a link failure occurs.
The other values can be:
INTERVENTIONREQUIRED
Specify NOPATHS if you want a swap to occur when a loss of access to a device
occurs. This includes loss of all channel paths to a device, a device being BOXed,
or when a paging device generates an intervention (intreq) condition. Intreq on a
paging device is included in a nopaths trigger, as such a condition would normally
result in a disabled WTOR and likely loss of an LPAR. If any of these conditions
occurs to a device in the group, NOPATHS tells AutoSwap to swap the group. EMC
recommends this CAX option if you want the protection offered by unplanned
swaps.
NOPATHS affects operations only when all paths are lost. If many paths fail,
performance may be severely degraded; but, an unplanned AutoSwap is not
triggered. You may choose to use AutoSwap’s SWAP command to transfer the
workload to the alternate Symmetrix systems.
SYNCLINKFAILURE
IMPORTANT
Use SYNCLINKFAILURE with caution. SYNCLINKFAILURE is not sensitive to
Concurrent SRDF or SRDF/Star configurations.
UNPLANNEDOPTIONS=FBAUSRNRDY
This parameter is for unplanned swaps involving FBA devices only. It specifies that FBA
R2 devices should be made Not Ready on the channel.
CSD=NRDY|USRNRDY
Used to specify or override the default of RNR (RDF-Not-Ready) state at completion of an
AutoSwap event. The state is set during AutoSwap checkpoint 2 processing in order to
prevent any further access to the R1 device.
The CSD parameter option has been added to allow users the ability to specify the
post-swap state of the R1 devices that were swapped.
CSD option argument values can be:
NRDY
(Default) This is optional and is the CAX default. Specifying NRDY sets R1 devices
to RNR state during the AutoSwap event.
USRNRDY
Sets the R1 devices that were swapped to User-Not-Ready state during the
AutoSwap event.
Note: Currently, high priority processing supports the CSD=USRNRDY option from
MFE 7.5 and some levels at MFE 7.3 and 7.4. Where an AutoSwap host with High
Priority devices without this support is detected, message CGRS642W will be
displayed and CSD=NRDY will be for High Priority devices instead.
CAXOPTS 51
Relevant ConGroup Parameters
[NO]AllowOnlineToDevice|[NO]AllOnTo
This option facilitates the Host Read Only (HRO) support on a TO device by allowing
ONLINE target (TO) devices in a AutoSwap or CAX group to be acceptable or
unacceptable to AutoSwap.
The default value of “NoAllowOnlineToDevice” results in the AutoSwap validation
process failing the group activation if there are any ONLINE target (TO) devices.
“AllowOnlineToDevice” is used in conjunction with the HRO Only attribute provided by
the ResourcePak Base SCF.DEV.ATTR.HRO.INCLUDE keyword specified in the
ResourcePak Base SCFINI parameter file. HRO is a local to the host where the
parameter is specified and prevents any write processing occurring to the TO device.
“AllowOnlineToDevice” allows ONLINE target (TO) devices to be present in the
AutoSwap group, but only where a corresponding SCF.DEV.ATTR.HRO.INCLUDE.LIST
also includes the target (TO) device. Where the R2 is ONLINE prior to an AutoSwap
SWAP the Read Only (R/O) state is ensured by Symmetrix.
However, following an AutoSwap SWAP the R2 becomes Read Write (R/W). Specifying
the target (TO) device in the SCF.DEV.ATTR.HRO.INCLUDE.LIST ensures that the Read
Only state is preserved following the swap.
For an ONLINE target (TO) device where there is no corresponding ResourcePak Base
SCF.DEV.ATTR.HRO.INCLUDE.LIST AutoSwap group validation fails and message
CGRS274E is displayed.
CGRS274E (00009)(PID 00002) 'TO' device 06712 has an invalid state,
RS 00000013 (online not Host R/O).
There are special considerations on the AutoSwap SWAP back where HRO is specified.
When the target (TO) device is not swapped and remains ONLINE, the device is set to
NRDY due to the ChangeSourceDevice (CSD) specification and becomes unavailable.
Additional processing is required following such a SWAP back to make the device
available again for access.
For further information on Host Read Only support refer to the ResourcePak Base
Product Guide.
[NO]AllowOnlineUndefinedDevice|[NO]ALLONUNDEFDEV
This option allows FROM devices, which are online but excluded or not discovered by
MFE ResourcePak base, to be acceptable (AllowOnlineUndefinedDevice) or
unacceptable (NOAllowOnlineUndefinedDevice) to AutoSwap during group validation
processing. This can occur when SCF.DEV.EXCLUDE is specified in the SCFINI file or
MFE ResourcePak base was unable to discover the device due to IO timeouts or some
other access issue.
Example
In this example the ConGroup AutoSwap Extension is required. If ConGroup AutoSwap is
not operating or the consistency group device group fails validation, ConGroup does not
enable the consistency group.
CAX=(CAXOPTS=CASOPT1)
CAXOPTS CASOPT1=(AUTOCOND=NOPATHS,CFW=OFFVAL,LOSTO
ONSWAP=SYSRESET(0FF0),ALLOWSYSTEMSCOUNTMISMATCH)
The CAX statement declared that CAXOPT1 specifies the ConGroup AutoSwap options for
this consistency group.
◆ AUTOCOND=NOPATHS specifies that if an I/O failure with a “no paths available”
condition for any of the devices in the consistency group device group that the failure
event actions begin.
◆ CFW=OFFVAL turns off cachefastwrite at the time of group validation. No CFW
processing is attempted during the swap event.
◆ LOSTOwnerpolicy ONSWAP=SYSRESET(0FF0) specifies that if all communication, both
channels and SRDF, is lost processing continues, if possible, on the owner system.
Other systems that can no longer communicate with the owner system enter a
non-restartable wait state 0FF0.
◆ ALLOWSYSTEMSCOUNTMISMATCH specifies that failure response actions do not occur
even if the system count at the time of the event does not match the system count at
validation time.
If the failure event is the inability to communicate from the R1 to the R2 a normal
ConGroup trip occurs and the ConGroup AutoSwap Extension is disabled. This is because
the R2 data is no longer current. A swap should never occur to “stale” data.
CAXOPTS 53
Relevant ConGroup Parameters
COUPLEDS_ALLOWED
Purpose
COUPLEDS_ALLOWED allows you to add volumes with couple datasets to a consistency
group, or explicitly prevent addition of volumes with couple datasets to the consistency
group.
Syntax
COUPLEDS_ALLOWED=NO|YES
Arguments
NO
(Default) Prohibits volumes with couple datasets to be explicitly added to the
consistency group.
YES
Allows volumes with couple datasets to be explicitly added to the consistency group.
If you include the ConGroup consistency group specific configuration parameter
DEVICE_LIST=ALL and also specify COUPLEDS_ALLOWED=YES, then volumes with
couple datasets are NOT added to the consistency group.
