ShadowImage Guide
ShadowImage Guide
SVOS
MK-23VSP1B014-01
Overview of ShadowImage
The ShadowImage® (SI) local replication software for Hitachi Virtual Storage Platform One Block (VSP
One Block) uses local mirroring technology that you can use to create and maintain full copies of data
volumes within a storage system.
Using ShadowImage volume copies (for example, backups, secondary host applications, data mining,
testing) allows you to continue working without stopping host I/Os to the production volumes.
ShadowImage enables you to maintain storage system-internal copies of data volumes on the VSP
One Block storage systems for purposes such as data backup or duplication. The secondary volumes
(S-VOLs) are created within the same storage system as the primary volumes (P-VOLs) at hardware
speeds. ShadowImage is a key component of Hitachi Vantara's solutions and service offerings.
When you create a ShadowImage pair, the P-VOL remains available for read/write during the initial
copy. After the initial copy is complete, subsequent write operations to the P-VOL are duplicated at the
S-VOL.
The P-VOLs and S-VOLs remain paired until they are split. When you split a pair, the P-VOL continues
to be updated by host I/O, but the data in the S-VOL is not updated. The S-VOL contains a mirror
image of the P-VOL at the time the pair was split.
The S-VOL data is consistent and usable, available for read/write access by secondary host
applications.
Changes to the P-VOLs and S-VOLs are managed by differential bitmaps.
You can resynchronize the pair by updating the data from the P-VOL to the S-VOL or from the
S-VOL to the P-VOL in a reverse resynchronization.
A typical configuration consists of a storage system, a host connected to the storage system, the SI
software, a primary or source volume (P-VOL), and secondary or target volumes (S-VOLs), and
interface tools for operating SI.
Interface tools
If a logical volume assigned as a namespace on the NVM subsystem with the NVMe-oF connection is
specified as a P-VOL or an S-VOL of the SI pair, CCI must be used.
CCI is a tool that uses the command line interface to perform operations that are necessary to use
storage systems. You can either run pair commands directly from a host, or you can script CCI
commands to have pair operations performed automatically.
For more information about using CCI, see the Command Control Interface User and Reference Guide.
Consistency groups
Use a consistency group (CTG) to perform tasks on the SI pairs in the group at the same time,
including CTG pair-split tasks. Using a CTG to perform tasks ensures the consistency of the pair status
for all pairs in the group.
Volume pairs
A volume pair consists of a P-VOL and one to three layer-1 (L1) pair S-VOLs.
Because S-VOLs are updated asynchronously, the P-VOL and S-VOLs might not be identical except
immediately after a split. If a pair is split, any further updates to the P-VOL will not be reflected in the S-
VOL.
Cascaded pairs
Cascaded pairs are volume pairs created in the first and second layer. A pair made up of an L1 S-VOL
and a layer-2 (L2) S-VOL is an L2 pair. You can pair each SI L1 S-VOL with two L2 S-VOLs. You can
pair nine L1 and L2 S-VOLs with a P-VOL.
In cascaded pairs, the P-VOL for an L1 pair is a root volume and the S-VOL is a node volume. The P-
VOL for an L2 pair is the S-VOL of an L1 pair, a node volume, and the S-VOL is a leaf volume.
Creating a pair causes the storage system to start the initial copy. During the initial copy, the P-VOL
remains available for read and write operations from the host. After the initial copy, the storage system
periodically copies the differential data in the P-VOL to the S-VOL. Subsequent write operations to the
P-VOL are regularly duplicated to the S-VOL. The data in the P-VOL is copied to the S-VOL.
Initial copy is performed when you create a copy pair. Data on the P-VOL is copied to the S-VOL for
the initial copy using the following workflow.
The storage system goes through the following workflow to create an initial copy:
1. The S-VOLs are not paired. You create the copy pair.
Note: The storage system accepts read/write for unpaired volumes.
2. The initial copy is in progress (COPY(PD) status). The storage system copies the P-VOL data
to the S-VOL.
3. The initial copy is complete and the volumes are paired (PAIR status).
A P-VOL continues receiving updates from the host during the initial copy.
Update copy is performed to asynchronously copy new data (differential data) from the P-VOL of a
copy volume to the S-VOL.
The storage system goes through the following process to create an update copy:
1. The storage system marks I/O to the P-VOL in PAIR status as differential data and stores the
location of the data in bitmaps for transfer to the S-VOL.
2. After there are write I/O operations to a P-VOL, the storage system starts the update copy
operation.
Note: The timing of the update copy operation is based on the amount of differential data that
accumulates and the elapsed time since the previous update.
Updated data is copied asynchronously. Therefore, even when the initial copy completes and the pair
status changes to PAIR, data in the P-VOL and in the S-VOL might not be the same. If you want data
in the P-VOL and in the S-VOL to match at a certain date and time, you must split the pair.
Note: Data in the P-VOL and in the S-VOL might not be the same if the host did not write data to the P-
VOL during copy operation. To ensure matching data in the P-VOL and the S-VOL, you must split the
pair to change its status to PSUS.
The following table lists and describes the differences between data backup and recovery for
ShadowImage (SI) and Thin Image Advanced.
Item ShadowImage Thin Image Advanced
P-VOL physical failures P-VOL data can be recovered using P-VOL data cannot be guaranteed.
such as hard disks the S-VOL.
P-VOL logical failures P-VOL data can be recovered using P-VOL data can be recovered
such as data update the S-VOL. using the S-VOL.
errors or viruses
Capacity required for More capacity is required for Less capacity is required for
backup backup because all data in the P- backups because only differential
VOL is retained. data of the P-VOL is retained.1
Notes:
1. For snapshot pairs. When a cloned pair is created, all data in the P-VOL is retained and more
capacity is required for backup.
2. For snapshot pairs. When a cloned pair is created, the P-VOL and the S-VOL can be
separated and the P-VOL performance is not affected.
3. When a pool is full (the depletion threshold is exceeded in a pool for which the capacity for TI
pairs is limited), data in all S-VOLs using the pool cannot be guaranteed.
Recommended usage
To maintain backed up data for long periods, save it on magnetic tapes or other media. For temporary
backups use Thin Image Advanced or SI. When backing up data to magnetic tapes use SI. To reduce
the capacity necessary for backups use Thin Image Advanced, but note that this affects P-VOL
performance.
Use SI to minimize the impact from P-VOL physical failures. If you need four or more generations of
backups, use both SI and Thin Image Advanced as shown in the following figure.
System requirements
Item Requirement
License capacity The installed license capacity must be greater than or equal to the combined
size of all P-VOLs and S-VOLs.
CCI Required if you are running commands through an in-band Fibre Channel
connection.
For more information about CCI and running commands through an in-band
Fibre Channel connection, see the Command Control Interface User and
Reference Guide.
Values:
L1 pairs: 1 to 3
L2 pairs: 1 or 2
Volume Capacity: The P-VOL and S-VOL must be the same size in
blocks.
Values:
L1 pairs: 0, 1, or 2
Item Requirement
L2 pairs: 1 or 2
You can share volumes with other software applications (see Sharing
ShadowImage volumes).
Item Requirement
Consistency groups Maximum per storage system: 2,048 (including SI and Thin Image
Advanced)
Note: You cannot place SI and Thin Image Advanced pairs in the same
CTG.
CTG IDs are numbers between 0 and 7FF. CTG IDs 00 to 7F (or 0 to 127)
are used for SI and Thin Image Advanced. CTG IDs 80 to 7FF (or 128 to
2,047) are used for Thin Image Advanced only.
The capacity of P-VOLs and S-VOLs must be less than or equal to the installed license capacity.
Volume capacity is counted only once, even if you use the volume more than once. You do not need to
multiply the capacity by the number of times a volume is used (For example, a P-VOL used as the
source volume for three pairs is counted only once).
For an external volume, the total volume capacity is counted, but for a DP-VOL (a virtual volume used
in Dynamic Provisioning), the pool capacity being used by the volume is counted. For the volume with
the capacity saving enabled, the licensed capacity is the data capacity before saving. The pair with the
available licensed capacity exceeded cannot be created.
After you start performing pair tasks, monitor your capacity requirements to keep the used capacity
within the capacity of the installed license.
You can continue using ShadowImage volumes in pairs for 30 days after licensed capacity is
exceeded, regardless of whether an external volume or a DP-VOL is used. After 30 days, the only
allowed operation is pair deletion. The problem of exceeding the licensed capacity can be resolved by
reducing the capacity to less than or equal to the licensed capacity. However, the warning message will
remain displayed. The warning message is updated once a day, therefore when the problem is
resolved, the warning message will automatically disappear within 24 hours.
For more information about licenses, see the System Administrator Guide.
Planning ShadowImage
Planning ShadowImage helps you to understand the limits of what ShadowImage can do with your
storage system.
Complete the following steps to determine the maximum number of SI pairs that you can create in your
storage system:
1. Determine the number of differential and pair tables your storage system needs to create SI
pairs:
Calculate the number of tables for volumes.
Query the number of differential tables required to create SI pairs.
Note: These calculations assume that you are only using SI in the system.
2. Determine the maximum number of SI pairs that you can create on your storage system.
Differential tables and pair tables are required to create SI pairs. Differential tables are tables that
manage the differential bitmaps, and pair tables are tables that contain the information needed to
manage SI pairs. Create enough tables to handle the SI pairs you plan to create.
The storage system uses a single pair table for up to 36 differential tables.
The maximum number of SI pairs you can create is half the number of system volumes in the table if
the P-VOLs and S-VOLs are in a one-to-one relationship.
For example, if the maximum amount of system volumes you can create is 65,536, then you can only
create 32,768 SI pairs. If there are more S-VOLs than P-VOLs, then the number of allowed SI pairs
decreases.
The maximum number of differential tables and pair tables in a storage system depends on the amount
of installed shared memory. Ensure that you have sufficient shared memory to handle the number of
pairs you plan to create.
The following tables show the number of differential tables and pair tables according to the amount of
installed shared memory.
Extension 1 419,200
VSP One B26, VSP One B28:
32,768
Extension 2 419,200
VSP One B28: 32,768
You can calculate the number of differential tables and pair tables that you will need for each pair.
Note: If a DP-VOL exceeds 4 TB, differential tables are placed in the hierarchy memory instead of the
shared memory, and differential tables in the shared memory are not used. Therefore, it is not
necessary to calculate the number of differential tables for DP-VOLs over 4 TB.
