3PAR Inc.
4209 Technology Drive
Fremont, CA 94538 U.S.A.
Part No. 320-200175 Rev A
October 2009
Revision Notice
This is the first release of this manual. A complete revision history is provided at the end of this document.
Changes
The material in this document is for information only and is subject to change without notice. While reasonable efforts have been
made in the preparation of this document to assure its accuracy, 3PAR Inc. assumes no liability resulting from errors or omissions in
this document or from the use of the information contained herein.
3PAR reserves the right to make changes in the product design without reservation and without notification to its users.
Updates to the Documentation at 3PAR Central
Any updates to this document, or to other 3PAR technical documents, can be found by logging on to 3PAR Central’s Document
Control System from 3PAR’s Support page at http://support.3PAR.com.
3PAR Technical Support and Services
Contact your local service provider for technical support and services at http://www.3PAR.com/services.html.
Sales and Ordering Information
For sales and ordering information, contact:
3PAR Inc.
4209 Technology Drive
Fremont, CA 94538 U.S.A.
Telephone: 510–413–5999
Fax: 510–413–5699
E-mail: salesinfo@3PAR.com
Reader Comments and Suggestions
Please E-mail your comments and suggestions about this document to readercomments@3PAR.com.
3PAR, InServ, InForm, InSpire and Serving Information are registered trademarks of 3PAR Inc.
Microsoft, Windows, and Windows NT are either registered trademarks or trademarks of Microsoft Corporation.
UNIX is a registered trademark of The Open Group.
All other trademarks and registered trademarks are owned by their respective owners.
Federal Communications Commission Radio Frequency Interference Statement
WARNING: Changes or modifications to this unit not expressly approved by the party responsible for compliance could void the user’s
authority to operate the equipment.
This device complies with Part 15 of FFC Rules. Operation is subjected to the following two conditions (1) this device may not cause
harmful interference, and (2) this device must accept any interference received, including interference that may cause undesired
operation.
This equipment has been tested and found to comply with the limits for a Class A digital device, pursuant to Part 15 of the FCC rules.
These limits are designed to provide reasonable protection against harmful interference when the equipment is operated in a
commercial environment. This equipment generates, uses, and can radiate radio frequency energy and, if not installed and used in
accordance with the instruction manual, may cause harmful interference to radio communications. Operation of this equipment in a
residential area is likely to cause harmful interference in which case the user will be required to correct the interference at his own
expense.
volumes. For TPVVs, all data and snapshot space is allocated on demand from a CPG. For fully
provisioned virtual volumes, only the snapshot space is allocated on demand from the CPG.
2.12.1 Snapshots and Common Provisioning Groups
A Common Provisioning Group (CPG) is a virtual pool of logical disks that allows multiple
volumes to share the CPG's resources and allocate space on demand. However, CPGs still
require careful planning and monitoring to prevent them from becoming so large that they set
off the system's built-in safety mechanisms. These safety mechanisms are designed to prevent a
CPG from consuming all free space on the system, but they only work properly on systems that
are planned carefully and monitored closely.
CAUTION: Refer to the 3PAR InForm OS Concepts Guide for a complete list of
warnings and cautions regarding CPGs.
Remote Copy and Thin Provisioning
2.33
Remote Copy User’s GuideInForm OS Version 2.3.1
NOTE: While it is possible for a CPG to have up to 4095 volumes, it is strongly
recommended that no more than 32 volumes be associated with a single CPG. The
reasons for this limit are as follows:
■ Virtual volumes in the same CPG can share the same logical disk. In the unlikely
event that the logical disk is damaged (because of multiple simultaneous disk
failures, for example), all the volumes associated with that logical disk will be
unavailable.
■ Virtual volume performance might suffer from too much interleaving within the
logical disks.
For example, when Remote Copy requires that the system create a new snapshot of a volume,
the volume’s CPG might need to allocate additional space to that volume. That space is drawn
from the system’s common storage pool, at which point it is placed into the CPG’s individual
pool and then allocated to the volume. After the resynchronization takes place and the
snapshot is no longer useful, Remote Copy deletes the snapshot from the system. At that time,
the space formerly allocated for that snapshot is returned to the volume’s free space.
Contiguous free snapshot data space can be reclaimed and returned to the CPG using
3PAR Thin Copy Reclamation. Refer to the InForm OS Concepts Guide for information about
3PAR Thin Copy Reclamation.
2.34
2.12.2 Thinly Provisioned Virtual Volumes
NOTE: Thinly Provisioned Virtual Volumes (TPVVs) do not have snapshot space by
default. In order to use TPVVs with Remote Copy., you must create TPVVs with
snapshot space.
Thinly Provisioned Virtual Volumes (TPVVs) associated with the same CPG, draw space from
that pool as necessary, by allocating space on demand in small increments. As the volumes that
draw from the CPG require additional storage, the system automatically creates additional
logical disks and adds them to the pool until the CPG reaches the user-defined allocation limit
that restricts its maximum size.
Before creating a TPVV, you must first create a CPG as described in the 3PAR InForm OS CLI
Administrator’s Manual. After the CPG is established, use the instructions in the 3PAR InForm
OS CLI Administrator’s Manual to create volumes that draw from that pool.
Remote Copy and Thin Provisioning
InForm OS Version 2.3.1Remote Copy User’s Guide
NOTE: When a TPVV is configured as a primary Remote Copy volume, the initial
synchronization of the primary volume sends only the allocated data pages to the
backup storage server.
NOTE: When a TPVV is configured as a primary volume in a Remote Copy group,
the secondary volume should have no data written to it prior to adding it the
Remote Copy group, or it must match the primary volume in order for the primary
and secondary volumes to match during initial synchronization.
2.12.3 Fully Provisioned Virtual Volumes
Fully provisioned virtual volumes can also draw resources from a CPG. However, whereas a
TPVV draws all space from a CPG’s logical disk pool, fully provisioned virtual volumes only
draws snapshot data space from the pool.
Before creating a fully provisioned virtual volume, you must first create a CPG as described in
the 3PARInForm OS CLI Administrator’s Manual. After the CPG is established, use the
instructions in the 3PAR InForm OS CLI Administrator’s Manual to create volumes that draw
from that pool.
2.13 Error Handling
The following sections describe how Remote Copy handles various failures.
NOTE: For a complete disaster recovery scenario for both synchronous and
asynchronous periodic mode Remote Copy, see Appendix C, Example Setup and
Disaster Recovery.
2.13.1 Single Link Failure
For redundancy, two communication links should be configured between InServ Storage
Servers in each Remote Copy pair. When both links are active, both links carry data. Data for a
specific I/O is sent over a single link, but data for a different I/O can be sent over different links,
particularly when there are multiple volumes being replicated.
Error Handling
2.35
Remote Copy User’s GuideInForm OS Version 2.3.1
When a link between storage servers is broken, an alert is issued on each storage system as
soon as the link failure is detected. If the other link remains active, all data is sent on that
remaining link, though a reduction in throughput (bandwidth) might occur. No other
noticeable errors should occur.
2.13.2 Double Link Failure
When both links between InServ Storage Servers in a Remote Copy pair are broken, the two
storage systems cannot communicate. As a result, both storage systems declare each other
down and both systems generate alerts to indicate the failure of the other system. Systems
handle double link failures differently according to whether they are being used for
synchronous or asynchronous periodic remote copies.
NOTE: See Volume Group Modes on page 2.22 for descriptions of synchronous
and asynchronous periodic volume groups.
2.13.2.1 Synchronous Volume Groups
■ If the no_fail_wrt_on_err policy is enabled for a synchronous volume group (this is the
default), replication is stopped for the group and the system creates snapshots (virtual
copies) of all the primary volumes that have completed the initial synchronization.
2.36
The no_fail_wrt_on_err (default) will not display in the showrcopy command output,
but is displayed with fail_wrt_on_err.