Note: The Consistency Groups for z/OS Product Guide describes DEVICE_LIST.
Comments
◆ There are seven types of couple datasets:
• SYSPLEX
• CFRM
• WLM
• OMVS
• ARM
• SFM
• LOGR
AutoSwap only supports LOGR couple datasets. If the primary LOGR couple dataset is
to be used as part of a consistency group, you should set up the LOGR structure in the
Coupling Facility for duplexing to staging datasets. After the consistency group has
tripped, the contents of the primary LOGR couple dataset, the LOGR staging datasets,
and the LOGR logstream datasets contain the information necessary to restart CICS.
◆ If you specify DEVICE_LIST=ALL, then volumes with couple datasets are not added to
the consistency group unless you take one of the following steps:
– Identify the couple devices for specific inclusion in the group by using
DEVICE_LIST=device.
– Specify COUPLEDS_ALLOWED=YES.
Example
COUPLEDS_ALLOWED=YES
GLOBAL
Purpose
GLOBAL has one argument, OWNER. OWNER specifies the SMF ID of the LPAR that controls
the consistency group and the ConGroup AutoSwap functions. If a failure occurs that
results in all communication being lost between the primary and secondary sites, the
system specified in OWNER proceeds while the other site halts.
Syntax
GLOBAL=(OWNER=SMFid)
Argument
OWNER=SMFid
The SMF ID of the owner LPAR. This argument must be the same in ConGroup
parameter files for all instances of ConGroup that use the defined consistency groups.
PAGEDEV_ALLOWED
Purpose
PAGEDEV_ALLOWED allows you to add volumes with page datasets to a consistency group,
or explicitly prevent addition of volumes with page datasets to the consistency group.
Syntax
PAGEDEV_ALLOWED=NO|YES
Arguments
NO
YES
Allows volumes with page datasets to be explicitly added to the consistency group.
Comments
If you specify the ConGroup, consistency group specific configuration parameter,
DEVICE_LIST=ALL, then volumes with page datasets are NOT added to the consistency
group unless you take the following actions. You should identify the page devices for
specific inclusion in the group by using DEVICE_LIST=device, and specify
PAGEDEV_ALLOWED=YES.
Note: The Consistency Groups for z/OS Product Guide describes DEVICE_LIST.
Example
PAGEDEV_ALLOWED=YES
GLOBAL 55
Relevant ConGroup Parameters
This chapter provides guidance for using AutoSwap commands and arguments.
◆ Overview................................................................................................................. 58
◆ DEFINE GROUP ....................................................................................................... 60
◆ DELETE GROUP ....................................................................................................... 71
◆ DISPLAY ................................................................................................................. 71
◆ SET ........................................................................................................................ 75
◆ SETSWAP ................................................................................................................ 80
◆ SWAP ..................................................................................................................... 84
◆ VALIDATE GROUP .................................................................................................... 86
Note: The Mainframe Enablers Message Guide provides more information about the error
and event messages that AutoSwap can produce.
Command Reference 57
Command Reference
Overview
The following sections provide reference pages for the following AutoSwap commands:
◆ DEFINE GROUP
◆ DELETE GROUP
◆ DISPLAY
◆ SET
◆ SETSWAP
◆ SWAP
◆ VALIDATE GROUP
You can enter these AutoSwap operator commands through the ConGroup operator
console.
Note: ConGroup commands are described in the Consistency Groups for z/OS Product
Guide.
Note: There is no SAF checking for commands that are read from the startup configuration
file (e.g. CONFIGCA).
To change the automatic verification, refer to the Mainframe Enablers Installation and
Configuration Guide for instructions on how to customize the security interface.
The resource information includes the following:
Class XFACILIT
Attribute Update for commands that affect AutoSwap, or Read for commands
that are display only.
Where:
<command name>
One of the operator enterable AutoSwap commands.
Syntax Conventions
The parameter syntax shown in this chapter follow these conventions:
◆ italics = An argument.
◆ [ ] = Enclosed item is optional.
◆ item1|item2 = Item1 and item2 are alternative values. Either item1 or item2 can be
entered.
◆ Keywords appear in uppercase (for example, FROM). They must be spelled exactly as
shown.
◆ Variables appear in lowercase and italics (for example, column-name). They represent
user-supplied names or values in the syntax.
◆ Default values are indicated with an underline. For example, (YES|NO) indicates NO
is the default value.
◆ If punctuation marks, parentheses, arithmetic operators, or other such symbols are
shown, you must enter them as part of the syntax.
Overview 59
Command Reference
The 4 digit device portion of the device number is identical for the same device pair. If the
source (FROM) was device 01234 then the corresponding target (TO) device would be
11234. In an AutoSwap group all of the 3390D devices must be contained in the same
subchannel set. Some system volumes, like the IPL device, cannot be contained within a
subchannel set other than 0.
Once the AutoSwap swap is complete, the target (TO) devices are now the base devices
even though it is still in a non-0 subchannel set. The source (FROM) devices are now the
3390D SPECIAL device type and are now inaccessible to any normal application usage.
When an IPL is required following a swap to devices in subchannel sets 1-3, the correct
devices need to be selected by z/OS as the base devices during IPL processing. The IODF
statement of the LOADxx member of SYSn.IPLPARM or SYS1.PARMLIB allows devices in a
subchannel set other than 0 to be selected as the base, or active, devices. Refer to the
IBM manual, z/OS 1.12 MVS Initialization And Tuning Reference for more information.
Note: To set the option to use multiple subchannel addressing, refer to the description of
SCF.DEV.MULTSS in the Initialization chapter of the ResourcePak Base Product Guide.
DEFINE GROUP
Purpose
Use DEFINE GROUP to define a set of devices to be monitored or swapped as a group. Use
of this command is not recommended because it creates a group that is NOT managed by
ConGroup. These groups have a significant limitation. Because they are not
ConGroup-managed, they have no unplanned swap mode.
Syntax
DEFine GROUP|GRP groupname [arguments][SDAS options]
Parameters
groupname
The name of the group to which you are assigning devices. The groupname can be
from 1 to 8 alphanumeric and special characters.
Arguments
Define Group arguments include:
Page
Parameter and range of valid values link
EXClude CCUU|CUU=(ccuu_lower[-ccuu_upper]| 61
[ccuu_lower[-ccuu_upper][,ccuu_lower[-ccuu_upper]]
)
EXClude VOLumes=(volser|volser_mask| 61
volser|volser_mask[,volser|volser_mask])
INClude VOLumes=(volser|volser_mask| 61
volser|volser_mask[,volser|volser_mask])
INClude CCUU|CUU=(ccuu_lower[-ccuu_upper]| 62
[ccuu_lower[-ccuu_upper][,ccuu_lower[-ccuu_upper]])
RDFGROUP|RDFGRP=ragrp#|CONFIGUREDDEVICE|CONF|CONFDEV 62
REPlace 62
Descriptions
EXClude CCUU|CUU=
(
ccuu_lower[-ccuu_upper]|
[ccuu_lower[-ccuu_upper][,ccuu_lower[-ccuu_upper]]
)
Exclude cuu devices or a range of cuu devices.