For example, if the capacity of the divided volume is 3,019,898,880 KB, the calculation is as
follows:
3. Use the following formula for calculating total number of the pair tables per pair:
577 ÷ 36 = 16.0277...
Round up 16.0277 to the nearest whole number and it will become 17.
Therefore, total number of the pair tables per pair is 17 in this example.
One or more pair tables can be used for each pair. However, the pair tables assigned to a pair
cannot be used by a different pair until the pair is deleted.
After you determine the necessary number of differential and pair tables, determine whether the
planned number of pairs can be created.
The number of differential and pair tables for a pair depends on the volume capacity.
1. Calculate how many differential tables are required for the planned number of pairs:
number-of-differential-tables-per-pair × planned-number-of-pairs
2. Calculate how many pair tables are required for the planned number of pairs:
number-of-pair-tables-per-pair × planned-number-of-pairs
3. Verify that the storage system has enough differential tables and pair tables available.
The number of differential tables calculated in step 1 must not exceed the number of
available differential tables
The number of pair tables calculated in step 2 must not exceed the number of available
pair tables
Note:
When creating ShadowImage pairs with multiple volumes that have different capacities,
apply the total numbers of differential tables calculated for each volume.
When creating ShadowImage pairs with multiple volumes that have different capacities,
apply the total numbers of pair tables calculated for each volume.
The number of available differential tables and pair tables depends on the amount of
installed shared memory.
Example
if you plan to create 20 pairs in a storage system that has 57,600 differential tables, use the following
calculation:
If the capacity of the volume is 3,019,898,880 KB, you will need 577 differential tables and 17 pair
tables.
and
17 × 20 = 340 ≤ 8,192
The following dependencies affect the performance of pairs that are in the process of Quick Split
(PSUS(SP) status) or Steady Split (COPY(SP) status):
OFF Internal volume Differential data is copied to the cache in the storage
system. The pair is split (PSUS status).
External volume
ON Internal volume
* This column indicates whether the top pool volume allocated to the S-VOL is internal or external.
Performing pair tasks, such as creating, splitting, and resynchronizing SI pairs, can affect host server I/
O performance on the storage system.
that the primary volume and secondary volume belong to different parity groups for
DDP.
If the primary volume and secondary volume are within the same parity group for DDP
and you need to perform copy operations on multiple pairs, perform operations for one
pair at a time.
If the system becomes overloaded, add drives to the parity group for DDP or increase
the number of Channel Boards (CHB) or Disk Boards (DKB).
If you plan to simultaneously use multiple software applications, make sure that your storage
system is configured for optimal performance, such as sufficient cache. Concurrent use affects
the performance and operation of the other software applications.
If use AIX® host servers, for best performance, place the P-VOLs on one AIX® host server and
the paired S-VOLs on another.
Using a single AIX® host server for the P-VOLs and S-VOLs changes the P-VOL and S-VOLs
to the same Port VLAN ID when you create or resynchronize the SI pairs. If the Port VLAN IDs
of the P-VOLs and S-VOLs are the same and you reboot the host server, the volumes can be
misidentified. For example, the S-VOL can be identified as the P-VOL.
Quick Restore. You can maximize performance when you restore pairs using Quick Restore
(see Workflow for maximizing Quick Restore performance).
For more information about the guidelines for maximizing host server I/O performance while performing
pair operations, see Optimizing host server I/O performance.
Creating, splitting, and resynchronizing SI pairs can affect host server I/O performance. You can take
the following actions to minimize the impact of pair operations on host server I/O performance:
Complete the following steps to maximize performance when restoring pairs with Quick Restore:
1. Use the same RAID level and drive type for the P-VOL and S-VOL.
2. Restore the pairs using Quick Restore.
3. Resume the original RAID levels if they were changed by a reverse resynchronization action.
You can share volumes on which you set Data Retention Utility access attributes.
You can create SI pairs using volumes on which you set Data Retention Utility access attributes.
Note: Performing SI tasks does not change Data Retention Utility access attributes.
The volume access attributes that you have specified for the SI pair determines the SI pair tasks that
you can perform. The pair tasks you can perform are vary, depending on whether you set Data
Retention Utility access attributes.
The following table lists the volume access attributes specified for the SI pair and the pair tasks that
you can perform when using CCI to set access attributes.
A pair's status determines the access attributes you can set for existing P-VOLs or S-VOLs. Depending
on the SI pair status, you might not be able to set access attributes for SI P-VOLs and S-VOLs for Data
Retention Utility. Access attribute settings also depend on whether you set Data Retention Utility
access attributes when using CCI.
The following table lists which attributes you can set depending on the pair status when using CCI to
set access attributes.
S-VOL YES NO
S-VOL YES NO
S-VOL YES NO
S-VOL YES NO
If you are using HDP volumes as SI P-VOLs or S-VOLs, the capacity of the HDP pool allocated to the
volume is added to the SI licensed capacity. A volume with capacity saving enabled can be used as a
P-VOL or S-VOL.
Restrictions
Data compressed or deduplicated by the capacity saving function is copied to a volume after
compression and deduplication are released. The capacity saving function is not performed
immediately for copied data. Before creating or resynchronizing a ShadowImage pair, make
sure that the available capacity in the copy destination volume is greater than the used capacity
in the copy origination volume before capacity saving. For details, see the Provisioning Guide
for your storage system.
If you create SI pairs using a volume for which the capacity saving function is used,
compressed or deduplicated data is copied. Because of this, copy or I/O performance may be
degraded.
When the capacity saving function is used, management information is stored in a pool. As a
result, there may be difference between a P-VOL and an S-VOL in the number of used pages
or licensed capacity.
Because the S-VOL uses the same pool capacity as the P-VOL, best practice is to avoid the
following volume combinations:
Using only the S-VOL as a Dynamic Provisioning volume (DP-VOL).
Using the P-VOL as a DP-VOL with the Data Direct Mapping attribute, and using the S-
VOL as a normal DP-VOL.
You cannot perform a Quick Restore if only the P-VOL or only the S-VOL is a DP-VOL or the
capacity saving of the P-VOL or S-VOL is enabled. For Quick Restore, you must use DP-VOLs
for both the P-VOL and S-VOL.
The deduplication system data volume cannot be used as a P-VOL or S-VOL.
You can use a maximum size Dynamic Provisioning volume as an SI P-VOL or S-VOL. For
information about the maximum size for Dynamic Provisioning volumes, see the Provisioning
Guide for your storage system.
You must use the block specification for SI volumes shared with DP-VOLs. You cannot use TB,
GB, or MB. For example, 4 TB = 8,589,934,592 blocks.
When you create an SI pair using a DP-VOL greater than 4,194,304 MB (8,589,934,592
blocks), the differential data is managed in a pool associated with the SI pair volume. In this
case, the required pool capacity for managing the differential data varies depending on the
product configuration, with a maximum of four pages for every 4,123,168,604,160 bytes.
For details about whether zero pages can be reclaimed when you specify SI P-VOLs or SI S-
VOLs on Command Control Interface, see the Provisioning Guide for your storage system.
If you perform an SI paircreate or pairresync operation while zero pages are being reclaimed
(including operations by WriteSame or Unmap, or by rebalancing the usage level), it results in
zero-page reclamation interruption.
If reclaiming zero pages is in progress on a P-VOL or an S-VOL during SI pair creation, the pair
might not be created. Therefore, avoid the time period when reclaiming zero pages is in
progress when creating a pair.
If you create an SI pair during WriteSame or Unmap command processing, the pair creation
may fail. In this case, create the pair again after the WriteSame or Unmap command
processing finishes.
If the WriteSame or Unmap command was issued to the SI P-VOL or S-VOL, zero pages are
not reclaimed by the command.
Zero pages are not reclaimed for the SI P-VOL or S-VOL even if you rebalance the usage level.
If the copy origination page of an SI pair is not allocated, but the copy destination page is
already allocated, zero data is copied to the copy destination. In this case, zero data pages are
not reclaimed. If you want to reclaim zero data pages, split the SI pair to change its status to
PSUS, and then reclaim zero data pages of the applicable volume.
The following table lists the volumes that can be shared between global-active device (GAD) and
ShadowImage.
(CCI command)
P-VOL S-VOL
YES
GAD P-VOL YES
YES
GAD S-VOL YES
(CCI command)
P-VOL S-VOL
The following tables list the relationship between GAD pair statuses and ShadowImage tasks when
sharing GAD pair volumes and ShadowImage pair volumes.
Note:
If a pair that cannot be split is in a consistency group (CTG), pairs in the CTG are suspended
(PSUE status) when executing a CTG pair split function for ShadowImage.
To split a ShadowImage pair that links with a GAD pair and obtain a backup, stop I/O for the
target backup volume, and then split the ShadowImage pair.
(CCI command)
COPY PAIR PSUS PSUE
(RL) (RL)
(paircreate)
Create and split a pair YES YES YES YES YES YES
(paircreate -split)
(pairsplit)
(pairresync)
(pairresync -restore)
(CCI command)
COPY PAIR PSUS PSUE
(RL) (RL)
(pairresync -restore)
(pairsplit -E)
(pairsplit -S)
*This task cannot be performed if the migration source volume of Volume Migration is contained in
the same volume.
(CCI command)
COPY PAIR SSUS PSUE SSWS
(RL)
(paircreate)
(paircreate -split)
(pairsplit)
(pairresync)
(CCI command)
COPY PAIR SSUS PSUE SSWS
(RL)
(pairresync -restore)
(pairresync -restore)
(pairsplit -E)
(pairsplit -S)
*This task cannot be performed if the migration source volume of Volume Migration is contained in
the same volume.
NO NO YES NO YES NO
Create a pair
(paircreate)
NO NO YES NO YES NO
Split a pair
(pairsplit)
(pairsplit -E)
(pairsplit -S)
NO NO NO NO YES
Create a pair
(paircreate)
NO NO NO NO YES
Split a pair
(pairsplit)
(pairsplit -E)
(pairsplit -S)
LUN Manager tasks do not affect SI tasks. You can assign volumes that are under secure ports or that
are assigned to World Wide Name (WWN) groups and/or LUN groups to SI pairs. You can also use
volumes that are assigned to SI pairs in LUN Manager tasks, such as assignment to WWN groups and/
or LUN groups.
A host cannot access SI S-VOLs except when you split the pair.