When the links are restored, Remote Copy automatically brings the links back into service,
but does not restart the replication for the volume groups with the
no_fail_wrt_on_err
restart the replication for such volume groups unless the auto_recover policy is set (see
Setting Remote Copy Volume Group Policies on page 4.4).
NOTE: For synchronous groups, when the links are restored, you can copy any
writes from the primary to the secondary groups by issuing the
policy. The startrcopygroup command must be issued to
startrcopygroup command on the system that holds the primary group to
automatically resynchronize the primary and secondary groups.
■ Snapshots are created for all primary volumes, but no snapshots are created for secondary
volumes while replication is stopped.
Error Handling
InForm OS Version 2.3.1Remote Copy User’s Guide
In addition to taking snapshots, the offset and length of the I/O will fail to complete on the
secondary volumes because the link failures were recorded in the nonvolatile memory. The
failed I/O must be recorded separately because the snapshot is taken after that I/O has been
written for the primary volumes but before it was acknowledged and completed on the
secondary. When replication is restarted for the volume, the failed I/O is read from the
primary volume and sent to the system that holds the secondary volume. In addition, all
differences between the base volume and the snapshot taken when the replication was
stopped are also sent over in order to resynchronize the secondary volume with the primary
volume.
■ If the fail_wrt_on_err policy is enabled for a synchronous volume group, failing host
writes does not cause the mirror to become out of sync. A write error is returned to all host
writes as long as the links are down. When the links come back up, the writes are
completed normally. This policy can be enabled with the
setrcopygroup command.
2.13.2.2 Asynchronous Periodic Groups
For asynchronous periodic groups, when a double-link failure occurs, replication is stopped on
both storage servers. No new snapshots are created. If a resynchronization was in progress
when the failure occurred, the snapshots on the secondary system are automatically promoted.
NOTE: For asynchronous periodic groups, when the links are restored, replication
can be restarted by issuing the startrcopygroup command. The volumes are
resynchronized at the next scheduled resynchronization time. To resynchronize
prior to the next scheduled time, use the
initiate immediate resynchronization.
2.13.3 Remote Copy Failure Timeouts
There are two primary timeouts used by Remote Copy regarding link failures:
■ a link heartbeat timeout, which is used to declare an unresponsive link as down.
■ a target timeout, which is used to declare a target system down after all links have gone
down.
The value of these timeouts varies depending on the mode of the groups per target. If the
Remote Copy configuration contains asynchronous periodic mode volume groups, then the
link heartbeat timeout is 60 seconds and the target timeout is 200 seconds. If the Remote Copy
configuration contains synchronous mode volume groups, the target timeout is 15 seconds and
the heartbeat timeout is 10 seconds.
syncrcopy command to manually
Error Handling
2.37
Remote Copy User’s GuideInForm OS Version 2.3.1
When synchronous replication is used, low timeout values are used in an attempt to ensure
that replication is stopped before the host I/O begins to time out in case of a network failure. If
the network used by Remote Copy has transient problems, the lower timeout values can result
in repeated link failure alerts and the stopping of all Remote Copy volume groups. If
asynchronous periodic volume groups are in use, then this increased sensitivity is unnecessary
and the longer timeout values allow Remote Copy to continue in the presence of transient
network problems.
2.13.4 Storage Server Failures
If one of the storage servers in a Remote Copy pair fails, it is detected by the other storage
server as a concurrent failure of both communication links and is treated the same as a Double
Link Failure, as described on page 2.36. When such a failure is detected, the storage server
generates an alert.
2.13.4.1 Failure of a Target
When a target has failed and the data on the secondary volumes needs to be accessed, the role
of the groups containing those volumes must be reversed.
2.13.4.2 Failure of a Secondary Target
The failure of a secondary target in a Remote Copy pair is indistinguishable from the failure of
all communication links to the remote site. The treatment of this error condition is the same as
the handling of a complete network failure or a double link failure, as described in Double
Link Failure on page 2.36. After the secondary target comes back up, if any updates were made
while the target was down, those updates are synchronized after replication is restarted by
issuing the
startrcopygroup command on the storage server.
2.13.5 Write Errors
When a write encounters an error on the primary volume, the error is returned to the host as
an I/O failure. If a write completes successfully on the primary storage server but a write error
occurs on the backup volume, it is handled based on the replication mode and the error
handling policy for the volume group.
NOTE: See Volume Group Modes on page 2.22 for descriptions of synchronous
and asynchronous periodic volume groups.
2.38
Error Handling
InForm OS Version 2.3.1Remote Copy User’s Guide
■ If the no_fail_wrt_on_err policy is enabled for a synchronous volume group (this is the
default), the host I/O completes without error. Replication is stopped for the volume group
and a snapshot is taken of each volume in the volume group. In addition, the offset and
length of the I/O that failed is recorded in nonvolatile memory. An alert is generated to
indicate the I/O error on the secondary storage server and to report the stopping of the
volume group.
The failed I/O needs to be recorded separately because the snapshot is taken after the I/O
was written on the primary volume but before it could be completed on the backup
volume. When replication is restarted for the volume, the failed I/O is read from the
primary volume and sent to the backup. In addition, all differences between the base
volume and the snapshot taken when the replication was stopped are also sent over to
resynchronize the backup volume with the primary volume.
■ If the fail_wrt_on_err policy is enabled for a synchronous volume group, then a write
error is returned to the host when either the primary or the backup volume reports an I/O
error, but the replication is not stopped. If the host retries the failed write and subsequent
host writes continue to encounter errors while writing to the backup volume, errors
continue to be returned to the host.
■ auto_recover does not automatically recover from stopped groups as a result of write
errors. It only automatically recovers from link failures.
2.13.6 Read Errors
During the initial synchronization of a backup volume with its primary volume, Remote Copy
either reads the base volume (for synchronous volume groups) or creates a read-only snapshot
of the base volume (for asynchronous periodic volume groups). If a read request fails for any
reason, the synchronization process cancels and the secondary volume is marked
An alert is issued, indicating that the synchronization has failed. However, this does not stop
the replication. To get the volumes back in synchronization, manually issue the
command for synchronous volume groups as well as asynchronous periodic groups.
When a synchronous volume group is stopped and started after it has been fully synced, a
resynchronization is automatically started. This process compares a read-only snapshot taken
when the group was stopped with the current base volume. Any changed pages are read from
the base volume and sent over to the secondary. This type of read failure also results in the
cancellation of the resynchronization process with the secondary volume marked
The group is not stopped. To recover from this, you can manually issue the
command.
NotSynced.
syncrcopy
NotSynced.
syncrcopy
Error Handling
2.39
Remote Copy User’s GuideInForm OS Version 2.3.1
As long as you have set the synchronization period using the setrcopygroup command, an
asynchronous periodic volume group is automatically resynchronized at the scheduled
synchronization interval. When you set the resynchronization period for the primary group, it
is automatically mirrored to the secondary.
NOTE: There is no default synchronization period. Therefore, if you have not
specified a synchronization period using the setrcopygroup command,
automatic resynchronizations will not take place. The minimum interval for
periodic resynchronization is five minutes.
During the resynchronization, two read-only snapshots of the primary volume are compared
and any pages that are different are read from the newer snapshot. A failure of such a read
results in the cancellation of the resynchronization process, with the backup volume marked
Stale. At this point, the snapshots on the secondary volume are automatically promoted. The
Stale state indicates that the secondary has a valid point-in-time copy of the primary,
however the last attempt at synchronization failed. If there is no error during the next
scheduled synchronization and it completes successfully, the volume state becomes
you need to synchronize the volumes immediately rather than waiting for the next scheduled
synchronization time, you can manually issue the
syncrcopy command.
Synced. If
2.40
Other synchronization states might be indicated as follows:
◆ New - Remote copy for the volume has not started.