EXClude VOLumes=
(
volser|volser_mask|
volser|volser_mask[,volser|volser_mask]
)
Exclude a volser or volser mask.
In volser_mask, the mask characters are *, % ,?, where:
* Represents any number of characters (including a null string).
% or ? Matches a single character.
Note: If a volser contains special characters that conflict with the masking characters
(*, %, ?), the volser must be contained in quotes. For example: VOLUMES='VOL%01'
INClude VOLumes=
(
volser|volser_mask|
volser|volser_mask[,volser|volser_mask]
)
The INCLUDE argument only specifies volumes which are on the R1 side of an SRDF
R1/R2 set. You do not have to identify the associated R2 device(s) to AutoSwap. This
identification is automatic. By the time you issue the DEFINE GROUP command,
DEFINE GROUP 61
Command Reference
EMCSCF has performed a roll call and created an internal list of z/OS device
addresses, their matching Symmetrix ID and device numbers, and has a knowledge of
the R1/R2 relationships.
In volser_mask, the mask characters are *, % ,?, where:
* Represents any number of characters
(including a null string).
% or ? Matches a single character.
Note: If a volser contains special characters that conflict with the masking characters
(*, %, ?), the volser must be contained in quotes. For example: VOLUMES='VOL%01'
INClude CCUU|CUU=
(
ccuu_lower[-ccuu_upper]|
ccuu_lower[-ccuu_upper][,ccuu_lower[-ccuu_upper]]
)
Include cuu devices or a range of cuu devices.
RDFGROUP|RDFGRP=ragrp#|CONFIGUREDDEVICE|CONF|CONFDEV
This argument is used for concurrent SRDF devices, where the ragrp# indicates which
R2 to swap to. CONFIGUREDDEVICE is used if only 1 of the 2 concurrent R2 SRDF
devices is defined to the host. If the ragrp# is used, it must be valid for all concurrent
RDF devices defined in the group.
REPlace
Replace groups that have already been defined. This argument does not replace
groups that are active.
You should use REPLACE when a consistency group’s content has changed. When a
group is validated, AutoSwap communicates the group definition to all other LPARs,
thus achieving a global knowledge of group membership.
If a group with this group name is already active (validated and not subsequently
deleted), the validation with the new group definition fails.
You may change the definition of active groups using DELETE followed by DEFINE
GROUP.
SDAS options
Separate each SDAS option with a space.
Note: If you are going to a new line, begin the line with a continuation character.
Continuations are indicated with a '-' after the last value on a line, for example:
...
DEF GRP PROD INC CUU=312C-312F -
CFW=RES PREVAL PROCCNT=3 RETAIN REPLACE
...
Page
SDAS options Link
[NO]ALLOWCOUPLEDATASETS 64
[NO]AllowConcurrentCopy|[NO]ACC 64
[NO]AllowOnlineToDevice|[NO]AllOnTo 65
[NO]AllowOnlineUndefinedDevice|[NO]ALLONUNDEFDEV 65
[NO]AllowSnapSession|[NO]ALLSNAP 66
[NO]AllowSystemsCountMismatch|[NO]ASCM 66
CFW=OFFVAL[IDATION]|NO|OFF|RES[UME]|ALLOW 67
CHANGESOURCEdevice|CSD=[NRDY|NONRDY|NRDYAFTER | 67
USRNRDY] [NOVOLP
|VOLUMEPrefix=volser_pref|VOLP=volser_pref]
CROSSSYSTEMTIMEOUT|XSYSTO=[300|timeinterval] 67
FORCE=[NOLINK|LOSTSYStem] 68
MAXRC=maxrc 68
NOPREVALidate|NOPV|PREVALidate|PV 68
[PROCESSCOUNT=|PCNT=|PROCCNT=]64|process_count 68
RETAIN=SWAPCmplt|NORETAIN 69
Notes:
Default value SWAPCmplt for a compliment group.
Default value NORETRAIN for a normal consistency group.
[No]ROUTEMESSAGEtoowner [ALL|Warn|Error] 69
NOSWAPimmediate|SWAPimmediate 69
ValidateINTerval=timeinseconds 69
NOVOLserCHeck|VOLserCHeck 69
DEFINE GROUP 63
Command Reference
Descriptions
[NO]ALLOWCOUPLEDATASETS
Allows (or disallows with NOALLOWCOUPLEDATASETS) devices to be swapped where
they contain couple datasets.
Exercise caution when specifying the ALLOWCOUPLEDATASETS option. The only couple
datasets that should be processed are LOGR.
When defined in a CAX group, the ALLOWCOUPLEDATASETS specification is made
through the ConGroup COUPLEDS_ALLOWED specification. This is a ConGroup
argument specified in the definition of the consistency group.
• To set ALLOWCOUPLEDATASETS, use the ConGroup COUPLEDS_ALLOWED=YES
specification.
• To set the AutoSwap NOALLOWCOUPLEDATASETS option, use the ConGroup
COUPLEDS_ALLOWED=NO specification.
NOALLOWCOUPLEDATASETS is the default.
Note: The Consistency Groups Product Guide discusses the restrictions on couple
datasets specification.
[NO]AllowConcurrentCopy|[NO]ACC
Note: Both CC and Snap create sessions in which a local point-in-time-copy of data is
created. The location of this data depends on whether the data has been updated or
not. This mapping is kept in the local Symmetrix system. Remote Symmetrix systems
(at the other end of SRDF links) have no knowledge of these sessions. If you swap
devices involved in CC or Snap sessions, the sessions are lost.
Note: If you do not want to lose of these sessions, EMC suggests you don’t place these
sessions on swappable devices. AutoSwap can not prevent the use of CC or Snap
sessions. But AutoSwap can identify their existence, and may be used to control the
swap environment.
Note: You may use NOACC and NOALLSNAP to identify devices with active CC or Snap
sessions. If you specify NOACC or NOALLSNAP, a validation fails on devices with active
CC or SNAP sessions.This can be useful in a planned swap situation. If you intend
allowing unplanned swaps, then you can use ACC and ALLSNAP to ensure that
accidental use of CC or Snap does not interfere with unplanned swapping. You can use
VALIDATE to detect and report on the existence of CC and Snap sessions.
[NO]AllowOnlineToDevice|[NO]AllOnTo
This option facilitates the Host Read Only (HRO) support on a TO device by allowing
ONLINE target (TO) devices in a AutoSwap or CAX group to be acceptable or
unacceptable to AutoSwap.