You can share SI P-VOLs and S-VOLs with Resource Partition Manager by specifying them in a
Resource Partition Manager resource group.
For more information about Resource Partition Manager, see the System Administrator Guide for your
storage system.
The resource group must be assigned to a user group for which you have privileges.
You should understand the relationship between Thin Image Advanced pair statuses and
ShadowImage operations when the P-VOL of a Thin Image Advanced pair is shared as the P-VOL or
an S-VOL of a ShadowImage pair.
Note:
Pair tasks with Thin Image Advanced primary volume shared with
ShadowImage primary volume
P-VOL S-
VOL
S-VOL NO NO
Notes:
1. You must create the SI pair before you create the HTI Advanced pair.
Create a pair YES YES YES YES YES YES NO YES YES
(paircreate)
Create and YES YES YES YES YES YES NO YES YES
split a pair
(paircreate -
split)
Split a pair YES YES YES YES YES YES NO YES YES
(pairsplit)
Quick Restore NO NO NO NO NO NO NO NO NO
a pair
(pairrespync -
restore)
Suspend a YES YES YES YES YES YES YES YES YES
pair (pairsplit -
E)
Delete a pair YES YES YES YES YES YES YES YES YES
(pairsplit -S)
*After issuing the raidcom modify snapshot -snapshot_data split command for a Thin Image Advanced pair for
which a consistency group is configured, verify that the status of all pairs in the Thin Image Advanced
consistency group is PSUS or unpaired, and then execute the SI task. If you do not, the Thin Image Advanced
snapshot data and P-VOL data when the storage system accepted the raidcom modify snapshot -snapshot_data
split command will not be consistent. For details about Thin Image Advanced pair status and operations, see the
Thin Image Advanced User Guide.
Pair tasks with Thin Image Advanced primary volume shared with
ShadowImage secondary volume
Create a pair1 NO NO NO NO NO NO NO NO NO
(paircreate)
Create and NO NO NO NO NO NO NO NO NO
split a pair
(paircreate -
split)
Quick Restore NO NO NO NO NO NO NO NO NO
a pair3
(pairrespync -
restore)
Delete a pair YES YES YES YES YES YES YES YES YES
(pairsplit -S)
Notes:
1. When you use an S-VOL of an SI pair as the P-VOL of a Thin Image Advanced pair, you must create the
SI pair first and then create the Thin Image Advanced pair.
2. After issuing the raidcom modify snapshot -snapshot_data split command for a Thin Image Advanced pair
for which a consistency group is configured, verify that the status of all pairs in the Thin Image Advanced
consistency group is PSUS or unpaired, and then execute the SI task. If you do not, the Thin Image
Advanced snapshot data and P-VOL data when the storage system accepted the raidcom modify
snapshot -snapshot_data split command will not be consistent. For details about Thin Image Advanced
pair status and operations, see the Thin Image Advanced User Guide
3. When the host accesses the Thin Image Advanced S-VOL, the Thin Image Advanced P-VOL data might
be used to return a reply. For this reason, you cannot Quick Restore to exchange the SI P-VOL and S-
VOL (which is the Thin Image Advanced P-VOL).
You can share SI P-VOLs and S-VOLs with TC P-VOLs and S-VOLs. However, you cannot share SI S-
VOLs with TC S-VOLs.
The write operation on the TC P-VOL takes more time when you share an SI P-VOL with a TC S-VOL.
For more information about splitting pairs, see Other pair splitting methods.
Restrictions
When you share an SI P-VOL with a TC S-VOL, if you split the SI pair during write I/Os to the
TC P-VOL, only part of write I/Os might be written to the SI S-VOL. To keep data consistency in
the SI S-VOL, stop I/Os to the TC P-VOL first, and then split the SI pair.
You can use SI cascaded pairs as TC pairs. TC does not distinguish between node and leaf
volumes in cascaded pairs. Both are considered S-VOLs.
If you plan to Quick Restore the SI pair, you must first suspend the TC pair.
You can perform a CTG pair-split on SI pairs that share volumes with TC S-VOLs.
For more information about CTG pair-split for shared volumes, see Using consistency group
pair-split with shared volumes
If you want to use the SI S-VOL as the TI P-VOL, create the SI pair first, and then create the TI
pair.
For more information about sharing SI and TC volumes, see the VSP One Block TrueCopy® User
Guide.
The following table describes ShadowImage operations when SI P-VOLs are shared with TC P-VOLs
or S-VOLs.
ShadowImage operation
TC
Create Split Suspend copy Release Resync pairs
pair
pairs pairs operation pairs
Normal Quick Reverse Quick
status
copy resync copy restore
The following table describes ShadowImage operations when SI S-VOLs are shared with TC P-VOLs.
ShadowImage operation
TC
Create Split Suspend copy Release Resync pairs
pair pairs pairs operation pairs
Normal Reverse Quick
copy Quick copy restore
status
resync
You can use SI cascaded pairs as UR pairs. In cascaded pairs, UR does not distinguish
between node and leaf volumes. Both are considered as S-VOLs.
If you plan to Quick Restore the SI pair, you must first suspend the UR pair.
You can perform a CTG pair-split on SI pairs that share volumes with UR S-VOLs.
For more information about using CTG pair-split on pairs with shared volumes, see Using
consistency group pair-split with shared volumes.
For more information about sharing SI and UR volumes, see the related appendix in the VSP One
Block Universal Replicator User Guide.
The following table describes ShadowImage operations when SI P-VOLs are shared with UR P-VOLs
or S-VOLs.
ShadowImage operation
UR
Create Split Suspend copy Release Resync pairs
pair
pairs pairs operation pairs
Normal Quick Reverse Quick
status
copy resync copy restore
The following table describes ShadowImage operations when SI S-VOLs are shared with UR P-VOLs.
ShadowImage operation
UR
Create Split Suspend copy Release Resync pairs
pair pairs pairs operation pairs
Normal Reverse Quick
copy Quick copy restore
status
resync
You can share SI P-VOLs and S-VOLs with UR P-VOLs and S-VOLs that are in 3DC configurations
with three UR sites.
The following figures show examples of sharing SI and UR volumes in the following configurations:
You can perform ShadowImage operations on Universal Replicator volumes that are used in a 3DC
multi-target configuration.
The following tables specify whether SI operations can be performed when UR volumes are shared
with SI volumes in a 3DC multi-target configuration.
You can perform ShadowImage operations on Universal Replicator volumes that are used in a 3DC
multi-target delta resync configuration.
The following table specifies whether SI operations can be performed when a UR S-VOL and a UR S-
VOL for delta resync are shared with SI P-VOLs in a 3DC multi-target delta resync configuration.
Table. 3DC multi-target delta resync: Sharing a UR S-VOL and a UR S-VOL for delta resync with SI P-
VOLs
UR pair UR ShadowImage operation
status delta
(S-VOL) resync Create Split Suspend Release Resync pair
pair pair pair pair pair
Normal Quick Reverse Quick
mirror
copy resync restore
status
You can perform ShadowImage operations when ShadowImage and Universal Replicator are used in a
3DC cascade configuration.
The following tables specify whether SI operations can be performed when UR volumes are shared
with SI volumes in a 3DC cascade configuration.
Table. 3DC cascade: Sharing a UR P-VOL with an SI P-VOL on the primary storage system
UR pair ShadowImage operation
status
Create Split pair Suspend Release Resync pair
(P-VOL)
pair pair pair
Normal Quick Reverse Quick
copy resync restore
You can perform ShadowImage operations on Universal Replicator volumes that are used in a 3DC
cascade configuration with delta resync.
The following tables specify whether SI operations can be performed when UR volumes are shared
with SI volumes in a 3DC cascade delta resync configuration.
Table. 3DC cascade delta resync: Sharing a UR P-VOL on the primary system and a UR S-VOL for
delta resync with SI P-VOLs
UR pair UR ShadowImage operation
status delta
(P-VOL) resync Create Split Suspend Release Resync pair
pair pair pair pair pair
Normal Quick Reverse Quick
mirror
copy resync restore
status
The following figures provide configuration examples when ShadowImage, TrueCopy, and Universal
Replicator are used.
For the above configurations, you cannot perform Quick Restore for SI pairs.
The following image illustrates a 3DC multi-target configuration with the following pairs:
For details about 3DC multi-target configurations, see the VSP One Block Universal Replicator User
Guide.
The following tables describe ShadowImage operations according to the status of TC pairs and UR
delta resync pairs in the storage system of the TC secondary site.
PJNS
PAIR YES YES YES YES YES YES NO NO
PJES
PAIR YES YES YES YES YES YES NO NO
SJNS
PAIR YES YES YES YES YES YES NO NO
The following table describes ShadowImage operations according to the status of TC and UR pairs in
the storage system of the TC or UR local site.
PSUE
PSUE
PSUE
You can create SI pairs using Universal Volume Manager external volumes.
For more information about Universal Volume Manager external volumes, see the VSP One Block
Universal Volume Manager User Guide.
Access to the external volumes used for SI pairs, only from or through the local storage system. Do not
access, for example, from a host connected with the external storage system, or by using a copy
functionality on the external storage system.
You can migrate the following types of SI pair volumes as Volume Migration source volumes:
To assign another type of SI pair volume as a Volume Migration source volume, you must delete the SI
pair first.
Volume Migration volumes cannot be used in SI pair tasks; you must release a volume in Volume
Migration before you can use it as an SI volume.
You must set up and prepare the P-VOLs and S-VOLs for SI pairing before creating SI pairs. The
volumes you use must meet the requirements for pair volumes.
The following table provides example volumes to guide you in preparing volumes for SI pairing.
For information about system option modes, see the System Administrator Guide. Contact customer
support if you want to ask for settings.
The local replica options provide additional settings to enable you to tailor your local replication
environment to your specific operating requirements. The local replica options include the Copy
Threshold option that your service representative can set and the options that you can set.
When you want to enable and disable the Copy Threshold option, contact your service
representative.
Note the following when you use the Quick/Steady Split Multiplexing, Reverse Copy
Multiplexing, and Normal Resync Multiplexing options:
The increase in copy volume increases the amount of data to be written to the target
volume.
If the parity group performance is less than the pair split or resync performance, the
amount of data waiting to be written to cache memory (cache write pending ratio) may
increase until it exceeds 60%, and the copy processing might be placed in the wait
state. In such a case, consider the copy order to reduce the number of pairs for which
copy processing is executed concurrently for volumes provisioned by the same parity
group.