◆ Syncing - The secondary volume is currently being synchronized with the primary.
◆ Synced - The primary and secondary volumes are currently in sync (for periodic mode
volumes this indicates the last synchronization.
◆ NotSynced - The primary and secondary volumes are not in sync.
◆ Stopped - The primary and secondary volumes were in sync the last time the group was
started, but now may not be due to the group being stopped.
◆ Stale - The secondary has a valid point-in-time copy of the primary, however the last
attempt at synchronization failed.
Error Handling
InForm OS Version 2.3.1Remote Copy User’s Guide
TCP/IP Network
Node Pair
Node Pair
Links
Links
or
FC SAN
2.14 Requirements and Restrictions
Remote Copy requires the use of a minimum of two InServ Storage Servers.
■ For Remote Copy over IP (RCIP), each storage server in the Remote Copy configuration must
have two nodes with 3PAR-supported GigE PCI adapters (a maximum of one adapter per
node and two adapters per system). Two adapters are required for each storage server in
order to support redundant TCP/IPlinks between systems. This limitation does not apply to
1-to-N Remote Copy. For 1-to-N Remote Copy, the primary storage server can have four
RCIP interfaces; two to one secondary system and two to the other secondary system.
■ For Remote Copy over Fibre Channel (RCFC), each storage server should have a pair of Host
Bus Adapters (HBAs) installed. The HBA pair is required for load sharing and fault
tolerance. These adapters must be reserved for exclusive use by Remote Copy. The HBAs in
each storage server connect those systems through FC SAN using Fibre Channel cable
connections (see Figure 2-18).
Figure 2-18. Remote Copy Links
■ For RCIP configurations, the management interface cannot reside on the same network as
the RCIP network.
■ The maximum delay times are as follows:
Requirements and Restrictions
◆ RCIP - 150 seconds
◆ RCFC - 1.5 milliseconds
◆ RCFC over IP - 60 milliseconds
2.41
Remote Copy User’s GuideInForm OS Version 2.3.1
Systems used by Remote Copy must adhere to the following restrictions:
■ All storage servers in the configuration must be configured at the same InForm OS level.
■ Each storage server is limited to two Remote Copy interfaces.
This limitation does not apply to 1-to-N Remote Copy. For 1-to-N Remote Copy, the primary
storage server can have four RCIP interfaces; two to one secondary system and two to the
other secondary system.
■ RCFC over IP networks is only allowed for Remote Copy in asynchronous periodic mode.
In addition, the following general restrictions apply to the use of Remote Copy:
■ Self-mirroring configurations are not supported. In other words, a storage server cannot be
used to replicate its own primary volumes.
■ Multihop configurations are not supported. In other words, a primary volume group
cannot be replicated to a backup system and then replicated again from the backup system
to a third storage server.
■ Under normal operating conditions, one system can only be the primary server for one
other system.
2.42
This limitation applies mainly to N-to-1, 1-to-N, and bidirectional configurations. See N-to-1
Restrictions on page 2.43 and 1-to-N Restrictions on page 2.44 for additional details. An
exception to this rule is granted for situations where targets must be reversed during
disaster recovery.
■ All volumes used with Remote Copy must be Thinly Provisioned Virtual Volumes (TPVVs)
with snapshot space or fully provisioned virtual volumes. In addition, the following
restrictions apply:
◆ The maximum number of mirrored volumes system-wide in synchronous mode is 800.
◆ The maximum number of mirrored volumes system-wide in asynchronous periodic mode
is 2400.
◆ The maximum number of volumes per Remote Copy group in synchronous and
asynchronous periodic mode is 100.
◆ The maximum number of volumes per Remote Copy group in asynchronous periodic
mode is 2400.
◆ The maximum size of a mirrored volume is 16 TB.
Requirements and Restrictions
InForm OS Version 2.3.1Remote Copy User’s Guide
Backup Server
Primary Servers (Maximum of 4)
Direction of data
replication under
normal operating
conditions
2.14.1
N-to-1 Restrictions
In addition to those requirements and restrictions described in Requirements and Restrictions
on page 2.41, the following also apply to N-to-1 Remote Copy configurations:
■ In an N-to-1 configuration, one remote storage server participates in a maximum of four
pairs, one for each primary storage server (Figure 2-19). One of these pairs might also be
bidirectional.
Figure 2-19. N-to-1 Maximum Supported Configuration
■ N-to-1 Remote Copy can be used in both Remote Copy over IP (RCIP) and Remote Copy over
Fibre Channel (RCFC) configurations.
■ Any storage server acting as a backup for two or more storage servers (acting as primary
systems) must have at least four controller nodes.
■ The backup server in the N-to-1 configuration must use only S-Class (P4) or T-Class controller
nodes. E-Class and two node F-Class controller nodes cannot be used.
■ Only asynchronous periodic mode volume groups are supported.
Requirements and Restrictions
2.43
Remote Copy User’s GuideInForm OS Version 2.3.1
Primary Server
Backup Servers (Maximum of 2)
Direction of data
replication under
normal operating
conditions
2.14.2 1-to-N Restrictions
In addition to those requirements and restrictions described in Requirements and Restrictions
on page 2.41, the following also apply 1-to-N Remote Copy configurations:
■ In a 1-to-N configuration, one primary storage server participates in a maximum of two
pairs, one for each backup storage server (Figure 2-20). One of these pairs might also be
bidirectional.
2.44
Figure 2-20.
■ 1-to-N Remote Copy can be used in either Remote Copy over IP (RCIP) and Remote Copy
1-to-N Maximum Supported Configuration
over Fibre Channel (RCFC) configurations.
■ For RCFC configurations, any storage server acting as a primary must have at least four
controller nodes. For RCIP configurations, the primary and backup servers must have a
minimum of two controller nodes each.
■ Primary systems with only two controller nodes must be InServ S, T, or F-Class Storage
Servers.
■ InServ E-Class Storage Servers can only serve as backup servers.
Requirements and Restrictions
InForm OS Version 2.3.1Remote Copy User’s Guide
2.14.3 Synchronous Long Distance Restrictions
For Synchronous Long Distance Remote Copy configurations, a maximum of two storage
servers can be used as backup servers for one primary storage server. Additionally, the
following restrictions apply to Synchronous Long Distance Remote Copy configurations:
■ All storage servers must have at least four controller nodes each.
■ Only unidirectional functionality allowed between Remote Copy pairs.
■ Only one synchronous and one asynchronous periodic target is allowed.
■ Only two targets are allowed.
Requirements and Restrictions
2.45
Remote Copy User’s GuideInForm OS Version 2.3.1
2.46
Requirements and Restrictions
InForm OS Version 2.3.1Remote Copy User’s Guide
3
Remote Copy Setup
In this chapter
3.1 Available Setup Methods3.2
3.2 Preparing for Setup3.2
3.3 Gathering Necessary Information3.3
3.4 Setting the Remote Copy Transport Layer3.8
3.4.1 Setting Up Remote Copy Over IP3.8
3.4.2 Setting Up the Remote Copy Interface for RCIP3.9
3.4.2.1 Setting the Gateway3.9
3.4.2.2 Verifying Connectivity3.10
3.4.2.3 Increasing MTU (Optional)3.11
3.4.3 Setting Up Remote Copy Over Fibre Channel3.13
3.4.4 Setting Up RCFC Over an IP Network3.15
3.5 Setting Up Remote Copy3.16
3.5.1 Setting Up Unidirectional Remote Copy3.17
3.5.2 Setting Up Bidirectional Remote Copy3.25
3.6 Synchronous Long Distance Remote Copy Setup3.28
Remote Copy Setup
3.1
Remote Copy User’s GuideInForm OS Version 2.3.1
3.7 Initial Synchronization Using Tape Backup3.38
3.1 Available Setup Methods
This chapter explains how to set up and configure 3PAR Remote Copy. Use these instructions if
you are setting up Remote Copy for the first time. However, depending on your level of
experience, there are three methods you can choose to set up a Remote Copy configuration:
■ If this is your first experience with setting up Remote Copy, particularly for N-to-1 or
1-to N configurations, or when N-to-1 or 1-to-N and bidirectional Remote Copy
configurations are planned, you might consider using the Preparing for Setup and
Gathering Necessary Information sections that follow.