The default value of “NoAllowOnlineToDevice” results in the AutoSwap validation
process failing the group activation if there are any ONLINE target (TO) devices.
“AllowOnlineToDevice” is used in conjunction with the HRO Only attribute provided by
the ResourcePak Base SCF.DEV.ATTR.HRO.INCLUDE keyword specified in the
ResourcePak Base SCFINI parameter file. HRO is a local to the host where the
parameter is specified and prevents any write processing occurring to the TO device.
“AllowOnlineToDevice” allows ONLINE target (TO) devices to be present in the
AutoSwap group, but only where a corresponding SCF.DEV.ATTR.HRO.INCLUDE.LIST
also includes the target (TO) device. Where the R2 is ONLINE prior to an AutoSwap
SWAP the Read Only (R/O) state is ensured by Symmetrix.
However, following an AutoSwap SWAP the R2 becomes Read Write (R/W). Specifying
the target (TO) device in the SCF.DEV.ATTR.HRO.INCLUDE.LIST ensures that the Read
Only state is preserved following the swap.
For an ONLINE target (TO) device where there is no corresponding ResourcePak Base
SCF.DEV.ATTR.HRO.INCLUDE.LIST AutoSwap group validation fails and message
CGRS274E is displayed.
CGRS274E (00009)(PID 00002) 'TO' device 06712 has an invalid state,
RS 00000013 (online not Host R/O).
There are special considerations on the AutoSwap SWAP back where HRO is specified.
When the target (TO) device is not swapped and remains ONLINE, the device is set to
NRDY due to the ChangeSourceDevice (CSD) specification and becomes unavailable.
Additional processing is required following such a SWAP back to make the device
available again for access.
For further information on Host Read Only support refer to the ResourcePak Base
Product Guide.
[NO]AllowOnlineUndefinedDevice|[NO]ALLONUNDEFDEV
This option allows FROM devices, which are online but excluded or not discovered by
MFE ResourcePak base, to be acceptable (AllowOnlineUndefinedDevice) or
unacceptable (NOAllowOnlineUndefinedDevice) to AutoSwap during group validation
processing. This can occur when SCF.DEV.EXCLUDE is specified in the SCFINI file or
MFE ResourcePak base was unable to discover the device due to IO timeouts or some
other access issue.
DEFINE GROUP 65
Command Reference
[NO]AllowSnapSession|[NO]ALLSNAP
Allows or prohibits (default) swaps of devices with snap sessions.
[NO]AllowSystemsCountMismatch|[NO]ASCM
Specifies whether or not to bypass the requirement that all other LPARs with an
established path group to the device belonging to the consistency group have EMCSCF
and ConGroup running.
During validation processing, the number of participating LPARS and path groups is
taken. Then, at swap time, the number of participating LPARS and path groups is taken
again.
If you specify AllowSystemsCountMismatch, the LPAR count at swap time does not
have to match the number of path groups for the device established at validation
time.1 The swap can continue.
Note: This option is recommended when you are using AutoSwap to protect against
failure events or unplanned outages. This is because loss of communication with one
or more LPARs may be a part of the failure scenario. This should not disable AutoSwap.
Note: This option is recommended for use in planned swaps. If your swap is planned,
and not the result of failure, the number of LPARs involved should not change. If it
does, there either has been a failure or an operational error.
1. If LPARs and pathgroups for LPARs cannot be found at validation time, the message CGRS195I
(ESWP195I for AutoSwap) is issued. Refer to the Mainframe Enablers Message Guide.
Note: The Mainframe Enablers Messages and Codes Guide provides more information
about messages you can receive.
CFW=OFFVAL[IDATION]|NO|OFF|RES[UME]|ALLOW
CHANGESOURCEdevice|CSD=[NRDY|NONRDY|NRDYAFTER] [NOVOLP
|VOLUMEPrefix=volser_pref|VOLP=volser_pref]
Where:
NRDY Sets the source device to NRDY status during the swap (at
checkpoint 2) to prevent access to the source device.
USRNRDY Sets the source device to USR-NRDY status during the swap
(at checkpoint 2) to prevent access to the source device.
NONRDY
Leaves the device in a RDY status upon the swap.
NRDYAFTER
Changes the device to NRDY following checkpoint 4 at the
successful completion of the swap.
VOLUMEPrefix Changes the volser prefix of the source device. The
VOLUMEPrefix must be one or two-characters in length. If
NRDY is specified then it will be changed to NRDYAFTER.
CROSSSYSTEMTIMEOUT|XSYSTO=[300|timeinterval]
Specifies the cross host timeout value. This is the time AutoSwap waits for
acknowledgement from other AutoSwap instances on other hosts. The default timeout
interval is 300 seconds.
DEFINE GROUP 67
Command Reference
FORCE=[NOLINK|LOSTSYStem]
NOLINK
You use this argument to define the action to be taken when an LPAR disappears
during swap processing. Before you swap, validation counts the number of running
LPARs. During swap processing, AutoSwap uses multiple checkpoints to
synchronize the activity between all the LPARs involved in the swap.
If an LPAR does not return status during checkpoint processing and you do not
specify FORCE=LOSTSYS, the swap aborts.
This option is complementary to the ALLOWSYSTEMSCOUNTMISMATCH option in
that the ALLOWSYSTEMSCOUNTMISMATCH has meaning during the validation for
path comparison processing and the FORCE=LOSTSYSTEM has meaning during the
swap processing. Either, both, or neither, option can be specified based on site
requirements.
The FORCE options are not merged with the global options.
MAXRC=maxrc
Specifies the MAXIMUMALLOWED return code.
NOPREVALidate|NOPV|PREVALidate|PV
NOPREVALidate|NOPV
(Default) Prohibit pre-validation of the group. This allows AutoSwap to validate the
group fully in the event of a manual swap.
PREVALidate|PV
Perform the first seven steps of the swap process, allowing the full swap to process
in a much shorter time.
[PROCESSCOUNT=|PCNT=|PROCCNT=]64|process_count
Specifies the number of swaps processed concurrently within a group. The default
PROCESSCOUNT is 64. Zero sets the maximum process count.
Note: This value cannot be changed with ConGroup CAX. ConGroup CAX always uses a
PROCESSCOUNT of zero.
Specifies the quiescent period (in seconds). The number of seconds can be between
zero (0) and 300. The default value is 15 seconds. A value of zero is equivalent to
specifying MIH (missing interrupt handler). If the resolved timeout value is less than
15 seconds, AutoSwap uses 15 seconds.
RETAIN=SWAPCmplt|NORETAIN
RETAIN=SWAPCmplt
(Default if the group is a normal consistency group) NORETAIN prevents a group, its
device definitions and AutoSwap options from being kept.
[No]ROUTEMESSAGEtoowner [ALL|Warn|Error]
Allows autoswap messages to be consolidated in a single place (the owner system
SYSLOG). This enables you to better locate issues in AutoSwap processing. (This alias
is ROUTEMSG.)