If the number of volumes to be resynchronized or split increases, the maximum number
of jobs for pair split or resynchronization that can execute concurrently for a pair
decreases. In this case, when you split or resynchronize many pairs concurrently with
this option set, the pair split or resynchronization performance might not be changed.
The maximum number of copy operations that can be run concurrently is 126 (including
the number for ShadowImage). This number includes initial copying, resynchronization,
update copying, and differential data copying. When copy operations other than pair
split and resynchronization are running concurrently, if you set this option, the pair split
or resynchronization performance might not be changed.
The maximum number of pairs that can be copied concurrently in the background is
504. If the number of pairs being copied reaches 504, the next pair to be copied waits
until one of the copy processing finishes.
The Quick/Steady Split Multiplexing option is enabled for Steady Split and Quick Split.
Reverse Copy Multiplexing is enabled for Reverse Copy.
Normal Resync Multiplexing is enabled for Normal Resync.
If the Increase Background Copy Performance option is enabled, the amount of copy
data increases over the entire storage system while the pair is in the PAIR status.
Therefore, you can decrease the amount of differential data when the pair-split
operation starts. However, the MP operating ratio and cache write pending rate increase
because the amount of copy data increases in the storage system. Understand the
trade-offs when you use this option.
Differences between the Host I/O Performance option and the Copy Pace Ext. options:
The Host I/O Performance option reduces the impact on host I/O by lowering the copy
activity when pairs are in COPY(PD), PAIR, COPY(SP), PSUS(SP), COPY(RS),
COPY(RS-R) status.
The Copy Pace Ext. options reduce the impact on host I/O by lowering the copy activity
when pairs are in the PAIR status.
These options can be set concurrently. If you set both options, you can benefit from both.
Default
No. Option Description
OFF
1 Swap & Freeze Saves ShadowImage data as-is
immediately after Quick Restore. Used
with the Quick Restore, inhibits the
update copy operation after Quick
Restore, and the paired S-VOL in PAIR
status is not updated and remains
unchanged.
OFF
2 Host I/O Performance Gives weight to host I/O responses
rather than the copy time of a volume.
This option controls SI copy operations
and improves host I/O responses. This
option suppresses copy operations at
any time regardless of workload.
Default
No. Option Description
OFF
9 to 13 HOST I/O priority mode 1 to 5 Makes the copy processing for SI pairs
fail for the host I/O that causes copy
processing for SI pairs so that the host I/
O is prioritized. This function is
performed if the cache write pending rate
does not decrease and the host I/O does
not finish within the time period specified
by the option. After the copy processing
fails, all data is copied as the
resynchronization. See Time until the
copy processing fails.
OFF
20 Copy Pace Ext. Slower1 Reducing the copy volume in the PAIR
status curbs the influence to the I/O
21 Copy Pace Ext. Slower2
performance of the host server. This
option is available to all SI pairs in the
22 Copy Pace Ext. None
PAIR status. The I/O performance of the
host server is improved most effectively
with Copy Pace Ext. None, followed by
Copy Pace Ext. Slower2, and Copy Pace
Ext. Slower1. Copy Pace Ext. None
takes precedence over Copy Pace Ext.
Slower2, and Copy Pace Ext. Slower2
takes precedence over Copy Pace Ext.
Slower1.
Default
No. Option Description
ON
24 Quick/Steady Split Multiplexing Accelerates SI pair split. The number of
jobs used for concurrent copy processing
(ShadowImage) for each pair is changed from 1 to 24.
ON
25 Reverse Copy Multiplexing Accelerates resynchronization
(secondary to primary) of SI pairs.
(ShadowImage)
The number of jobs used for concurrent
copy processing for each pair is changed
from 1 to 24.
ON
26 Normal Resync Multiplexing Accelerates resynchronization (primary
to secondary) of SI pairs.
(ShadowImage)
The number of jobs used for concurrent
copy processing for each pair is changed
from 1 to 24.
The time until the copy processing fails differs depending on the configuration.
When ShadowImage works alone, the available setting rage of the time is 1 to 31 seconds.
When the P-VOL of ShadowImage works with other software:
Configuration Available setting range of
the time until copy
processing fails (seconds)
Note:
1. Only TrueCopy, Universal Replicator, and global-active device apply. For other
software, the configuration is regarded as the same as when ShadowImage works
alone.
2. The function of these options is disabled when host I/O is performed on the P-VOL of
Universal Replicator.
When you create a pair, the storage system performs an initial copy to copy data in the P-VOL to the S-
VOL. You can create the SI pair and immediately split the pair so that you can access the S-VOL. You
can also create a cascaded pairs.
Creating an SI pair causes the MP unit that is responsible for processing the P-VOL LDEV's I/O to
assume processing responsibility for the S-VOL LDEV's I/O operations.
The P-VOL and S-VOL must be the same size in blocks. If the capacity is displayed in GB or
TB, a small difference between P-VOL and S-VOL capacity might not be displayed.
You can copy an encrypted volume to an unencrypted volume. There is no guard logic to
enforce copying encrypted P-VOLs to only encrypted S-VOLs. Unless there is a specific reason
for the data to become unencrypted, make sure you maintain the encryption by using only
encrypted S-VOLs.
The larger the volume capacity, the longer the time from creation of a pair until it transitions to
PAIR status. For virtual volumes, the more pages allocated to a virtual volume, the longer the
time from pair creation to PAIR status. Even when no pages are allocated to a virtual volume, if
the volume capacity is 256 TB, it takes at least an hour until the status changes to PAIR after
pair creation.
The value of the T10 PI attribute must be the same for the P-VOL and S-VOL.
When you simultaneously create an L1 pair and an L2 pair, if you specify the -split option in
CCI, the L2 pair split begins before the L1 pair status changes to PSUS. Because of this, the
operation might fail.
Initial copy is performed with a maximum of 24 multiplicities (number of jobs for which
concurrent copy can be performed) for each pair.
If the -split option is specified in the paircreate command, the multiplicity is set to 1. Because of
this, the processing time becomes longer than when the -split option is not specified.
If Quick/Steady Split Multiplexing (ShadowImage) option is enabled in the local replica option,
the maximum number of multiplexing becomes 24 even when the -split option is specified.
If only the P-VOL capacity of an L1 pair has been expanded, an L2 pair cannot be created.
Make sure to create the L2 pair after expanding all of the volume capacity used in the pairs.
1. If you are creating L1 and L2 pairs, check the status (see L1, L2 pair status and supported pair
tasks).
2. Specify the volume that you want to duplicate. This becomes the primary volume (P-VOL).
3. Identify the volume that will contain the copy. This becomes the secondary volume (S-VOL).
4. Create the pair by associating the P-VOL and the S-VOLs.
Cache maintenance can reduce overall performance and should be scheduled during times of
low system activity.
Maintenance of physical disk drive that provisions LDEVs used by SI can be performed without
impacting SI.
If a physical device failure occurs, the pair status is not affected because of the RAID
architecture.
If a physical device failure requires the storage system to use dynamic sparing or automatic
correction copy, the pair status is not be affected.
If an LDEV failure occurs, the storage system suspends the pair.
If an SI pair is using an LDEV, certain activities are not allowed. You can only block (for
maintenance), format, or restore an LDEV that is in use by only a pair in PSUE status.
Splitting an SI pair suspends the pairing of the P-VOL and S-VOLs until a resync or delete operation is
performed. Host updates to the P-VOL continue and are tracked as delta tracks in the bitmap. The S-
VOL data is available and can be accessed.
Splitting an SI pair ensures data consistency and that the data in the S-VOL at the time of the split is
usable. The S-VOL contains a mirror image of the original volume at that point in time, and it is
available for read/write access by secondary host applications.
The P-VOL for a split pair continues to be updated, but the S-VOL remains unchanged. The differential
data that accrues while the pair is split is stored in the differential bitmaps. Changes to the P-VOL and
S-VOLs are managed in these differential bitmaps. The differential data accrues until you
resynchronize the pair, which copies the differential data to the S-VOL.
If you have assigned an SI P-VOL or S-VOL to a volume reserved for Volume Migration, splitting the
volume cancels migration.
If you are splitting SI pairs with shared TC or UR volumes, see the restrictions (see Requirements_
restrictions_ and guidelines for using consistency group pair-split with shared volumes).
If you are sharing SI S-VOLs with UR P-VOLs and the R-JNL has a timeout period that ends after the
split time, the storage system might not detect the journal data. In this case, the SI split operation runs
after the timeout period.
Default: 6 hours
To ensure data consistency in the P-VOL and the S-VOL after a split, stop write I/Os from the
host to the P-VOL in advance. By this method, you can keep the P-VOL from being updated
during a split, and ensure data consistency between the P-VOL and the S-VOL.
For more information about checking I/O performance, see the System Administrator Guide.
If you split a pair during initial copy, the initial copy operation is canceled, and it is performed
again after the split. For initial copy, the maximum copying multiplicity (the number of jobs used
for concurrent copy processing) for each pair is 24, but the number might change to 1 after the
split. As a result, if you split a pair during initial copy, copy processing might take longer than
splitting a pair after initial copy completes.
The copy processing to split a pair requires some time to check if the differential data exists
even if there is no differential data. The amount of time to check the differential data takes
longer as the volume capacity increases, and it might take 10 minutes or more for 256 TB.
By default, for pair split, the maximum multiple copy processing for each pair is set to 24. However, if
you set the -splitoption for the paircreate command, the multiplicity can be changed to 1. As a result,
the processing time will be longer compared to when the -split option is not specified. Additionally, if the
Quick/Steady Split Multiplexing (ShadowImage) option of the local replica option is enabled, the
multiplexing will be maximum of 24 even when the -split option is specified.
If you share SI P-VOLs in a storage system with TC or UR S-VOLs, you can use consistency group
(CTG) pair-split to keep SI S-VOLs consistent.
When an SI P-VOL shares a UR or TC S-VOL to create pairs, you can assign the same CTG ID to the
SI pairs in order to use the CTG pair-split function to maintain consistency among the SI S-VOLs.
When you perform a CTG pair-split on shared UR P-VOLs, the following operations occur:
1. UR restores the journal data that was created before you restored the split time to UR S-VOLs
(SI P-VOLs).