■ Users familiar with 3PAR’s Remote Copy setup can use the 3PAR Remote Copy Setup Tool
for RCIP (320-1461) or the 3PAR Remote Copy Setup Tool for RCFC (320-1472) for a 1-to-1
configuration to help configure, test and visualize the setup for Remote Copy over IP (RCIP)
and Remote Copy over Fibre Channel (RCFC), respectively. See the Document Control
System (DCS) to obtain these documents and follow the directions found in the document
that best suits your configuration.
■ Expert users might prefer to use Appendix A, Quick Setup Guide.
The final section of this chapter demonstrates how to perform an initial synchronization when
using tape rather than synchronizing a secondary volume with the primary volume over the
Remote Copy links.
3.2 Preparing for Setup
Before proceeding with Remote Copy setup, review Requirements and Restrictions on
page 2.41. The following guidelines and restrictions also apply to Remote Copy setup:
■ Synchronous mode replication is recommended for distances no greater than a
Metropolitan Area Network (MAN). A MAN is a region larger than that covered by a Local
Area Network (LAN) but smaller than that covered by a Wide Area Network (WAN).
■ The physical connections between all storage servers used with Remote Copy must be
through an IP-capable network, point to point, or an FC SAN network.
3.2
Available Setup Methods
InForm OS Version 2.3.1Remote Copy User’s Guide
■ When configuring the Gigabit Ethernet (GigE) interface, the GigE interface and the
management Ethernet port of an InServ Storage Server controller node should be on
different IP subnets. If they are configured on the same subnet, Remote Copy packets
might go over the management port and will not be available to the Remote Copy
software.
■ All virtual volumes used with Remote Copy must be Thinly Provisioned Virtual Volumes
(TPVVs) with snapshot space, which require the use of 3PAR Thin Provisioning, or fully
provisioned virtual volumes. Both of these volume types draw space from a Common
Provisioning Group (CPG). See Remote Copy and Thin Provisioning on page 2.33 and the
3PAR InForm OS Concepts Guide for details about TPVVs, fully provisioned virtual volumes,
and CPGs.
■ You must create the volumes on each storage system in the Remote Copy configuration
separately by using the normal volume creation commands before creating volume groups
for use with Remote Copy. The volumes on the primary server must be the same size as
those on the backup server. See the 3PAR InForm OS CLI Administrator’s Manual for
instructions on creating volumes.
3.3 Gathering Necessary Information
Determine the setup that best meets your desired configuration (N-to-1, 1-to-N, bidirectional,
unidirectional, and so on) and use Figure 3-1 through Figure 3-4 to help gather the necessary
information to continue your Remote Copy setup as described in the sections that follow.
For 1-to-N and N-to-1 configurations, or when N-to-1 and bidirectional Remote Copy are used
within the same configuration, you can use Figure 3-3 or Figure 3-4 (depending on Remote
Copy transport layer) to supply information for additional systems. You might want to make a
copy of Figure 3-3 or Figure 3-4 (depending on the Remote Copy transport layer) in order to fill
in information for all links.
In preparation for setting up Remote Copy, obtain the IP addresses or 64-bit World Wide Name
(WWN) address information for the interfaces of the storage servers from your network
administrator and be prepared to provide 3PAR technicians with other details about the
network connections, as necessary. Contact your network administrator to obtain the gateway
IP addresses, netmask, and IP addresses for each link. In addition, you must verify that the
firewall settings allow access to port 5785 between the Remote Copy systems.
Gathering Necessary Information
3.3
Remote Copy User’s GuideInForm OS Version 2.3.1
Network
System 1
Backup
Gateway
Active
Gateway
System 2
Basic Remote Copy Pair Information:
TCP/IP
After the Remote Copy configuration is complete, you can use the InForm CLI
showport -rcip command or the showport -rcfc command to confirm the addresses you
specified during the setup.
Use the following figures to help visualize and organize your desired Remote Copy
configuration:
Figure 3-1. Network and Remote Copy Over IP Setup Information (Initial Pair)
3.4
Gathering Necessary Information
InForm OS Version 2.3.1Remote Copy User’s Guide
N:S:P
IP Address
Netmask
Target Name:
Target Name:
Gateway IP
1. Remote Copy Pair Transport Layer:
Node*
Target Interface IP
2. Remote Copy Pair Link Configuration:
Node*
MTU
Target Interface IP
System 1:
System 2:
System 1:
*The Node from which the Link Originates.
Speed of Gigabit Interface:
Left HBA:
Right HBA:
N:S:P
IP Address
Netmask
Gateway IP
MTU
System 2:
Speed of Gigabit Interface:
Left HBA:
Right HBA:
(e.g. System 2)
(e.g. System 1)
(e.g. System 2 Left HBA IP)
(e.g. System 2 Right HBA IP)
(e.g. System 1 Left HBA IP)
(e.g. System 1 Right HBA IP)
Use the information gathered from the previous boxes to populate the following boxes.
Figure 3-2. Network and Remote Copy Over IP Setup Information (Initial Pair)
Gathering Necessary Information
3.5
Remote Copy User’s GuideInForm OS Version 2.3.1
Target Name:
1. Remote Copy Transport Information for Additional Systems:
Node*
2. Additional Link Configuration:
Target Interface IP
System: _____ (e.g. System 1)
*The Node from which the Link Originates.
N:S:P
IP Address
Netmask
Gateway IP
MTU
System 3:
Speed of Gigabit Interface:
Left HBA:
Right HBA:
N:S:P
IP Address
Netmask
Gateway IP
MTU
System 4:
Speed of Gigabit Interface:
Left HBA:
Right HBA:
N:S:P
IP Address
Netmask
Gateway IP
MTU
System 5:
Speed of Gigabit Interface:
Left HBA:
Right HBA:
(e.g. System 1 Left HBA IP)
(e.g. System 1 Right HBA IP)
Node*
Target Interface IP
(e.g. System 4 Left HBA IP)
(e.g. System 4 Right HBA IP)
Node*
Target Interface IP
(e.g. System 3 Left HBA IP)
(e.g. System 3 Right HBA IP)
Node*
Target Interface IP
(e.g. System 1 Left HBA IP)
(e.g. System 1 Right HBA IP)
(e.g. System 3)
Target Name:
System: _____ (e.g. System 3)
(e.g. System 1)
Target Name:
System: _____ (e.g. System 1)
(e.g. System 4)
Target Name:
System: _____ (e.g. System 4)
(e.g. System 1)
(and so on for System 5)
Use the information gathered from the previous boxes to populate the following boxes.
3.6
Gathering Necessary Information
Figure 3-3. Network and Remote Copy Over IP Setup Information (Additional Systems)
InForm OS Version 2.3.1Remote Copy User’s Guide
N:S:P
Network
WWN
System 1
System 2
Target Name:
Target Name:
Basic Remote Copy Pair Information:
1. Remote Copy Pair Transport Layer:
N:S:P
Target Interface WWN
2. Remote Copy Pair Link Configuration:
N:S:P
Target Interface WWN
System 1:
System 2:
System 2:System 1:
N:S:P
WWN
Fibre Channel SAN
Left HBA
Right HBA
Left HBA
Right HBA
e.g. System 2 Left HBA WNN
e.g. System 2 Right HBA WWN
e.g. System 1 Left HBA WWN
e.g. System 1 Right HBA WWN
e.g. System 2
e.g. System 1
Use the information gathered from the previous boxes to populate the following boxes.