The keywords are:
ALL All nonowner AutoSwap messages are routed to the owner
system SYSLOG.
Warn Only W and E (warning and error) messages are routed to the
owner system SYSLOG.
ERROR Only E (error) messages are routed to the owner system
SYSLOG.
ValidateINTerval=timeinseconds
Specifies validating a group periodically as the server is running.
NOVOLserCHeck|VOLserCHeck
(Default) NOVOLSERCHECK instructs AutoSwap not to perform, volume serial
verification during the device validation; that is, this argument ensures that the FROM
and TO device volume serials are equivalent.
VOLSERCHECK instructs AutoSwap to perform volume serial verification during the
device validation; that is, this argument ensures that the FROM and TO device volume
serials are equivalent.
DEFINE GROUP 69
Command Reference
In the synchronous R1R2 environment this test is not required as the Symmetrix
system ensures that the volumes serials are equal as the devices are always
equivalent.
Note: The arguments for the original group and its complement do not have to be the
same.
Comments
◆ The DEFINE GROUP command has a special COMPLEMENT argument that allows you to
define a complement group to any currently defined swap group. The group defined
by COMPLEMENT is a set of devices to which normal operations is restored after an
outage.
The format for defining a complement group with DEFINE GROUP and the
COMPLEMENT argument is as follows:
DEFINE GROUP complementgroup COMPLEMENT originalgroup [SDAS options]
Where:
complementgroup
The name for the complement group. The groupname can be from one (1) to eight
(8) alphanumeric and special characters.
originalgroup
The name of the original group. This name also has the same requirements as that
of any consistency group: no more than eight (8) characters.
SDAS options
The optional SDAS options defined for the complement group. The options
specified in this statement apply to the complement group, not the original group.
The arguments for the original group and its complement do not have to be the
same. For example:
DEFINE GRP swapback COMPLEMENT origgrp AO ASCM CFW=RES
Note: Chapter 3 in this guide and to the Consistency Groups for z/OS Product Guide
provide more information.
Example
None.
DELETE GROUP
Purpose
Use DELETE GROUP to delete groups no longer needed to monitor devices. Also use
DELETE GROUP to delete an active group that cannot be replaced with the REPLACE
command.
Syntax
DELete GROUP|GRP groupname|groupname_mask
Parameters
groupname
The name of the group. The groupname can be from one (1) to eight (8) alpha-numeric
and special characters.
groupname_mask
The mask for the groupname.
In groupname_mask, the mask characters are *, % ,?, where:
* Represents any number of characters (including a null
string).
% or ? Matches a single character.
Example
The following example deletes the group PROD:
DEL GRP PROD
DISPLAY
This section describes the DISPLAY commands. The DISPLAY commands are:
◆ DISPLAY GROUP
◆ DISPLAY SDASOPTIONS
◆ DISPLAY OPTIONS
◆ DISPLAY GLOBALOPTIONS
◆ DISPLAY STARTPARMS
DISPLAY GROUP
Purpose
Use DISPLAY GROUP to display groups that have been successfully defined.
Syntax
DISPLAY GROUP|GRP groupname|groupname_mask [arguments]
DELETE GROUP 71
Command Reference
Parameters
groupname
The name of the group. The groupname can be from one (1) to eight (8) alphanumeric
and special characters.
groupname_mask
The mask for the groupname.
In groupname_mask, the mask characters are *, % ,?, where:
* Represents any number of characters (including a null string).
% or ? Matches a single character.
Arguments
Optional arguments that can be any of the following:
SDASOPTions|SOPT|OPTIONS|OPT
Displays the AutoSwap options used to define the group, or the AutoSwap default
values. For example:
Or displays the detail level (DET) view of the group. For example:
DIS GRP PROD DETAILS
sort options
[SORT[PID|PHASE|VOLSER|FROMDEVN|TODEVN|FROMSYMD|TOSYMD|
NO][ASCENDING|DESCENDING]]
Sorts the device pairs in the detail level display of a group by the specified column.
The NO option produces an unsorted display. The default sort option is PID
ASCENDING.
For example:
DIS GRP PROD DET SORT VOLSER ASCENDING
filter options
[FILTER FROMDEVN(xxxx-yyyy)|TODEVN(xxxx-yyyy)|
FROMSYMDEV(xxxx-yyyy)[,CTRLID(xxxxxxxxxxxx)]|
TOSYMDEV(xxxx-yyyy)[,CTRLID(xxxxxxxxxxxx)]|
PID(xxxxx-yyyyy)|VOLSER(mask)|CTRLID(xxxxxxxxxxxx)|
ALTSS|BASESS]
Filters the device pairs in the detail level display of a group, showing only the
groups that match on the specified value in the specified column.
FILTER ALTSS displays only the device pairs with TO devices that are in the alternate
subchannel set.
FILTER BASESS displays only the device pairs with TO devices that are in the active
base subchannel set.
For example:
DIS GRP PROD DET FILT FROMSYMD(009C-009F)
DIAG
Displays diagnostic information about a group. For example:
DIS GRP PROD DIAG
FIND findstring
Displays GROUPs containing findstring.
EXCLUDE findstring
Prohibits displaying GROUPs containing findstring.
For a DIS GRP DET command, the FIND and EXCLUDE arguments apply to the device pair
entries. The findstring applies to both lines.
Note: With the DISPLAY GROUP command, the OPTIONS argument works similarly to
SDASOPTIONS.
Example
The following example displays only the summary values for an AutoSwap group:
DIS GRP * DET F TOTAL
DISPLAY SDASOPTIONS
Purpose
Use DISPLAY SDASOPTIONS to display the default AutoSwap options in effect globally.
These can be changed by DEFINE GROUP commands. Note that use of this command is not
recommended because it operates outside of ConGroup.
Syntax
DISPLAY SDASOPTions|SOPT
Example
The following example displays all AutoSwap options in effect:
DISPLAY SOPT
DISPLAY OPTIONS
Purpose
Use DISPLAY OPTIONS to display all relevant options (SDASOPT, GOPT, and SPARMS) in a
single display.
Syntax
DISPLAY OPTions
DISPLAY 73
Command Reference
Example
The following example results in all relevant option displays (SDASOPT, GOPT, and
SPARMS) being appended and presented in a single display.
DISPLAY OPT
DISPLAY GLOBALOPTIONS
Purpose
Use DISPLAY GLOBALOPTIONS to display the global debugging and diagnostics in use.
Syntax
DISPLAY GLOBALOPTions|GOPT
Example
The following example displays the current global debugging and diagnostics:
DISPLAY GOPT
DISPLAY STARTPARMS
Purpose
Use DISPLAY STARTPARMS to display the current subsystem (LPAR) name.