Note: If an SI pair is suspended due to a failure, the split time and the actual task start time
must be the same in order to restore UR journal data that you created before the split time to
the UR/SI volume after the split. The task start time is determined by the amount of journal data
in the journal volume at the time of the split.
For example, if the journal volume contains data that needs one hour to be restored, the
starting time of the split operation delays for an hour.
1. Define a CTG for the ShadowImage pairs to be split using the Consistency Group pair split
function.
2. Create the pairs by specifying the Consistency Group pair split option.
3. Confirm that all the ShadowImage pairs within the Consistency Group are in the PAIR or
COPY(PD) state.
4. Split the pairs within the Consistency Group.
These pairs will be split simultaneously.
Requirements, restrictions, and guidelines for using consistency group pair-split with shared
volumes
You can share SI P-VOLs with TC or UR S-VOLs but there are requirements, restrictions, and
guidelines for using consistency group (CTG) pair-split in these cases.
Requirements
If you are sharing SI P-VOLs with TC or UR S-VOLs, CTG pair-split has the following requirements:
Restrictions
If you are sharing SI P-VOLs with TC or UR S-VOLs, CTG pair-split has the following restrictions:
Guidelines
If you are sharing SI P-VOLs with TC or UR S-VOLs, use the following guidelines when performing a
CTG pair-split:
To perform a CTG pair-split on SI pairs, pair operations must be done using CCI.
The pair status for all of the SI pairs in the CTG determines if you can perform a consistency group
(CTG) pair-split. If all of the SI pairs in the CTG are paired (PAIR status), you can perform a CTG pair-
split.
The following table describes when you can perform a CTG pair-split, based on the SI pairs in the CTG
that are not paired (a status other than PAIR), and the resulting pair status after you perform the pair-
split.
The status of the pairs in Can you perform Status after you perform a
the CTG that have a CTG pair-split
a CTG pair-split?
status other than PAIR
The status of the pairs in Can you perform Status after you perform a
the CTG that have a CTG pair-split
status other than PAIR a CTG pair-split?
Notes:
1. If you share a UR S-VOL and an SI P-VOL, the command might end normally after you
perform a CTG pair-split. Ensure the status of the pairs within the CTG have changed to
PSUS (use the pairdisplay command).
2. Consistency is guaranteed only for SI pairs in PAIR or COPY(PD) status.
The following are examples of when you can perform a CTG pair-split based on status of the SI pairs in
the CTG:
Example 1
The are six SI pairs in a CTG. Two of the pairs are paired (PAIR status), two are in COPY
status, and two are in PSUS status. In this case, you can perform a CTG pair-split, and doing
so changes the status of all of the pairs in the CTG to PSUS. However, the S-VOLs that were in
PSUS might not be consistent with the other volumes in the CTG.
Example 2
There are two SI pairs in a CTG and one is paired (PAIR status) and the other is in the process
of being resynchronized (COPY status). In this case, you cannot perform a CTG pair-split.
Example 3
The are six SI pairs in a CTG. Two of the pairs are paired (PAIR status), two are in the process
of Quick Split (PSUS status), and two are in PSUE status. In this case, the CTG pair-split ends
abnormally and the status of all of the pairs in the CTG remains the same.
You can resynchronize split (PSUS status) or suspend (PSUE status) pairs. Resynchronization
changes the status of the split volume pairs to PAIR. Resynchronizing a split pair copies the P-VOL's
differential data to the S-VOL and again pairs the S-VOL with the P-VOL. Resynchronizing a
suspended pair copies the entire P-VOL to the S-VOL and takes the same amount of time as the initial
copy operation.
Note: Resynchronizing a pair does not ensure data consistency. Data in the two volumes is consistent
only if the following conditions exist:
Note: If you perform a Quick Restore for a pair consisting of an encrypted volume and an unencrypted
volume, the encryption statuses of the volumes are reversed.
You can forward or reverse resynchronize pairs. A forward resynchronization resynchronizes from the
P-VOL to the S-VOL. A reverse resynchronization restores pairs by resynchronizing from the S-VOL to
the P-VOL.
Forward resynchronization
You can use one of the following methods to forward resynchronize pairs:
Normal Copy (Primary > Secondary): A full forward resynchronization from the P-VOL to the S-
VOL. During a Normal Copy, only the P-VOL is accessible to hosts for read/write operations.
Quick Resync (Primary > Secondary): A forward resynchronization from the P-VOL to the S-
VOL where data is not copied or resynchronized. The volumes are paired (PAIR status). The
update copy operation copies the differential data to the S-VOL.
During a Quick Resync, the P-VOL is accessible to hosts for read/write operations. Quick
Resync does not ensure data consistency, even if there is no host I/O during the
resynchronization.
Reverse resynchronization
Reverse Copy (Secondary > Primary): A full restoration from the S-VOL to the P-VOL. The
differential data is updated to the P-VOL.
During a Reverse Copy you can delete or suspend the pairs, but you cannot create, split, or
resynchronize pairs that share the same P-VOL. The P-VOL is inaccessible to hosts.
If you are sharing a TC or UR volume with an SI volume, you cannot create a TC or UR pair
with the shared volume.
Quick Restore from the Secondary to the Primary: A partial restoration that does not copy the
data but does the following:
Swaps the P-VOL and S-VOLs including their RAID levels and drive types.
Pairs the volumes (PAIR status*).
*The pair status changes to RCPY first, and then it changes to PAIR.
During a Quick Restore, the P-VOL and S-VOL are inaccessible. After a Quick Restore, the P-
VOL is accessible.
If you have a small amount of differential data, use Reverse Copy instead of Quick Restore
because Reverse Copy completes faster.
If you use volumes for which you set Data Retention Utility access attributes, Quick Restore
does not exchange the P-VOL and S-VOL access attributes.
For more information about using volumes for which you set Data Retention Utility access
attributes, see Sharing volumes and Data Retention Utility access attributes.
Note: To minimize the time it takes to Quick Restore an SI pair, do not perform LDEV
maintenance while the Quick Restore is processing.
You can delete or suspend the pair while you are restoring the pair using Quick Restore but you
cannot do the following:
An SI L2 pair
A pair volume for which you are formatting either internal volume using Quick Format.
For more information about formatting volumes using Quick Format, see the
Provisioning Guide for your storage system.
By default, for pair resynchronization (Normal Resync or Reverse Copy), the maximum
multiplicity of copy processing for each pair is set to 24. However, if you set the Normal Resync
Multiplexing option or Reverse Copy Multiplexing option of local replica options to OFF, the
multiplicity can be changed to 1. Make sure to set the Normal Resync Multiplexing option or
Reverse Copy Multiplexing option to OFF if the host I/O performance is prioritized.
For more information about checking I/O performance, see the System Administrator Guide.
If you plan to restore pairs using Reverse Copy, split or suspend the pairs sharing the same P-
VOL.
If the SI pair you plan to restore shares a volume with TC or UR, suspend the TC or UR pair.
For more information about splitting or suspending pairs, see Splitting ShadowImage pairs or
Suspending ShadowImage pairs, respectively.
Resynchronizing split pairs typically takes less time than resynchronizing suspended pairs (PSUE
status). Split pairs typically contain much less accumulated differential data than the total amount of
data in the P-VOL.
If you are using CCI to run commands, run the following command to resynchronize split pairs (PSUS
status): pairresync
When the following conditions apply, a ShadowImage pair is suspended, and the status of the pair is
changed to PSUE:
the storage system detects an error condition related to an update copy operation
the storage system cannot keep the pair mirrored
At this time, write I/Os to the P-VOL continue and all tracks of the P-VOL are saved as differential data.
When the pair is resynchronized, the pair status changes to Resync/PENDING, and the entire P-VOL
is copied to the S-VOL. For a split pair, resynchronization is completed in a short time. But it takes the
same period of time for a suspended pair to be resynchronized as the initial copy operation.
Change the status of the pair to PSUE by using the pairsplit -E command.
Delete the SI pairs that you no longer need. Deleting a pair unpairs the P-VOL and S-VOL but does not
delete their data. You can use the volumes of deleted pairs in another pair.
The pair must be unpaired and not in the process of being deleted and the volumes are not in the
process of being unpaired (SMPL(PD) status).
Review the following list to understand what happens after an SI pair is deleted.
If an SI pair is deleted, the volume status changes from SMPL(PD) to SMPL. You can use the CCI
pairdisplay command to check the volume status, but this command cannot differentiate SMPL from
SMPL(PD). To differentiate them in CCI, use the inqraid command to check if the volume is used by SI.
If it is used by SI, the status is SMPL(PD). If it is not used by SI, the status is SMPL.
1. Ensure that all of the write I/O operations to the P-VOL have completed and that all secondary
host applications that access the P-VOL have stopped.
2. Set the P-VOL offline.
3. Verify that the SI pair is unpaired and not in the process of being unpaired.
4. Split the SI pair.
5. Delete the SI pair.
The MP unit assigned to an S-VOL cannot be used if a pair is created because the MP unit assigned to
the P-VOL is also assigned to the S-VOL. Deleting the pairs returns the allocation of processor
responsibility to the state it was before the pairs were created. However, the allocation cannot be
returned in either of the following cases:
A user changed the allocation of the MP unit assigned to an S-VOL cannot be used if a pair is
created because the MP unit for the P-VOL or S-VOL of the SI pair.
When the pair is deleted, the cache write pending rate of the MP unit assigned to an S-VOL
cannot be used if a pair is created because the MP unit to which the S-VOL belongs is 50% or
more, or the cache write pending rate of the original MP unit assigned to an S-VOL cannot be
used if a pair is created because the MP unit is 50% or more.
If the original MP unit assigned to an S-VOL cannot be used if a pair is created because the MP unit is
already removed, assign another MP unit assigned to an S-VOL.
You can expand the capacity of a DP-VOL used in a GAD, TrueCopy, Universal Replicator,
ShadowImage, or Thin Image Advanced pair created between or on the following storage systems
while keeping the pair status in each software product.
You can expand the capacity of a DP-VOL that is used as a ShadowImage pair volume and is not
shared with other replication software. To expand the DP-VOL capacity in a configuration where an SI
pair shares the same volumes with other software, see the User Guide for the remote replication that
Note: If a DP-VOL capacity is expanded from 4 TB or less to more than 4 TB, all of the allocated area
before the expansion is copied when the first pair resync operation is performed after the expansion, or
the first pair split operation is performed after a pair quick resync operation.