N:S:P
N:S:P
Figure 3-4. Network and Remote Copy Over Fibre Channel Setup Information
Gathering Necessary Information
3.7
Remote Copy User’s GuideInForm OS Version 2.3.1
3.4 Setting the Remote Copy Transport Layer
Prior to setting up Remote Copy, you must set up the Remote Copy transport layer as described
in this section. The method for setting the RC transport layer differs depending on whether the
configuration uses Remote Copy over IP (RCIP) or Remote Copy over Fibre Channel (RCFC).
However, note that this is the only part of the Remote Copy setup that differs between RCIP or
RCFC.
3.4.1 Setting Up Remote Copy Over IP
For RCIP configurations, each link between a Remote Copy pair is a physical link between a
controller node on one storage server and a controller node on the other storage server in the
pair. These physical links use a GigE port from each of the two nodes in the storage servers that
belong to the Remote Copy pair. Typically, these ports are configured during the initial setup
of the storage system.
If setup did not occur, or the configuration has changed, you will need to reconfigure the ports
before you can begin setting up Remote Copy. Ports are configured using the
controlport rcip command. See Appendix B, Remote Copy Commands for more
information on this command. See the InForm OS Command Line Interface Reference for
complete information on using this command.
3.8
CAUTION: When configuring the GigE interfaces, the GigE interface and the
management Ethernet port of an InServ Storage Server controller node should be
on different IP subnets. If they are configured on the same subnet, Remote Copy
packets might go over the management port and would not be available to the
Remote Copy software.
Setting the Remote Copy Transport Layer
InForm OS Version 2.3.1Remote Copy User’s Guide
3.4.2 Setting Up the Remote Copy Interface for RCIP
For RCIP, you must set up the Remote Copy GigE interfaces for each storage server in the
Remote Copy configuration.
Set the Remote Copy interface for all systems based on the information gathered from
Figure 3-1 as follows:
1 To set up the Remote Copy interfaces, use the following command for the Ethernet ports
on each storage server:
# controlport rcip addr <port_IP> <netmask> <N: S:P>
Are you sure you want to change the address for <N:S:P>?
select q=quit y=yes n=no: y
Remote Copy interface change successful.
In order to issue this command on a storage server, you will need an IP address assigned to
the Ethernet port (
location of an primary server Ethernet port, expressed as node:slot:port (<N:S:P>).
NOTE: Remember to set the interfaces for both ports on each storage server to be
used in the Remote Copy configuration.
<port_IP>), the netmask to be assigned (<netmask>), and the
2 Issue the
and netmask are set correctly.
# showport -rcip
N:S:P State ---HwAddr--- IPAddr Netmask Gateway MTU
Rate Duplex AutoNeg
0:3:1 ready 000423CBF68C 10.100.24.107 255.255.255.0 -- 1500
1Gbps Full Yes
1:3:1 ready 000423CBF693 10.101.24.107 255.255.255.0 -- 1500
1Gbps Full Yes
showport -rcip command again on each system to verify that the IP addresses
3.4.2.1 Setting the Gateway
If all of the storage servers to be used in the Remote Copy configuration are on the same
subnet, skip to Verifying Connectivity on page 3.10.
If the storage servers are not on the same subnet (the most common situation), you need to set
the gateways for each system as described in the steps that follow.
Setting the Remote Copy Transport Layer
3.9
Remote Copy User’s GuideInForm OS Version 2.3.1
1 Use the following command to set the system gateways for the Ethernet ports on storage
server:
# controlport rcip gw <gateway_IP> <N:S:P>
Are you sure you want to change the gateway for <N:S:P>?
select q=quit y=yes n=no: y
Remote Copy interface change successful.
In order to issue this command on a storage server, you will need the gateway IP address to
assign to the local Ethernet port (
(
<N:S:P>).
NOTE: Remember to set the interfaces for both ports on each storage server to be
used in the Remote Copy configuration.
2 For each storage server in the Remote Copy configuration, verify that the gateways have
been set correctly by checking the
output:
<gateway_IP>) and the location of a local Ethernet port
Gateway column in the showport -rcip command
3.10
# showport -rcip
N:S:P State ---HwAddr--- IPAddr Netmask Gateway MTU
Rate Duplex AutoNeg
0:3:1 ready 000423CBF68C 10.100.24.107 255.255.255.0 10.100.24.1 1500
1Gbps Full Yes
1:3:1 ready 000423CBF693 10.101.24.107 255.255.255.0 10.101.24.1 1500
1Gbps Full Yes
3.4.2.2 Verifying Connectivity
After you have set the Remote Copy interfaces between the two storage servers by configuring
the GigE ports, verify connectivity from any storage server as follows:
# controlport rcip ping 10.101.24.108 0:3:1
PING 10.101.24.108 (10.101.24.108) from 10.100.24.107 : 56(84) bytes of
data.
64 bytes from 10.101.24.108: icmp_seq=1 ttl=253 time=0.325 ms
--- 10.101.24.108 ping statistics --1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.325/0.325/0.325/0.000 ms
Setting the Remote Copy Transport Layer
InForm OS Version 2.3.1Remote Copy User’s Guide
This command performs a ping from the specified interface (<N:S:P>) to the specified IP
address on the target system (
<IP_address>). Issue this command twice (once for each
interface), for each system in the configuration.
NOTE: Because of the spanning tree and MAC address learning, you might need
to ping each server several times before all tables are updated and the ping
succeeds.
If the
controlport rcip ping command does not work:
◆ Verify that the GigE cards are all in the ready state using the showport -rc
command.
◆ Verify all IP addresses, netmasks, gateways, and physical connections.
◆ Use controlport rcip ping to test connectivity between both IP interfaces on
both systems.
◆ Verify the MTU settings are correct. See Increasing MTU (Optional) on page 3.11 for
more information.
◆ Verify the controlport rcip speed settings. See controlport on page B.9 for
details.
If the pings do not succeed, seek help from a Network Administrator.
CAUTION: Do not proceed with the rest of the Remote Copy setup until pings
succeed between the GigE ports in the two storage servers.
3.4.2.3 Increasing MTU (Optional)
Depending on the network setup, you might be able to increase the maximum transmission
unit (MTU) size of the GigE interfaces to increase throughput between storage servers. The
throughput and CPU utilization of GigE adapters can be improved if the adapters are
configured to use jumbo frames (9000-bytes frames) rather than the default 1500-byte frames.
When supported by the network, use an MTU value of 9000. However, note that this setting
requires that the IP network support jumbo frames end-to-end (along the entire path between
the two storage servers).
Setting the Remote Copy Transport Layer
3.11
Remote Copy User’s GuideInForm OS Version 2.3.1
NOTE: Depending on workload, you might see as much as a 50 percent increase in
throughput by using a 9000 byte MTU setting as opposed to a 1500 byte MTU
setting.
If the MTU value of 9000 is supported by the network, increase the MTU as follows:
:
# controlport rcip mtu 9000 <N:S:P>
Remote Copy change successful.
where <N:S:P> represents the location of a GigE port, expressed as node:slot:port.
NOTE: Remember to change the MTU setting for both ports in each storage
server to be used in the Remote Copy configuration.
2 After setting the MTU to 9000 bytes for all GigE ports, verify that the network supports the
increased MTU size from any storage server as follows:
1 For the GigE ports on each storage server, increase the MTU size to 9000 as follows:
This command performs a ping from the specified interface (
<N:S:P>) to the specified IP
address on the target system (<IP_address>) while specifying a packet size of 8972 and
preventing fragmentation of packets. Issue this command twice (once for each interface),
for each system in the configuration.
3.12
NOTE: Because of the spanning tree and MAC address learning, you might need
to ping each server several times before all tables are updated and the ping
succeeds.