Syntax
DISPLAY STARTPARMS|SPARMS
Example
The following example displays the current LPAR name:
DISPLAY SPARMS
DISPLAY examples
Consider the following examples:
◆ To find all devices marked as invalid to all groups, enter:
DISPLAY GRP * DET F INVALID
◆ To find all devices with UGL* volsers, but display those devices which are also invalid,
enter:
DISPLAY GRP * DET F UGL*INVALID
◆ To find all devices with UGL* volsers, but display those that are valid, enter:
DISPLAY GRP * DET F UGL*VALID
◆ To find all devices with UGL* volsers, but display those that are not invalid, enter:
DISPLAY GRP * DET X UGL*VALID
◆ To display all devices for PL* groups where they are not volser UGL*, enter:
◆ To display all devices for PL* groups where they are not volser U????1, enter:
DISPLAY GRP PL* DET X U????1
◆ To display all devices where it is an R2>R1 pair and a swap is in progress, enter:
DISPLAY GRP * DET F R2*SWAP*R1
For a DISPLAY GRP SUMM command, the FIND and EXCLUDE arguments apply to the group
entry display.
Consider the following examples:
◆ To display all groups which are idle, enter:
DISPLAY GRP * F IDLE
For a DISPLAY GRP SOPT command, the FIND and EXCLUDE applies to the options for the
group.
SET
This section describes the SET commands. The SET commands are:
◆ SET SDASOPTIONS
◆ SET PARMS
◆ SET [NO]DEBUG
◆ SET [NO]CAPS
◆ SET VERBOSE
◆ SET NOVERBOSE
◆ SET MAXLINECOUNT
◆ SET DEFAULTLINECOUNT
SET 75
Command Reference
SET SDASOPTIONS
Purpose
SET SDASOPTIONS sets the SDAS (AutoSwap) options globally for groups or for individual
swaps. The effect differs depending on whether you are setting options globally or within
a swap.
Note: You can use SET SDASOPTIONS in CONFIGCA DD init deck, as well. See the examples
below for more detail.
Syntax
seT SDASOPTions|SOPT SDAS_options
Parameters
SDAS_options
Are one or more options you want to assign. The SDAS (AutoSwap) options are those
listed as SDAS (AutoSwap) options under “DEFINE GROUP ” on page 60. However, you
cannot use the four DEFINE GROUP arguments, EXCLUDE, INClude, RDFGRP, or
REPLACE as SDAS (AutoSwap) options with SET SDASOPTIONS.
Example
SET PARMS
Purpose
Use SET PARMS to reprocess the commands specified in the CONFIGCA DD. You can use
this to bring in groups that have been deleted because of a swap or an operator
command.
Syntax
seT PARMS
Example
You can use the following command to reprocess parameters in the CONFIGCA DD:
F EMCCGRP,DAS T PARMS
SET [NO]DEBUG
Purpose
Use SET [NO]DEBUG to enable or disable debugging.
Syntax
seT DEBUG|NODEBUG
Parameters
DEBUG
Enables debugging information to be created. You should only set DEBUG if you are
instructed to do so by EMC Customer Support.
NODEBUG
Disables debugging information from being created.
Example
The following example disables debugging:
T NODEBUG
SET [NO]CAPS
Purpose
Use SET [NO]CAPS to whether AutoSwap logs displays in mixed or upper case. NOCAPS is
the default. That is, if you do not use this command, AutoSwap logs displays in mixed
case.
Syntax
seT CAPS|NOCAPS
SET 77
Command Reference
Example
The following example specifies that AutoSwap log displays in upper case:
T CAPS
SET VERBOSE
Purpose
Use SET VERBOSE to establish the level of detail (and indirectly the number) of messages
produced. The level can be set from zero (0) to 255, where zero is the least (terse) and 255
is the most (verbose). SET VERBOSE is generally only used at the request of EMC support
staff.
Syntax
seT VERBose verbose_level
Parameter
verbose_level
A number between zero (0) and 255. Refer to Table 1 for standard values and their
meanings.
0 Messages that are basic summaries of a condition or state. Such messages are
initially interesting, but describe a condition that occurs regularly, and thus
generates a large number of messages.
For example message pref238I indicates that a device is available for AutoSwap.
Message pref238I might be interesting initially but it becomes a nuisance under
normal operation.
Generally Verbose Level 0 messages are messages that were not originally
verbose, but are now verbose because of the volume of such messages that are
generated.
SET NOVERBOSE
Purpose
Use SET NOVERBOSE to remove verbose-level messaging. This is the recommended
setting unless requested by the EMC staff.
Syntax
seT NOVERBose
SET MAXLINECOUNT
Purpose
Use SET MAXLINECOUNT to specify the global maximum number of lines produced on long
displays. The default value for this command is 1000. The maximum value you can use is
99999.
The DEFAULTLINECOUNT value cannot exceed the MAXLINECOUNT value. Thus, you need to
examine the current MAXLINECOUNT value before you change the DEFAULTLINECOUNT
value.
Note: “Long Displays and Linecount Values” on page 89 provides more information.
Syntax
seT MAXLinecount linecount
Parameter
linecount
A numeric value that specifies the maximum number of lines to be produced on long
displays. You do not have to specify a line count value if you choose to use the default.
SET DEFAULTLINECOUNT
Purpose
Use SET DEFAULTLINECOUNT to specify the default number of lines for displays. The
default value for this command is 100. The maximum value you can use is 99999.
The DEFAULTLINECOUNT value cannot exceed the MAXLINECOUNT value. Thus, you need to
examine the current MAXLINECOUNT value before you change the DEFAULTLINECOUNT
value.
Syntax
seT DEFAULTLinecount linecount
Parameter
linecount
A numeric value that specifies the default number of lines to be produced on long
displays.
SET 79
Command Reference
SETSWAP
Purpose
You can use SETSWAP to disable swap processing for short periods of time. Typically this
would be done to avert an unplanned swap event which may conflict with customer
processing.
You can abbreviate SETSWAP as TS.
Note: Using the SETSWAP DISABLE command does not affect the trip operation of a
consistency group. Only the swap processing is disabled.
Syntax
seTSwap GROUP|GRP groupname ENAble|DISable parameter[,parameter[,...]]
Where:
[CoMmanDTimeout timeout]
[SUMmary|DETail]
[RUN|NORUN]
Parameters
CROSSSYSTEM
(Default) Process the request across all AutoSwap systems.
LocalSYStem
Only process the request on the local (command issuing) system.
This prevents a planned or unplanned swap from being initiated on this system. Other
AutoSwap systems that remain ENABLEd still allows the swap to be initiated. In this
instance, the DISABLEd systems also participate in the swap. However, where high
priority devices existed in the group prior to being DISABLEd these are no longer
performed at a high priority until the group is ENABLEd.
CoMmanDTimeout
Cross system communication timeout value in seconds. The default value is 60
seconds. If the command is not completed in this period of time then command
completion is forced.