Therefore, the first pair resync operation after the expansion, or the first pair split operation after the
pair quick resync operation will take a while to complete. The copy progress percentage might be
displayed as 100% even before the copy operation is complete.
Calculating the numbers of differential and pair tables for each pair
Note: If a DP-VOL exceeds 4 TB, the differential tables are placed in the hierarchy memory instead of
the shared memory, and the differential tables in the shared memory are not used. Therefore, it is not
necessary to calculate the number of differential tables for DP-VOLs over 4 TB.
Calculate how many differential tables are required for each pair using the following formulas:
number-of-differential-tables-for-expansion = number-of-differential-tables-
after-expansion - number-of-differential-tables-before-expansion
In the example, 'ceiling' refers to rounding up the value enclosed in parenthesis ( ) to the next
integer.
Calculate how many pair tables are required for each pair using the following formulas:
number-of-pair-tables-for-expansion = number-of-pair-tables-after-expansion
- number-of-pair-tables-before-expansion
In the example, 'ceiling' refers to rounding up the value enclosed in parenthesis ( ) to the next
integer.
3. Verify that the following numbers do not exceed the maximum number allowed on the storage
system:
The total numbers of current pair tables and pair tables required for expanding the
capacity.
The total numbers of current differential tables and differential tables required for
expanding the capacity.
Example:
Note: A maximum of 65,280 request IDs can be assigned to one storage system. If there is no
available request ID, the command fails with EX_IDEXHA. If that occurs, see the description of
the function of request IDs in the Command Control Interface Command Reference and release
request IDs.
4. Verify that expanding the P-VOL capacity in the SI pair is complete.
When using CCI, use the raidcom get command_status command to verify that the raidcom
extend ldev command processing is complete, and then verify that the LDEV capacity is
correct using the raidcom get ldev command.
Example:
5. Confirm that asynchronous command processing has completed, and then release a request
ID.
6. Expand the S-VOL capacity in the SI pair in the same way as the P-VOL capacity.
Note: If the S-VOL capacity in the SI pair cannot be expanded, perform the procedure in
Troubleshooting the expansion of DP-VOL capacity.
7. Verify that expanding the S-VOL capacity in the SI pair is complete in the same way as the P-
VOL capacity.
8. Verify that the volume capacity is not being expanded.
When using CCI, check if the P column in the pairdisplay command displays N.
Example:
The P column displays the status that the volume capacity is being expanded:
N: The volume capacity is not being expanded.
E: The volume capacity is being expanded. This status displays for approximately 30
seconds after the expansion of the P-VOL or the S-VOL capacity is complete, and then
this status is migrated to N.
When the SI pair shares the same volumes with other software, there is an order in which you must
expand the DP-VOL capacity. For more information about the specific expansion order, see the User
Guide for the remote replication that is used with the SI pair.
You can expand the capacity of DP-VOLs used in an SI pair and each software pair when an SI pair
shares the same volumes with the following software:
TrueCopy (TC)
global-active device (GAD)
Universal Replicator (UR)
Thin Image Advanced (HTI Advanced)
When an SI pair shares the same volumes with other software, the order in which you expand the DP-
VOL capacity consists of the following two expansion orders:
The following describes the guidelines for the order in which you expand the DP-VOL capacity when an
SI pair shares the same volumes with other software.
Figure. Expansion order when TC/UR/GAD pair shares the same volumes with local replication
1. When a remote replication (TC/UR/GAD) is included, expand first the S-VOL capacity on the
end of the remote replication, and finally expand the P-VOL capacity.
2. When a remote replication (TC/UR/GAD) shares the same volumes with a local replication (SI/
HTI Advanced), expand the capacity in each following group. Expand first the capacity of the S-
VOL group in the remote replication, and then expand the capacity of the P-VOL group in the
remote replication:
a. S-VOL group that consists of the S-VOL of the remote replication and all volumes of the
local replication with which the S-VOL shares the same volume.
b. P-VOL group that consists of the P-VOL of the remote replication and all volumes of the
local replication with which the P-VOL shares the same volume.
The expansion order in each group follows the expansion order in the local replication.
3. In the local replication (SI/HTI Advanced), expand the capacity in the order from the highest
layer to the lowest. The expansion order is not specified for the volumes in the same layer.
Refer to the following troubleshooting information if an issue occurs during the expansion of the DP-
VOL capacity.
Recovery procedure from failure if one volume capacity used in the SI pair
cannot be expanded
If only one volume capacity used in an SI pair can be expanded, while the other cannot be expanded,
the resync operation of the SI pair cannot be performed after the expansion because the capacity of
both the volumes is not the same.
When an SI pair is used with TC, GAD, UR, or HTI, if the capacity of the P-VOL and the S-VOL used in
one of the software products cannot be expanded, which causes the invalid state that the expansion of
the entire volume capacity is not complete.
The following describes the procedure for recovering from the invalid state.
1. Verify that both the P-VOL and the S-VOL used in an SI pair meet the prerequisites for
expanding the DP-VOL capacity described in Expanding DP-VOL capacity used in a
ShadowImage® pair.
When an SI pair is used with TC, GAD, UR, or HTI, the capacity of the P-VOLs and the S-VOLs
in all of the used software products must be expanded.
Verify that all of the volumes meet the prerequisites for expanding the capacity described in the
User Guide for each software product. If the prerequisites are not met, proceed to step 4.
2. When the prerequisites for expanding the DP-VOL capacity are met, retry the expansion, and
make sure that the capacity of the P-VOL and the S-VOL used in the SI pair is the same. If
there is not enough free space, increase the free space, and then expand the DP-VOL capacity.
When the SI pair is used with TC, GAD, UR, or HTI, retry the expansion of the volume capacity
in all of the used software products, and make sure that the capacity of the P-VOL and the S-
VOL used in each software product is the same.
4. Delete the SI pair, expand the volume capacity in the SMPL status, and then recreate the SI
pair.
If you want to restore it to the conditions before the expansion due to a capacity input error and
others, delete the SI pair, and then recreate the LDEV with the correct capacity. After that,
recreate the SI pair.
If the capacity of the P-VOL and the S-VOL used in the SI pair or the HTI pair is not the same,
you can only delete the SI pair or the HTI pair. You can read the S-VOL data used in the SI pair
or the HTI pair before the expansion. If you want to use the data before the expansion, delete
all pairs after the data reading is complete.
Recovery procedure from failure if one volume capacity used in the SI pair
has been expanded
Use the procedure to recover from a failure if only one volume capacity used in an SI pair can be
expanded, while a failure occurs on the other volume before the expansion.
Make sure to recover from the failure first by following the recovery procedure in Troubleshooting
ShadowImage.
However, depending on the operations during the recovery procedure, the following actions are
required as shown in the table below.
Resynchronizing an SI The resync operation cannot be performed because the capacity of the P-
pair VOL and the S-VOL used in the SI pair is not the same. Perform the
"Additional Steps" described below immediately before the SI pair resync
operation.
The SI pair cannot be created because the capacity of the target P-VOL and
Deleting an SI pair the S-VOL is not the same when recreating the SI pair. Perform the
Recovering from a "Additional Steps" described below immediately before the SI pair resync
operation.
volume failure
Recreating an SI pair
The SI pair cannot be created because the capacity of the target P-VOL and
Deleting an SI pair
S-VOL is not the same when recreating the SI pair. Make sure that the
capacity of the P-VOL and the S-VOL is the same when creating the
Deleting a volume
volumes.
Creating a volume
Recreating an SI pair
Additional Steps
Expand the capacity of volumes whose capacity has not been expanded, and then make sure that the
capacity of the P-VOL and the S-VOL used in the SI pair is the same.
When an SI pair is used with other software, make sure to expand the capacity of the P-VOLs and the
S-VOLs in all of the used software products. For more information about the expansion procedure, see
the User Guide for the remote replication used with the SI pair.
Monitoring ShadowImage
Monitor your SI system on an ongoing basis to keep track of pairs and volumes and their current and
past conditions. You can monitor pair status as well as pair synchronization rates, and you can check
the pair task history to review and confirm replication operations.
When managing pairs, you need to know the status of the pairs before performing any SI management
tasks.
The following table lists the pair tasks that you can perform for each status.
SMPL(PD) NO NO NO NO NO
COPY(RS-R)
NO NO NO YES YES
The pair tasks you can complete on cascaded pairs depend on the L1 and L2 pair status. The tasks
are dependent on the following:
The following table lists the L1 pair tasks that you can complete based on the related L2 status.
(Normal (Reverse
copy) copy)
The following table lists the L2 pair tasks that you can perform based on the related L1 status.
1. For L2 pairs, only Normal Copy or Quick Resync can be performed. You cannot perform a
reverse resynchronization (Reverse Copy or Quick Restore). For more information about
resynchronizing L2 pairs, see Resynchronizing or restoring ShadowImage pairs.
2. To split L2 pairs, the L1 pair status must be PSUS. For more information about the
prerequisites for splitting L2 pairs, see Splitting ShadowImage pairs.
The following table lists the node volume read/write L1 and L2 pair status.
COPY(PD) Read only Read Read only Read only Read Read only Read
only only only
PAIR
COPY(SP)
PSUS
COPY(RS) Read only Read Read only Read only Read Read only Read
only only only
COPY(RS-R)
PSUE
The following table lists the leaf vol read/write per L2 pair status.
L2 pair status
Read only Read only Read only Read/Write Read/Write Read only Read only
The SI pair tasks you can perform depend on the pair's status and the status of unaffected S-VOLs.
The following table lists the pair tasks you can perform based on the status of S-VOLs related to the P-
VOL in other pairs.
(Normal (Reverse
Copy) Copy)
(Normal (Reverse
Copy) Copy)
SMPL(PD) NO NO NO NO NO NO
The following table lists pair status names and descriptions, including the level of S-VOL access.
COPY(PD) CPPD1 The paircreate initial copy is in progress.2 Read/ Read only
The storage system accepts read/write to write
the P-VOL but does not accept write enabled
operations to the S-VOL.
PAIR PAIR The initial copy operation has completed Read/ Read only
and the volumes are paired. The storage write
system performs update copy operations. enabled
COPY
COPY(SP) The pair is in the process of being Steady Read/ Read only
Split. The storage system: write
enabled
1. Copies the differential data to the
S-VOL.2 The data in the S-VOL is
identical to the data in the P-VOL.