◆ If the pings succeed, the network supports the new setting. Proceed to Setting Up
Remote Copy on page 3.16.
Setting the Remote Copy Transport Layer
InForm OS Version 2.3.1Remote Copy User’s Guide
◆ If the pings fail, issue another set of pings without specifying a size. If this second
attempt succeeds, the new setting is not supported and you should revert to 1500 MTU
before continuing on to Setting Up Remote Copy on page 3.16.
3.4.3 Setting Up Remote Copy Over Fibre Channel
For Remote Copy over Fibre Channel (RCFC) configurations, each link between a Remote Copy
pair is a physical link between a controller node on one storage server and a controller node
on the other storage server in the pair. These physical links use up to a 4 Gbps Fibre Channel
adaptor port from each of the two nodes in the storage servers that belong to the Remote
Copy pair. Typically, these ports are configured during the initial setup of the storage system.
If the setup did not occur, or the configuration has changed, you will need to reconfigure the
ports before you can begin setting up Remote Copy. Ports are configured using the
controlport rcfc command. See Appendix B, Remote Copy Commands for more
information on this command. See the InForm OS Command Line Interface Reference for
complete information on using this command.
CAUTION: Each pair of Remote Copy over Fiber Channel links must exist in an
exclusive zone. Fabric zones cannot be shared.
RCFC requires a dedicated node pair per target. Each fibre channel interface on each node in
the primary InServ Storage Server must connect to a dedicated fibre channel interface on each
node in the backup InServ Storage Server. For example, in Figure 3-5:
■ Fibre channel interfaces on nodes 0 and 1 on the Primary InServ connect to fibre channel
interfaces on Backup InServ1 nodes 0 and 1, respectively.
■ Fibre channel interfaces on nodes 2 and 3 on the Primary InServ connect to fibre channel
interfaces on Backup InServ2 nodes 0 and 1, respectively.
Setting the Remote Copy Transport Layer
3.13
Remote Copy User’s GuideInForm OS Version 2.3.1
Primary InServ
Backup InServ1
Backup InServ2
Node 0
Node 1
Node 2 Node 3
Node 0 Node 1
Node 1
Node 0
Figure 3-5. Remote Copy over Fibre Channel Dedicated Node Pairs
3.4.3.1 Setting Up the Remote Copy Interface for RCFC
For Remote Copy over Fibre Channel (RCFC), you must set up the Remote Copy interface for
each storage server in the Remote Copy configuration.
Set the Remote Copy interface for all systems based on the information gathered from
Figure 3-4 as follows:
3.14
1 Set the connection type for each Fibre Channel adaptor port on the primary and backup
storage servers as shown in the following example:
# controlport config rcfc -ct point -f <N:S:P>
The Fibre Channel adaptor port, as specified by
an RCFC connection (
config rcfc) in point-to-point mode (-ct point). See the InForm
<N:S:P> (Node:Slot:Port), is configured as
OS Command Line Interface Reference for details about the controlport config
command.
2 Issue the
showport -rcfc command to obtain the RCFC port positions.
# showport -rcfc
Repeat steps 1 and 2 for each storage server to obtain all four RCFC port positions to be
used in the Remote Copy configuration. Record this information in Figure 3-4 on page 3.7.
Setting the Remote Copy Transport Layer
InForm OS Version 2.3.1Remote Copy User’s Guide
3 Verify the Fibre Channel transports are present on the Fibre Channel adaptor ports on each
system and the
command.
# showrctransport -rcfc
N:S:P Peer_Node_WWN Peer_Port_WWN State
3:2:1 2FF70002AC00005F 23410002AC00005F new
1:4:1 2FF70002AC00005F 21510002AC00005F new
4 Initialize the Fibre Channel adaptor ports on each storage server as shown in the following
example:
# controlport rcfc init -f <N:S:P>
State column displays ready by issuing the showrctransport -rcfc
5 Issue the
Fibre Channel adaptor ports are ready to begin setting up Remote Copy.
# showrctransport -rcfc
6 If any port displays that the
the State column displays ready as shown in the following example.
# creatercopytarget <target_name> IP <n:s:p>:<link_IP_addr>
<N:S:P>:<link_IP_addr>
◆ <target_name> is the target name (for example, InServ2)
Setting Up Remote Copy
◆ IP states that this is an IP link.
◆ <n:s:p> specifies the node on InServ1 that contains the IP connections to the target.
InForm OS Version 2.3.1Remote Copy User’s Guide
◆ <link_IP_addr> is the link IP address for the target system (for example,172.16.1.12
and 172.16.2.12).
NOTE: The command for RCFC target creation is:
#creatercopytarget <target_name> FC <node_WWN> <n:s:p:WWN>
<n:s:p:WWN>
NOTE: You can only configure one link per target definition per node. All
examples in this guide use the first pattern illustrated in Figure 2-9 on page 2.17,
where links are created between the lowest numbered node in each member of
the Remote Copy pair and the highest numbered node in each member of the
Remote Copy pair.
3.5.1.2 Setting Up the Backup Server
The commands used to set up the backup storage server (InServ2) are the same as those used
to set up the primary server (InServ1), except that they create the links in the opposite
direction.
To set up the backup storage server (InServ2)
1 Start Remote Copy as follows:
On InServ2:
# startrcopy
2 Define a target for InServ2:
On InServ2:
#
creatercopytarget <target_name> IP <n:s: p>:<link_IP_addr>
<node>:<link_IP_addr>
◆ <target_name> is the target name (in this example, InServ1)
◆ IP states that this is an IP link.
◆ <n:s:p> specifies the node that contains the IP connections to the target.
Setting Up Remote Copy
3.19
Remote Copy User’s GuideInForm OS Version 2.3.1
◆ <link_IP_addr> is the link IP address for the target system (for example,
172.16.1.11 and 172.16.2.11).
3 Check that the links have come up between the primary and backup storage servers by
using the
showrcopy command on each storage server:
NOTE: For complete details on the showrcopy command, including valid
synchronization, link, and group states, see showrcopy on page B.57.
On InServ2:
# showrcopy
Remote Copy System Information
Status: Started, Normal
Target Information
Name ID Type Status Options Policy
108 32 IP ready mirror_config
3.20
Link Information
Target Node Address Status Options
108 0 10.100.24.108 Up
108 1 10.101.24.108 Up
receive 0 10.100.24.108 Up
receive 1 10.101.24.108 Up
Group Information
Name Target Status Role Mode Options
group1 108 Started Primary Sync
auto_recover,fail_wrt_on_err
LocalVV ID RemoteVV ID SyncStatus LastSyncTime
vv0.0 1974 vv0.0 498 Synced NA
vv0.1 1975 vv0.1 499 Synced NA
vv0.2 1976 vv0.2 500 Synced NA
vv0.3 1977 vv0.3 501 Synced NA
vv0.4 1978 vv0.4 502 Synced NA
It might take several minutes for the links to come up between the two storage servers.
When the Target Information shows a Status of ready, and the Link Information
Setting Up Remote Copy
InForm OS Version 2.3.1Remote Copy User’s Guide
returns a Status of Up, the links have come up. Be sure to check that links are up on both
systems (the previous example shows output for InServ2 only).
NOTE: If there are any problems with the way that the links were created, use the
dismissrcopylink <target_name> <node>:<IP_address> command to
remove links and then try again.
After all links come up between the Remote Copy pair, you can proceed with creating
volume groups on the systems.
3.5.1.3 Creating Remote Copy Volume Groups
Before creating volume groups, you must create volumes on each storage server separately
through the normal volume creation commands. The volumes must be the same size, though
they can have different RAID levels and availability levels (e.g., Cage versus Magazine). In
addition, they must be TPVVs or fully provisioned virtual volumes. See the 3PAR InForm OS CLI
Administrator’s Manual for instructions on creating volumes.