SUMmary
(Default) If the command completes successfully then only generate output showing
the status change of the group.
If an error condition occurs then additionally generate output showing the hosts that
failed and the reason for the failure.
DETail
Output a summary line and a host line to show the status change for that host.
DISable
DISABLE a group or groups for SWAP processing. A message is displayed every 30
seconds to indicate that the group remains DISABLEd. The DISABLE period should be
kept to a minimum as no swap processing, either planned or unplanned, is allowed
during this period.
CAX devices in the affected groups become AutoPend in the DISPLAY GROUP DETAIL
output.
If an error condition occurs during DISABLE processing the group is returned to its
original state through backout processing. The original state was likely ENABLEd;
however, if the original state was DISABLEd, it remains DISABLEd. An additional
MLWTO is issued to display the status of the backout process.
ENAble
ENABLE a group or groups for SWAP processing. ENABLE processing results in a basic
validation of the group.
GROUP groupname
The groupname may be any active AutoSwap group and may contain masking
characters (* or %). For example GROUP * includes all AutoSwap groups.
Only active groups are applicable. If a group is marked ’inactive,’ then it is not
processed.
RUN
Check to see if the command is applicable and perform the required processing.
NORUN
Only check to see if the command is applicable. No active processing is performed.
Comments
None.
SETSWAP 81
Command Reference
Examples:
1. The following example1 disables and then enables a group.
F EMCCGRP,DAS,TS GRP PAGE DIS
CGRS598I (00104) SETSWAP DISABLE completed:
Group PAGE now DISABLED
Total groups processed : 1
Successful : 1
Failed : 0
2. The following example uses the DISPLAY GROUP command to display the state of the
current group. In this example the groups PAGE and PAGEC are disabled:
F EMCCGRP,DAS,D GRP PAGE%
CGRS162I (00106)
Group ID Owning System Host Defined Status
Name Identifier MM/DD/YY HH:MM:SS
-------- ----- ---- ---------------- -------- -------- --------
PAGE 00005 Group Owner 09/20/08 20:29:29 Idle *DISABLD
PAGEA 00006 Group Owner 09/20/08 20:29:29 Idle
PAGEB 00007 Group Owner 09/20/08 20:29:29 Idle
PAGEC 00008 Group Owner 09/20/08 20:29:29 Idle *DISABLD
PAGE2 00009 Group Owner 09/20/08 20:29:29 Inactive
PAGE3 00010 Group Owner 09/20/08 20:29:29 Inactive
PAGE4 00037 Group Owner 09/20/08 20:29:29 Inactive
Groups Matched : 7
3. The following example uses the DAS FIND DIS command to display only the groups
marked disabled. In this example the groups PAGE and PAGEC are disabled:
F EMCCGRP,DAS,D GRP PAGE% FIND DIS
CGRS162I (00106)
Group ID Owning System Host Defined Status
Name Identifier MM/DD/YY HH:MM:SS
-------- ----- ---- ---------------- -------- -------- --------
PAGE 00005 Group Owner 09/20/08 20:29:29 Idle *DISABLD
PAGEC 00008 Group Owner 09/20/08 20:29:29 Idle *DISABLD
Groups Matched : 7 Find Excluded : 5
4. The following example disables and then enables a group using the DETAIL output
F EMCCGRP,DAS,TS GRP PAGE DIS DET
CGRS598I (00107) SETSWAP DISABLE completed:
Group PAGE now DISABLED:
X04 (010435DE2096002F) : Warning, AutoSwap not active
X04 (010435DE20960062) : Request valid, RS 00
X06 (010635DE20960052) : Request valid, RS 00
X06 (010635DE20960054) : Warning, AutoSwap not active
X06 (010635DE2096004A) : Warning, AutoSwap not active
Total groups processed : 1
Successful : 1
Failed : 0
1. Because all the examples are ConGroup examples, the messages generated use the ConGroup
prefix CGRP.
SETSWAP 83
Command Reference
SWAP
This section describes the SWAP commands. The SWAP commands are:
◆ SWAP GROUP
◆ SWAP
SWAP GROUP
Purpose
Use SWAP GROUP to swap the UCBs of the devices defined in the group. The group does
not need to be pre-validated. The group is validated as it is swapped.
Syntax
SWAP|G GROUP|GRP groupname
[VALidate|VAL][CROSSSYSTEM|XSYS]
Parameter
groupname
The name of the group to swap. The groupname can be from one (1) to eight (8)
alphanumeric and special characters.
VALidate
Specifies to re-validate the group.
Note: VALIDATE is always performed unless the SWAP group contains the
PREVALIDATE specification. If you issue the VALIDATE command (page 86) for a
group, AutoSwap considers that group to have had the PREVALIDATE specification.
This means that if you issue the SWAP command for such a group, it is not
re-validated, unless you specify this VALIDATE option of the SWAP command.
CROSSSYSTEM|XSYS
When a define group is owned by another host, you must specify the
CROSSSYSTEM (or XSYS)argument on hosts other than the owning host in order to
initiate the request.
SWAP
Purpose
Use SWAP to swap individual devices or ranges of devices. You can also use SWAP to set
AutoSwap options.
Options you set through SWAP are merged with those set by the most recent SET
SDASOPTIONS command. If the same option is specified by SWAP and by SET
SDASOPTIONS, then the SWAP value overrides the similar specification for the SET
SDASOPTIONS command.
This version of SWAP maintains functional compatibility (not keyword compatibility) with
the functionality provided by previous versions of S/DAS.
Syntax
SWAP|G from-cuu to-cuu count [SDAS_options]
Parameters
from-cuu to-cuu
from-cuu and to-cuu are SRDF mirrors of each other. Concurrent devices may be
specified here since primary and secondary devices are explicitly stated.
count
The number of devices to be swapped. The default value is one (1).
SDAS_options
Are one or more options you want to assign. The options are the same as those listed
as arguments under “DEFINE GROUP ” on page 60 with the exception of EXCLUDE,
INClude, RDFGRP, and REPLACE.
Example
The following example lists the from-cuu (312E) and the to-cuu (34AE) and indicates that
four devices are to be swapped:
G 312E 34AE 4
SWAP 85
Command Reference
VALIDATE GROUP
Purpose
Use VALIDATE GROUP to validate groups explicitly to make sure that devices are still in
their intended state.
Note: After you issue the VALIDATE command for a group, AutoSwap considers that group
to have had the PREVALIDATE specification. This means that if you issue the SWAP
command for such a group, it is not re-validated, unless you specify the SWAP VALIDATE
option.
Syntax
VALidate GROUP|GRP groupname|groupname_mask
Parameters
groupname
The name of the group. The groupname can be from one (1) to eight (8) alphanumeric
and special characters.
groupname_mask
The mask for the groupname.
In groupname_mask, the mask characters are *, % ,?, where
* Represents any number of characters (including a null string).