2. Splits the pair.
PSUS3
PSUS(SP) The pair is in the process of being Quick Read/ Read/
Split. The differential data is copied to the write write
S-VOL in the background.2 You cannot enabled enabled
delete pairs in this status.
PSUS4
PSUS The pair has been split. The storage Read/ Read/
system stops performing update copy write write
operations but accepts write I/Os for the enabled enabled
S-VOL. The storage system keeps track
of updates to split P-VOLs and S-VOL so
that you can Quick Resync.
CPRS1
COPY(RS) The pairresync CCI command is in Read/ Read only
progress.2 The storage system does not write
accept write I/Os for S-VOL. enabled
Resynchronizing split pairs copies only
the differential data to the S-VOL.
RCPY
COPY(RS-R) The reverse pairresync CCI command is Read only Read only
in progress.2 The storage system copies
only the S-VOL differential data to the P-
VOL. The storage system does not
PSUE
PSUE The storage system does the following: Read/ Read only
write
Suspends the pair. enabled
Continues accepting read and
write I/Os to the P-VOL.
Stops update copy operations to
the S-VOL.
Marks the P-VOL as differential
data. Resynchronizing a pair
copies the P-VOL to the S-VOL.
Notes:
Troubleshooting ShadowImage
There are different aspects of SI that you can troubleshoot; for example, SIMs that are reported by
storage systems requiring maintenance.
Overview
The storage system reports all SIMS regarding ShadowImage operations. All SIMS are recorded in the
storage system and reported to the management client. Each SIM also triggers an SNMP trap that is
sent to the hosts.
For details about SNMP and SIM reference codes, see the VSP One Block System Administrator
Guide.
1. Execute the raidcom get ctg command to find the Consistency Group ID with the STS in
CHG state.
2. Specify the Consistency Group ID that is not used by any pair in CCI.
3. Create the pairs that will be the target of the Consistency Group pair split operation
using the CCI of the host server.
If the host server is down or has failed, many CTGs that contain no SI pair might be created
and there might not be enough available CTGs. Under such conditions, if you are using CCI to
run the paircreate command with CTG pair-split, the command might be rejected.
In this case, perform the following steps to delete the CTG that contains no SI pair and rerun
the paircreate command:
1. Execute the raidcom get ctg command to find the Consistency Group ID with the STS in
CHG state.
2. Specify the Consistency Group ID that is not used by any pair using CCI.
3. Create the ShadowImage pairs or HTI Advanced pairs that will be the target of the
Consistency Group pair split operation using the CCI of the host server.
If the CTG ID is 128 or more, create an HTI Advanced pair. For more information about
how to create an HTI Advanced pair, see the Thin Image Advanced User Guide.
The following are possible reasons why the status for some pairs that are assigned to a CTG
are not changed:
The UR pair is assigned to a CTG and the P-VOL and S-VOL have the same content.
The journal volumes for this pair are full.
The SI license is invalid.
The SI pair volumes are blocked.
The SI pair is in a status that does not allow you to run the pairsplit CCI command.
The SI pair is a part of cascaded pairs, and the other pairs in the cascaded pairs are in
a status that does not allow you to split the pairs.
You are using an SI pair volume in a TC or UR pair, and the TC or UR pair is in a status
that does not allow you to run the pairsplit CCI command.
If you cannot change the status, the pairsplit CCI command can end abnormally with the error
code EX_EWSTOT, which indicates timeout occurrence. You cannot change the pair status
during a timeout.
If a pinned track occurs on an SI P-VOL or S-VOL, the storage system suspends the pair. Contact
customer support for assistance in recovering pinned tracks.
The HOST I/O Performance option is enabled. Disable the option (see Options for
ShadowImage).
The S-VOL's drive or external storage Make the configuration of the S-VOL's drive or
performance is lower than the P-VOL's. external storage the same as the P-VOL's.
The P-VOL's drive or external storage has an Review the error and make the necessary
error. correction.
The S-VOL's drive or external storage has an Review the error and make the necessary
error. correction.
The cache write pending ratio of the MP unit to Review the configuration.
which the P-VOL and S-VOL belong exceeds
60%.
The available capacity of the pool to which the Review the configuration. Check if SIMs related
capacity saving-enabled P-VOL and S-VOL to Dynamic Provisioning pool failures, including
belong is 120 GB or less. SIM code 62axx (DP Pool actual usage rate full),
is reported. If reported, take corrective actions for
each SIM.
You can use the CCI operation logs to troubleshoot tasks that you have performed. The following
procedure describes CCI error codes and how to locate and interpret them.
3. Locate the description of the SSB2 error code in the following table. Unless otherwise indicated,
these codes apply to SSB1 codes 2e31, b9a0, b9a1, b9a2, b9a5, b9a6, b9ae, and b9af.
For more information about the errors that are not described in the table, contact customer support.
0C94 The SI pair operation cannot be performed because the data reduction shared volume
specified as the P-VOL is currently unavailable.
0C95 The SI pair operation cannot be performed because the data reduction shared volume
specified as the S-VOL is currently unavailable.
200D The pair task was rejected because the specified DP-VOL is not associated with a
pool.
201B The CTG pair-split was rejected because the UR pair is other than PAIR/PSUS/PSUE.
The UR S-VOL was the SI P-VOL included in the CTG on which the pair-split is being
performed.
2026 The Quick Restore operation was rejected because the cache mode of the specified P-
VOL is different from the cache mode of the external S-VOL.
202D The pair task was rejected because one of the following conditions applies to the P-
VOL:
It is used as a global-active device pair volume, and the volume status does not
allow the pair task.
It is a volume for which the global-active device reserve attribute is set.
202E The pair task was rejected because one of the following conditions applies to the S-
VOL:
It is used as a global-active device pair volume, and the volume status does not
allow the pair task.
It is a volume for which the global-active device reserve attribute is set.
2036 The pair task was rejected because the global-active device quorum disk was specified
as the SI P-VOL.
2037 The pair task was rejected because the global-active device quorum disk was specified
as the SI S-VOL.
2038 The operation failed because multiple GAD pairs had been created by using the SI P-
VOL.
2043 The volume you specified as a P-VOL was a volume using two mirrors included in a 3-
UR DC multi-target or cascade configuration. The operation was rejected because the
volume was used as a UR delta resync or data volume.
2044 The volume you specified as an S-VOL was a volume using two mirrors included in a
3-UR DC multi-target or cascade configuration. The operation was rejected because
the volume was used as a UR delta resync or data volume.
2047 The pair task was rejected because the current firmware version does not support the
specified P-VOL capacity.
2048 The pair task was rejected because the current firmware version does not support the
specified S-VOL capacity.
205B The pair was not created because the specified MU number is in use.
2060 The volume you specified as an SI P-VOL was a volume of a UR pair. The pair task
was rejected because the status of the UR pair is not in the required status.
2061 The volume you specified as an SI S-VOL was a volume of a UR pair. The pair task
was rejected because the status of the UR pair is not in the required status.
206d The Quick Restore operation was rejected because the volume you specified as the P-
VOL is used as a volume for a GAD pair, and it is the migration source volume of
Volume Migration.
206f The Quick Restore operation was rejected because the volume you specified as the P-
VOL is used as a volume for a GAD pair, and a new GAD pair is being created or
resynchronized.
2072 The pair task was rejected because of one of the following reasons:
- While restoring the HTI pair, you attempted to create, split, or resynchronize
the SI pair.
- After you specified the MU number an HTI pair was using, you attempted to
create, split, or resynchronize the SI pair.
When the P-VOL is used as a P-VOL of a Thin Image Advanced pair that is in
the PSUP status, the Reverse Copy operation in SI is attempted.
2073 The pair task was rejected because of one of the following reasons:
2077 The Quick Restore operation was rejected because the volume you specified as the P-
VOL is used as a volume for a GAD pair, and the volume attribute (VOL_ATTR in the
raidcom get ldev command) of the S-VOL of the GAD pair is different from that of the
S-VOL of a ShadowImage pair.
2078 Because the specified P-VOL was also a UR P-VOL for delta resync, one of the
following errors occurred:
The Reverse Copy operation was rejected because the UR pair status is not
PSUS.
The Quick Restore operation was rejected.
2079 The pair task was rejected because the specified S-VOL was also a UR P-VOL for
delta resync.
2086 The pair task was rejected because the initialization process is being performed.
2089 The Quick Restore operation was rejected because you are formatting the volume you
specified as a P-VOL using Quick Format.
For more information about formatting volumes using Quick Format, see the
Provisioning Guide of your storage system.
208A The Quick Restore operation was rejected because you are formatting the volume you
specified as an S-VOL using Quick Format.
For more information about formatting volumes using Quick Format, see the
Provisioning Guide of your storage system.
2093 The pair task was rejected because the T10 PI attributes for the P-VOL and S-VOL do
not match.
2097 The Quick Restore operation was rejected for one of the following reasons:
The P-VOL is also an HDP V-VOL, but the S-VOL is an external volume.
The P-VOL is an external volume, but the S-VOL also an HDP V-VOL.
20A2 The create pair task was rejected because the P-VOL is a DP-VOL for which the
capacity is increasing.
20A3 The pair was not created because the S-VOL is a DP-VOL for which capacity is
increasing.
20A4 The pair operation was rejected because you specified either of the following volumes
for the P-VOL:
20A7 The pair operation was rejected because you specified either of the following volumes
for the S-VOL:
20A9 The pair task was rejected because HTI is using the specified CTG ID.
20B0 The pair task was rejected because the volume you specified as the P-VOL is a DP-
VOL and its capacity is increasing.
20B1 The pair task was rejected because the volume you specified as the S-VOL is a DP-
VOL and its capacity is increasing.
20B4 The pair task was rejected because the volume you specified as the P-VOL is a DP-
VOL that is not associated with a pool.
20B5 The pair task was rejected because the volume you specified as the S-VOL is a DP-
VOL that is not associated with a pool.
20B7 The pair task was rejected because an LU path or a namespace on the NVM
subsystem with an NVM subsystem port added is not defined to the volume you
specified as the P-VOL.
20B8 The pair task was rejected because an LU path or a namespace on the NVM
subsystem with an NVM subsystem port added is not defined to the volume you
specified as S-VOL.
20C3
The pair task was rejected because the volume you specified as the P-VOL is a
deduplication system data volume.