NOTE: When a TPVV is configured as a primary volume in a Remote Copy group,
the secondary volume should have no data written to it prior to adding it the
Remote Copy group, or it must match the primary volume in order for the primary
and secondary volumes to match during initial synchronization.
Follow these steps to create the Remote Copy volume groups on the primary and backup
storage servers.
1 Set up a volume group on the primary storage server as follows:
<VV_name> is the name of the virtual volume (vv1) to be added to the primary volume
group (Group1) specified by the
<group_name> variable. You also need to specify the
<target_name> (for example, InServ2) and the name of the corresponding volume on
the target system (for this example, vv1_remote).
The
admitrcopyvv command adds the specified volume (vv1) to the primary volume
group (Group1). At the same time, it adds the specified volume on the target system
(vv1_remote
page 3.21 (Group1.r112).
) to the corresponding secondary volume group created in step 1 on
3.5.1.4 Start Copying
The final step in setting up unidirectional Remote Copy for a single Remote Copy pair is to start
replication for the volume groups.
NOTE: To perform the initial synchronization using tape backup, use the –-
nosync
that the volumes are already synchronized. This command can also be used for
newly created volumes that have not yet been exported to the hosts. See Initial
Synchronization Using Tape Backup on page 3.38 for complete instructions.
1 On the primary storage server, start replication for a volume group as follows:
On InServ1:
#
startrcopygroup <group_name>
where
<group_name> is the name of the primary volume group (Group 1).
The
startrcopygroup command is mirrored from InServ1 to InServ2 and begins the
copying of data. Because this is the first time this command has been run for this volume
group, the existing data is copied from the primary to the secondary volume group.
option of the startrcopygroup command. This command assumes
Setting Up Remote Copy
3.23
Remote Copy User’s GuideInForm OS Version 2.3.1
2 Issue the showrcopy command again to verify that the groups are synchronizing. The
groups are completely synchronized after the SyncStatus is Synced.
On InServ1:
# showrcopy
...
Group Information
Name Target Status Role Mode Options
sync_group_1 InServ2 Started Primary Sync
LocalVV ID RemoteVV ID SyncStatus LastSyncTime
localvv.0 391 remotevv.0 351 SyncedNA
localvv.1 392 remotevv.1 352 SyncedNA
3.24
Setting Up Remote Copy
InForm OS Version 2.3.1Remote Copy User’s Guide
3.5.1.5 Setting Up Additional Remote Copy Pairs
For configurations that use multiple Remote Copy pairs (also known as N-to-1 or 1-to-N type
configurations), after following the instructions in Setting Up Unidirectional Remote Copy on
page 3.17, repeat the entire process for each additional pair, up to a maximum of two
(for 1-to-N configurations) or four (for N-to-1 configurations) Remote Copy pairs.
Note that for N-to-1 configurations all pairs must share the same secondary storage server, and
for 1-to-N configurations all pairs must share the same primary storage server. See N-to-1
Restrictions on page 2.43 and 1-to-N Restrictions on page 2.44 for additional information and
restrictions regarding configurations with multiple Remote Copy pairs.
NOTE: The transport layer only needs to be set once per system. However, you
should verify the connectivity for each pair in the configuration.
3.5.2 Setting Up Bidirectional Remote Copy
In this section, as illustrated by Figure 3-7 on page 3.26:
■ The members of the Remote Copy pair are an eight-node cluster called InServ1 and a four-
node cluster called InServ2.
■ A virtual volume on InServ1 called vv1 is the only member of Group1.
■ A virtual volume on InServ2 is called vv1_remote and is the only member of
Group1.r112, the secondary group that is automatically created at the same time as
Group1.
Note that vv1 and vv1_remote must be the same size.
■ A virtual volume on InServ2 called vv2 is the only member of Group2.
■ A virtual volume on InServ1 is called vv2_remote and is the only member of
Group2.r125, the secondary group that is automatically created at the same time as
Group2.
Note that vv2 and vv2_remote must be the same size.
<VV_name> is the name of the virtual volume (vv2) to be added to the primary
volume group (Group2). You also need to specify the target name
that volume group (InServ1)and the name of the secondary volume (vv2.remote).
periodic mode in step 1, you might want to set up automatic
h to denote hours, or d to denote days, <target_name> is the target
<target_name> for
Setting Up Remote Copy
3.27
Remote Copy User’s GuideInForm OS Version 2.3.1
The admitrcopyvv command adds the specified volume (vv2) to the primary volume
group (Group2). At the same time, it also adds the secondary volume (vv2.remote) to
the corresponding secondary volume group on the target system (Group2.r125).
4 Start replication for the volume groups created in step 1:
On InServ2:
# startrcopygroup <group_name>
where
<group_name> is the name of the primary volume group (Group2).
The
startrcopygroup command begins the copying of data. Because this is the first time
this command has been run for this volume group, the existing data is copied from the
primary to the secondary volume group.
3.6 Synchronous Long Distance Remote Copy Setup
Use the procedures described in this section to set up Synchronous Long Distance Remote Copy
on your storage servers. See Synchronous Long Distance Configuration on page 2.12 and
Requirements and Restrictions on page 2.41 for information about Synchronous Long Distance
Remote Copy setups.
3.28
In this section, as illustrated by Figure 3-8:
■ The primary system is a four-node cluster called InServA and the backup systems are four-
node clusters called InServB and InServC.
■ InServA is the primary storage system
◆ Target name: InServB
◆ Links:
◆ Node 2 IP address 10.100.33.195
◆ Node 3 IP address 10.101.33.195
◆ Target name: InServC
◆ Links:
◆ Node 2 IP address: 10.100.33.63
◆ Node 3 IP address: 10.101.33.63
Synchronous Long Distance Remote Copy Setup
InForm OS Version 2.3.1Remote Copy User’s Guide
■ InServB is the backup storage system
◆ Target name: InServA
◆ Links:
◆ Node 2 IP address 10.100.33.96
◆ Node 3 IP address 10.101.33.96
◆ Target name: InServC
◆ Links:
◆ Node 0 IP address 10.100.33.63
◆ Node 1 IP address 10.101.33.63
■ InServC is the backup storage system
◆ Target name: InServA
◆ Links:
◆ Node 2 IP address 10.100.33.96
◆ Node 3 IP address 10.101.33.96
◆ Target name: InServB
◆ Links:
◆ Node 2 IP address 10.100.33.195
◆ Node 3 IP address 10.101.33.195
■ Synchronous Remote Copy is used between InServA and InServB. The following WWNs
are used:
◆ InServA:
◆ Node WWN: 2FF70002AC000060
◆ Node 0 WWN: 20110002AC000060
◆ Node 1 WWN: 21320002AC000060
◆ InServB:
◆ Node WWN: 2FF70002AC0000C3
Synchronous Long Distance Remote Copy Setup
3.29
Remote Copy User’s GuideInForm OS Version 2.3.1
IP Network
InServA
testgroup
testvv.A.0
Targets: InServB
Primary Group:
testgroup.r96
Secondary Group:
Node 0
Node 1
InServC
Target: InServA
10.100.33.63
10.101.33.63
InServC
Node 2
Node 3
InServB
Target: InServA
10.100.33.195
10.101.33.195
Synchronous Remote Copy between
InServA and InServB
Asynchronous periodic Remote Copy
*
*
testvv.B.0
testvv.A.1
testvv.A.2
testvv.B.1
testvv.B.2
testvv.C.0
testvv.C.1
testvv.C.2
between InServA and InServC
FC Network
InServB
InServC
Node 0
Node 1
Node 2
Node 3
10.100.33.96
10.101.33.96
Node 0
Node 1
Node 2
Node 3
◆ Node 0 WWN: 20210002AC0000C3
◆ Node 1 WWN: 21210002AC0000C3
■ Asynchronous Periodic Remote Copy is used between InServA and InServC.