% or ? Matches a single character.
Example
VAL GRP X
This appendix contains a worksheet you can use to gather the information you need for
your AutoSwap installation. You can also use this information when you enable and
customize AutoSwap.
◆ CAXOPTS configuration worksheet .......................................................................... 88
◆ Other global configuration parameters .................................................................... 88
Configuration Worksheet 87
Configuration Worksheet
Host:
The following work table allows you to collect information for the other AutoSwap
configuration parameters.
Host:
CAX=(CAXOPTS=options_set) ❏
COUPLEDS_ALLOWED=(NO | YES) NO ❏
GLOBAL=(OWNER=smfid)
PAGEDEV_ALLOWED=(NO | YES) NO ❏
This appendix describes how linecount settings can affect the displays of available
devices.
◆ Long displays.......................................................................................................... 90
Long displays
The following explains how the linecount settings affect displays of large numbers of
devices:
◆ If a linecount is specified, it is validated against the maxlinecount value. If the
specified value is greater than the maxlinecount value, the value for maxlinecount is
used.
◆ In general, a summary line is always displayed. This does not undergo the linecount
restriction (this being a useful feature for D GRP DET).
◆ A fixed heading line that is part of the message (for example, D GRP SUMMARY is not
included in the linecount.
◆ Multiline output is generated (batched) in parts. To limit MLWTO memory
usage/contention the displays are produced in 32K batches. Each continuation is
marked by a message line indicating continuation and a part number.
For example:
and so on...
◆ If at least one group display cannot be produced because the value is too small, the
following is appended to the associated message:
Line count too small. No groups displayed.
For example:
F PALSDAS2,D GRP T1 L 2 DET
ESWP217I (00033)
Line count too small. No groups displayed.
◆ If a group display is being truncated, then the 'more...' line is added to the end of the
MLWTO. The groups matched summary line indicates the number of groups displayed.
For example:
F PALSDAS2,D GRP * L 2
ESWP162I (00035)
Group ID Owning System Host Defined Status
Name Identifier 09/04/08 14:14:10
-------- ----- ---- ---------------- -------- -------- --------
A 00002 Group Owner 09/04/08 14:18:14 Inactive
A1 00003 Group Owner 09/04/08 14:18:14 Inactive
More....
Groups Matched : 2
Note: In the above example, the linecount does not include the fixed header line since
it is fixed; that is, it is only displayed once.
◆ Related parts of a message are excluded where the linecount is reached for any of the
related parts. This means that a complete coherent message is produced and not
truncated.
For example:
Note: The following two groups can only be displayed if a linecount of at least 27 is
specified (since there are 27 lines).
Long displays 91
Long Displays and Linecount Values
MAXRC=4
NoPrevalidate
ProcessCount=1
NoRetain
NoSwapImmediate
Group:A1, ID:00003, S/DAS options:
NoBypassOfflineDevices
CFW=Resume
ChangeSourceDevice=NRDY
NoVolumePrefix
MAXRC=4
NoPrevalidate
ProcessCount=1
Retain
NoSwapImmediate
More....
Groups Matched : 2
The following shows that the second group is not displayed (and thus truncated) when a
value lesser than 27 is given:
F PALSDAS2,D GRP * OPT L 26
ESWP163I (00040)
Group:A, ID:00002, S/DAS options:
NoPaths
NoBypassOfflineDevices
CFW=Resume
ChangeSourceDevice=NRDY
NoVolumePrefix
MAXRC=4
NoPrevalidate
ProcessCount=1
NoRetain
NoSwapImmediate
More....
Groups Matched : 1
◆ If a detail group display reaches the linecount in the detail output, then the
summaries are still produced and an indicator of the number of lines lost is output.
The summary is still calculated as it would have been had the Linecount not been
exceeded. This is useful because it allows you to use the Find and Exclude options
without having too much display. As for not truncating related data, the two lines
which you normally get for the From and To device is only displayed when both can be
displayed. That is, if the linecount is reached for the second line, then the first line is
not done.
For example:
F PALSDAS2,D GRP P DET L 7 F ' VALID'
ESWP217I (00046)
Group:P, ID:00006, Mode:Idle
TO device subchannel set; Active:NONE, Alternate:SS1
Creation Date (DD/MM/YY):09/04/08
Time (HH:MM:SS):14:18:15
PID Phase Volser| FROM/TO Device |Counts Status/
|Ty Devn CCA SSID Symd Ctrl#|Sys Pth Mode
----- ----- ------+-- ---- ---- -------- --- --+------ --------
Total Group Devices: 4 Missing Lines : 2
Selected: 1 Find Excluded : 3
Valid : 1 Invalid : 3
Swapped : 0 Failed Swap : 0
Offline : 0 Not Defined : 0
Groups Matched : 1
APPENDIX C
Swap Services Message Formatting
This appendix describes the message formatting for the EMC SWAP services messages
that are listed in the Mainframe Enablers Message Guide.
◆ Swap services conventions .................................................................................. ... 84
83
Swap Services Message Formatting
Table C-1
This chapter uses the following multi-prefix format for all message headers:
ESWPnnnE | CGRSnnnE| FMMSnnnE | SCFSnnnE
Look up the message by the prefix and code you receive.
Note: You may receive other prefixes than the ones listed above. If you do, go to EMC
support as described in “Where to get help” on page 7.
rrrrr
Is a request sequence number that identifies the AutoSwap command request on a
particular host. rrrrr is incremented each time a new command request is made.
All messages associated with the same request on the same host are prefixed by
the same value for rrrrr.
>hhhh
Is the SMFID of the host where the message was routed. Where the
RouteMessagetoOwner AutoSwap option is specified, messages routed from the
non-owner to the owner will contain the issuing host ID instead of a request
sequence number. The format is >hhhh, where hhhh is the SDMFID.
For example, >HIC3 indicates the message was routed from host HIC3.
ppppp
Is a PID that is a unique incrementing value for each SWAP validation or swap
process (that is, device pair) for the same group definition. This value always
follows the request sequence number or host ID to uniquely identify the messages
relating to the same device pair swap within the same AutoSwap group.
When a cross system validation or swap is performed, the same PID is used on all
hosts. The PID is set by the AutoSwap owner host when the group is created and
will remain the same for the life of the AutoSwap group.
Verbose levels
Some messages are only produced when the currently set SWAP verbose level is
greater than or equal to the verbose level of the message. Error messages and most
warning messages are always produced no matter what verbose level is set. The
following verbose levels are used to indicate particular SWAP processing:
Typographical conventions
Within a message, single quotes (‘ ‘) are used around the FROM and TO device
specification to denote and prevent confusion with the usage of the words FROM and
TO.
The default for messages is mixed-case. Use the SET CAPS command for uppercase
format.
The convention of [ ] indicates optional information is added to a message. A |
indicates an either (or) condition.