20C4
The pair task was rejected because the volume you specified as the S-VOL is a
deduplication system data volume.
20C5 The command was rejected because you were in the process of turning off the storage
system's power.
20D0 The P-VOL rejected the paircreate CCI command because the DP pool is initializing in
the DP-VOL.
20D1 The S-VOL rejected the paircreate CCI command because the DP pool is initializing in
the DP-VOL.
20DF The volume which was specified as the S-VOL cannot be used, because the volume is
undergoing online data migration.
20E9 The pair task was rejected because the volume you specified as the P-VOL is a S-VOL
for an existing pair, and the volume you specified as the S-VOL is the P-VOL for
20EC The volume which was specified as the P-VOL cannot be used, because the volume is
undergoing online data migration and Cache Through is specified as the cache mode.
20F2 The pair task was rejected because the physical serial numbers of the corresponding
storage systems do not match, even if the virtual serial numbers match, when the
serial numbers of the virtual storage machines are specified for the P-VOL and S-VOL.
20F4 The pair task was rejected for the volume specified as the P-VOL because either the
device model and serial number or the virtual LDEV ID for the virtual storage machine
is being changed.
20F5 The pair task was rejected for the volume specified as the S-VOL because either the
device model and serial number or the virtual LDEV ID for the virtual storage machine
is being changed.
20F6 The pair creation was rejected because one of the following applied to the volume
specified as the P-VOL.
20F7 The pair operation was rejected because one of the following applied to the volume
specified as the P-VOL.
20F8 The pair task was rejected for the volume specified as the S-VOL because the volume
is undergoing online data migration.
20FC The command was rejected because the volume specified as the P-VOL is undergoing
online data migration and the command is for creating a third pair.
2301 The pair task was rejected because there is not a sufficient amount of installed shared
memory or SI is not installed.
2306 The pair task was rejected because the LBA size of the specified P-VOL is not the
same as the size of the specified S-VOL.
2309 The pair was not created because the number of pairs exceeded the maximum number
of pairs.
230A The pair was not created because the volume you specified as the S-VOL is the P-
VOL of the SI pair that has an MU number of 0.
230B The pair task was rejected because the pair is being suspended or deleted.
Pair creation was rejected because the specified CTG ID had already been
used for an L1 pair.
Pair creation was rejected because the specified CTG ID had already been
used for an L2 pair.
Pair creation was rejected because the volume you specified as the P-VOL is
the S-VOL of the pair which is in the process of being Quick Split.
The Quick Restore operation was rejected because the VLL setting of the P-
VOL is different from that of the S-VOL.
The pair task was rejected because the specified P-VOL and S-VOLs were a
Compatible FlashCopy® pair.
The reverse resynchronization was rejected because the pair of the specified
P-VOL and S-VOLs is suspended (PSUE status).
The reverse resynchronization was rejected because the specified P-VOL and
the S-VOL is the L2 pair.
The CTG pair-split was rejected because some of the pairs in the CTG were
being resynchronized, split, or were already suspended.
The pair task was rejected because the pair status of the P-VOL, the S-VOL, or
both showed that the pair could not receive the issued command.
The Quick Restore or Reverse Copy operation was rejected because the
specified pair is an L2 pair.
2314 The pair was not created because the volume you specified as the S-VOL is the S-
VOL of another pair that has been split (PSUS status).
2322 The pair task was rejected because there is not a sufficient amount of installed shared
memory or initialization is not completed.
2324 The pair task was rejected because the number of slots of the volume you specified as
the P-VOL exceeded the upper limit.
2325 The pair task was rejected because the number of slots of the volume you specified as
the S-VOL exceeded the upper limit.
2326 The pair was not created because the volume you specified as the P-VOL had already
had three S-VOLs.
2327 The pair was not created because the node volume specified as the P-VOL had
already had two S-VOLs.
2328 The pair task was rejected because the pair configuration exceeded the number of the
layers of the cascade configuration.
2329 The pair task was rejected because the volume you specified as the S-VOL is the S-
VOL of an existing pair.
232A The pair was not created because pairs that would exceed the license capacity were
going to be created.
232F The pair task was rejected because the volume you specified as the P-VOL is
allocated as the destination of the Volume Migration.
2331 The pair task was rejected because the capacity of the specified P-VOL is not the
same as the capacity of the S-VOL.
2332 The pair was not created because the volume you specified as the P-VOL had already
had three S-VOLs.
2333 The pair task was rejected because the volume you specified as the P-VOL is not the
P-VOL of the existing pair.
233A The pairresync CCI command was rejected because the volume you specified as the
P-VOL is not an SI P-VOL.
233B The pair task was rejected because the volume you specified as the S-VOL is a root
volume.
233C The pair task was rejected because the volume you specified as the S-VOL is a node
volume, and the volume you specified as the P-VOL is not the P-VOL for the specified
S-VOL.
233D The pairsplit CCI command was rejected because the specified P-VOL and S-VOLs
were a L2 pair, and the L1 pair is not split (PSUS status).
233E The pair task was rejected because the volume you specified as the P-VOL is being
used as a TC P-VOL.
233F The pair task was rejected because the volume you specified as the S-VOL is the TC
P-VOL, and the pair is not split (PSUSstatus) or suspended (PSUE status).
2342 The pair task was rejected because the volume you specified as the S-VOL is the
destination of the Volume Migration.
2343 The pair was not created because the volume you specified as the S-VOL had already
been an S-VOL.
2344 The pair task was rejected because the volume you specified as the S-VOL for SI pair
tasks is not an S-VOL.
2346 The volume you specified as an SI S-VOL is a TC P-VOL. The pair task was rejected
because the TC pair is not in the required status.
2347 The volume you specified as an SI S-VOL was a TC S-VOL. The pair task was
rejected because the TC pair is not in the required status.
234B The pair task was rejected because the volume you specified as the S-VOL is the
volume of the Volume Migration.
234C The pair task was rejected because the volume you specified as the P-VOL is a
deduplication system data volume (Data Store).
234D The pair task was rejected because the volume you specified as the S-VOL is a
deduplication system data volume (Data Store).
2350 The pair task was rejected because the specified P-VOL and the S-VOL for SI pair
tasks is not a pair.
2351 The pair task was rejected because the volume you specified as the P-VOL and the
volume you specified as the S-VOL are the same.
2352 The pair was not restored because the specified P-VOL and S-VOLs is online to the
host.
2353 The pair was not deleted because the specified P-VOL and S-VOLs are in the process
of being Quick Split.
2354 The pairresync CCI command was rejected because the P-VOL and S-VOLs is in the
process of being Steady Split.
2357 The pair creation was rejected because the volume you specified as the S-VOL is the
P-VOL of the pair you are splitting or the P-VOL of the pair you are reverse
resynchronizing.
2358 The pairresync CCI command was rejected because the volume you specified as the
S-VOL is the P-VOL of the splitting pair.
235B The volume you specified as a P-VOL is a TC P-VOL. The reverse resynchronization
was rejected because the TC pair is not suspended (PSUE status) or split (PSUS
status).
235C The volume you specified as the P-VOL is a TC S-VOL. The reverse resynchronization
was rejected because the TC pair is not suspended (PSUE status) or split (PSUS
status).
235D The volume you specified as an S-VOL was a TC P-VOL. The reverse
resynchronization was rejected because the TC pair is not suspended (PSUE status)
or split (PSUS status).
236C The reverse resynchronization was rejected because the volume you specified as the
P-VOL has the S-VOL Disable attribute assigned by the Data Retention Utility.
236D The pair task was rejected because the volume you specified as the S-VOL has the S-
VOL Disable attribute assigned by the Data Retention Utility.
2370 The pair task was rejected because the volume you specified as the P-VOL is not
mounted.
2371 The pair task was rejected due to one of the following reasons:
2372 The pair task was rejected because the volume you specified as the P-VOL is being
formatted or shredded.
2373 The pair task was rejected because the volume you specified as the P-VOL is a
command device.
2380 The pair task was rejected because of one of the following reasons:
2381 The pair task was rejected due to one of the following reasons:
2382 The pair task was rejected because the volume you specified as the S-VOL is being
formatted or shredded.
2383 The pair task was rejected because the volume you specified as the S-VOL is a
command device.
2387 The pair was not created because the volume you specified as the P-VOL is the
volume for Volume Migration.
2390 The Quick Restore operation was rejected because the volume specified as the P-VOL
is a volume for which capacity saving function is enabled.
2391 The Quick Restore operation was rejected because the volume specified as the S-VOL
is a volume for which capacity saving function is enabled.
2394 The pair was not registered in the CTG because the number of the pairs assigned to
the CTG has exceeded the defined maximum number of pairs.
2395 The pair task was rejected because you are reverse resynchronizing the pair sharing
the specified volume as the P-VOL.
2396 The pair task was rejected because you are reverse resynchronizing the L1 pair
sharing the specified P-VOL as the root volume.
2397 The pair task was rejected because you are reverse resynchronizing the L2 pair
sharing the specified P-VOL or S-VOL as the node volume.
2398 The reverse resynchronization was rejected because the pair is not split (PSUS status)
or suspended (PSUE status).
2399 The reverse resynchronization was rejected because some of the pairs sharing the
specified volume as the P-VOL are not in split (PSUS status) or suspended (PSUE
status).
23BB The pair was not created because the volume you specified as the S-VOL could not be
used as the S-VOL because of Volume Security settings.
23EF The pair was not deleted because the P-VOL and S-VOL are in the process of being
Quick Split.
23F1 The pair was not created because the CTG identifier you specified is not supported.
9100 You cannot run the command because the system did not authenticate your user
information.
B911 The pair task was rejected because the specified volume did not exist.
B912 The pair was not created because the specified S-VOL does not exist.
B913 The pair task was rejected because the mirror ID is invalid.
The following table lists the SI pair tasks and options that CCI supports.
Command Option
-reset_system_opt
Command Option
The following table lists the SI consistency group (CTG) actions and options supported by CCI.
Command Option
Command Option
CTG pair-split
(undefined split time) No options pairsplit Not applicable
Resynchronize pairs
No options pairresync* Not applicable
Delete pairs
No options pairsplit* -S
Command Option
* You must use a CCI pair group to run the command on pairs in a CTG.
The following table lists the valid ranges of parameters for CCI options.
Parameter Range
CTG ID 0 to 127