3.30
Synchronous Long Distance Remote Copy Setup
Figure 3-8. Configuring Synchronous Long Distance Remote Copy Connections
InForm OS Version 2.3.1Remote Copy User’s Guide
3.6.1 Setting Up the Primary Server
To set up the primary storage server (InServA):
1 Start Remote Copy as follows:
On InServA:
# startrcopy
NOTE: In the following step, the creatercopytarget command must be issued
twice, once for InServB and once for InServC.
2 Define the targets on InServA:
On InServA:
# creatercopytarget <target_name> FC <node_WWN> <N:S:P>:<WWN> <N:S:P>:<WWN>
# creatercopytarget <target_name> IP <N:S:P>:<IP_addr> <N:S:P>:<IP_addr>
◆ <target_name> is the name of the Remote Copy target (for example, InServB and
InServC).
◆ FC states a Fibre Channel link.
◆ IP states an IP link.
For FC setup:
◆ <node_WWN> specifies the WWN of the target node (for example, 2FF70002AC0000C3
for InServB).
◆ <N:S:P>:<WWN> specifies the node, slot, and port of the Fibre Channel adapter port
on the primary system InServA and the WWN of the peer port on the target system
InServB (for example, 0:1:1:20210002AC0000C3 and 1:3:2:21210002AC0000C3).
For IP setup:
◆ <N:S:P>:<IP_addr> specifies the node on InServA that contains the IP connections
to the target (for example, 2:4:1and 3:4:1) and the IP address (
<IP_addr>) for the
target system (for example,10.100.33.63 and 10.101.33.63 for InServC).
Synchronous Long Distance Remote Copy Setup
3.31
Remote Copy User’s GuideInForm OS Version 2.3.1
3.6.2 Setting Up the Backup Servers
The commands used to set up the backup storage servers (InServB and InServC) are nearly
the same as those used to set up the primary server (InServA), except that they create the
links in the opposite direction and create a second target for failover purposes.
To set up the backup storage servers (InServB and InServC):
1 Start Remote Copy on InServB as follows:
On InServB:
# startrcopy
2 Define a target for InServB:
On InServB:
#
creatercopytarget <target_name> FC <node _WWN> <N:S:P>:<WWN> <N:S:P>:<WWN>
# creatercopytarget <target_name> IP <N:S:P>:<IP_addr> <N:S:P>:<IP_addr>
◆ <target_name> is the name of the Remote Copy target (for example, InServA and
InServC).
3.32
◆ FC states a Fibre Channel link.
◆ IP states an IP link.
For FC setup:
◆ <node_WWN> specifies the WWN of the target node (for example, 2FF70002AC0000C3
for InServA).
◆ <N:S:P>:<WWN> specifies the node, slot, and port of the Fibre Channel adapter port
on the primary system InServA and the WWN of the peer port on the target system
InServA (for example, 0:1:1:20110002AC000060 and 1:3:2:21320002AC000060).
For IP setup:
◆ <N:S:P>:<IP_addr> specifies the node, slot, and port on InServB that contains the
IP connections to the target (for example, 2:4:1 and 3:4:1) and the IP address
(
<IP_addr>) for the target system (for example,10.100.33.63 and 10.101.33.63 for
InServC).
Synchronous Long Distance Remote Copy Setup
InForm OS Version 2.3.1Remote Copy User’s Guide
3 Define a second target for InServB:
On InServB:
#
creatercopytarget <target_name> IP <N:S: P>:<link_IP_addr>
<N:S:P>:<link_IP_addr>
◆ <target_name> is the name of the Remote Copy target (for example, InServC).
◆ IP states that this is an IP link.
◆ <node> specifies the node that contains the IP connections to the target (for example
2:4:1and 3:4:1).
◆ <link_IP_addr> is the link IP address for the target system (for example,
10.100.33.96 and 10.101.33.96).
4 Start Remote Copy on InServC as follows:
On InServC:
# startrcopy
5 Define a target for InServC:
On InServC:
#
creatercopytarget <target_name> IP <N:S: P>:<IP_addr> <N:S:P>:<IP_addr>
◆ <target_name> is the name of the Remote Copy target (for example, InServA).
◆ IP states that this is an IP link.
◆ <N:S:P> specifies the node that contains the IP connections to the target (for example
0:2:1 and 1:2:1).
◆ <IP_addr> is the link IP address for the target system (for example, 10.100.33.96
and 10.101.33.96).
6 Define a second target for InServC:
On InServC:
#creatercopytarget <target_name> IP <N:S:P>:<IP_addr> <N:S:P>:<IP_addr>
◆ <target_name> is the name of the Remote Copy target (for example, InServB).
Synchronous Long Distance Remote Copy Setup
3.33
Remote Copy User’s GuideInForm OS Version 2.3.1
◆ IP states that this is an IP link.
◆ <node> specifies the node that contains the IP connections to the target (for example
0:1:1 and 1:3:2).
◆ <IP_addr> is the link IP address for the target system (for example, 10.100.33.195
and 10.101.33.195).
3.6.3 Verifying Synchronous Long Distance Remote Copy Setup
X To verify the Synchronous Long Distance Remote Copy setup on the primary and backup
servers, issue the
◆ On InServA:.
# showrcopy
Remote Copy System Information
Status: Started, Normal
Target Information
Name ID Type Status Options Policy
InServB 6 FC ready 2FF70002AC0000C3 mirror_config
InServC 7 IP ready mirror_config
showrcopy command on InServA, InServB, and InServC.
3.34
Link Information
Target Node Address Status Options
InServB 0:1:1 20210002AC0000C3 Up
InServB 1:3:2 21210002AC0000C3 Up
InServC 2:4:1 10.100.33.63 Up
InServC 3:4:1 10.101.33.63 Up
receive 0:1:1 20210002AC0000C3 Up
receive 1:3:2 21210002AC0000C3 Up
receive 2:4:1 receive Up
receive 3:4:1 receive Up
Synchronous Long Distance Remote Copy Setup
InForm OS Version 2.3.1Remote Copy User’s Guide
◆ On InServB:
# showrcopy
Remote Copy System Information
Status: Started, Normal
Target Information
Name ID Type Status Options Policy
InServA 4 FC ready 2FF70002AC000060 mirror_config
InServC 5 IP ready mirror_config
Link Information
Target Node Address Status Options
InServA 0:2:1 20110002AC000060 Up
InServA 1:2:1 21320002AC000060 Up
InServC 2:4:1 10.100.33.63 Up
InServC 3:4:1 10.101.33.63 Up
receive 0:2:1 20110002AC000060 Up
receive 1:2:1 21320002AC000060 Up
receive 2:4:1 receive Up
receive 3:4:1 receive Up
◆ On InServC:
# showrcopy
Remote Copy System Information
Status: Started, Normal
Target Information
Name ID Type Status Options Policy
InServA 3 IP ready mirror_config
InServB 4 IP ready mirror_config
Link Information
Target Node Address Status Options
InServA 0:2:1 10.100.33.96 Up
InServA 1:2:1 10.101.33.96 Up
InServB 0:1:1 10.100.33.195 Up
InServB 1:3:2 10.101.33.195 Up
receive 0:2:1 receive Up
receive 1:2:1 receive Up
Synchronous Long Distance Remote Copy Setup
3.35
Remote Copy User’s GuideInForm OS Version 2.3.1
3.6.4 Creating Synchronous Long Distance Remote Copy Volume Groups
NOTE: In the following procedure, it is assumed that the volumes testvv.A.x
have already been created on InServA, volumes
created on InServB, and volumes testvv.C.x have already been created on
InServC.
NOTE: In the following procedure, two target are used, InServB and InServC,
which define the configuration as Synchronous Long Distance Remote Copy.
1 Create and admit the volume group and targets on InServA: