Configuration and pair management using CCI
You can use CCI commands to configure a global-active device system and manage GAD pairs.
Global-active device system configuration
The following illustration shows a completed GAD system configuration, which includes the following key components:
- Host server connected to the primary and secondary storage systems with management software (alternate path or cluster or both, depending on the system configuration).
- Primary and secondary storage systems at the primary and secondary sites. The following system components are configured on the storage systems:
- The volumes that will become the primary and secondary volumes of GAD pairs
- The GAD feature installed on the primary and secondary storage systems
- A virtual storage machine on the secondary storage system. Virtual storage machines can be configured on both the primary and secondary storage systems, but in this sample system configuration only the secondary storage system has a virtual storage machine.
- A resource group on the secondary storage system. Resource groups can be configured on both the primary and secondary storage systems, but in this sample configuration only the secondary storage system has a resource group.
- An external volume group on each storage system for the quorum disk
- Remote paths between the storage systems
- A CCI command device on each storage system
- External storage system with the quorum disk connected to both storage systems using Universal Volume Manager
Primary storage system settings
The primary storage system components used in the procedures and examples in this chapter have the following settings.
|
Model |
Serial number |
|
VSP G1x00 and VSP F1500 |
11111* |
|
* The serial number for the VSP Gx00 models and VSP Fx00 models is a six-digit number that begins with "4" (for example, 411111). | |
|
Actual LDEV ID |
Capacity |
Port attribute |
Port name |
LU number |
|
22:22 |
1,024,000 blocks |
Target |
CL1-A |
0 |
|
Port name |
Port attribute* |
|
CL3-A |
Initiator |
|
CL4-A |
RCU Target |
| *You do not need to set the port attributes (Initiator, RCU Target, Target, External) on the VSP Gx00 models and VSP Fx00 models. | |
|
Actual LDEV ID |
Port attribute |
Port name |
External volume group number |
Path group ID |
LU number |
Quorum disk ID |
|
99:99 |
External |
CL5-A |
1-1 |
1 |
0 |
0 |
|
Pair type |
Pair group name |
Device name (LDEV ID) |
Mirror ID |
CTG ID |
|
GAD |
oraHA |
dev1(22:22) |
0 |
2 |
Secondary storage system settings
The secondary storage system components used in the procedures and examples in this chapter have the following settings.
|
Model |
Serial number |
|
VSP G1x00 and VSP F1500 |
22222* |
|
* The serial number for the VSP Gx00 models and VSP Fx00 models is a six-digit number that begins with "4" (for example, 422222). | |
|
Actual LDEV ID |
Capacity |
Port attribute |
Port name |
LU number |
|
44:44 |
1,024,000 blocks |
Target |
CL1-C |
0 |
|
Port name |
Port attribute* |
|
CL3-C |
RCU Target |
|
CL4-C |
Initiator |
| *You do not need to set the port attributes (Initiator, RCU Target, Target, External) on the VSP Gx00 models and VSP Fx00 models models. | |
|
Actual LDEV ID |
Port attribute |
Port name |
External volume group number |
Path group ID |
LU number |
Quorum disk ID |
|
88:88 |
External |
CL5-C |
1-2 |
1 |
0 |
0 |
|
Resource group name |
Virtual storage machine | |
|
Model |
Serial number | |
|
HAGroup1 |
VSP G1x00 and VSP F1500 |
11111* |
|
* The serial number for the VSP Gx00 models and VSP Fx00 models is a six-digit number that begins with "4" (for example, 411111). | ||
|
Host group ID |
Host group name |
Usage |
|
CL1-C-0 |
1C-G00 |
For the S-VOL |
|
LDEV ID |
Parity group |
Capacity |
|
77:77 |
13-4 |
100 GB |
|
Pool ID |
Pool name |
Pool volume |
|
0 |
HA_POOL |
77:77 |
|
Pair type |
Pair group name |
Device name (LDEV ID) |
Mirror ID |
CTG ID |
|
GAD |
oraHA |
dev1(44:44) |
0 |
2 |
CCI server configuration
The CCI server configuration used in the procedures and examples in this chapter has the following settings.
Instance number | Configuration definition files | Usage |
0 | horcm0.conf | For the operation of the primary storage system |
1 | horcm1.conf | For the operation of the secondary storage system |
100 | horcm100.conf | For the operation of the primary storage system from the viewpoint of the virtual storage machine (serial number: 11111) |
101 | horcm101.conf | For the operation of the secondary storage system from the viewpoint of the virtual storage machine (serial number: 11111) |
For operations involving virtual storage machines, the parameters specified in the
raidcom command and the objects displayed by the
raidcom command are based on the virtual ID. In the procedures and
examples in this chapter, there is no virtual storage machine defined in the primary
storage system, but you can operate the primary storage system as if there is a virtual
storage machine with the same serial number and model as the primary storage
system.
External storage system settings
The external storage system used in the procedures and examples in this chapter has the following settings.
|
Model |
Serial number |
|
VSP G1x00 and VSP F1500 |
33333* |
|
* The serial number for the VSP Gx00 models and VSP Fx00 models is a six-digit number that begins with "4" (for example, 433333). | |
|
Storage system at destination |
WWN |
|
Storage system at the primary site |
50060e8007823520 |
|
Storage system at the secondary site |
50060e8007823521 |
Workflow for creating a GAD environment
- Initial state
- Adding the external storage system for the quorum disk
- Verifying the physical data paths
- Creating the command devices
- Creating the configuration definition files
- Starting CCI
- Connecting the primary and secondary storage systems
- Creating the quorum disk
- Setting up the secondary storage system
- Updating the CCI configuration definition files
- Creating the GAD pair
- Adding an alternate path to the S-VOL
Initial state
The initial state before GAD configuration consists of one host, one primary storage system, and one secondary storage system.
- Primary and secondary storage systems:
- For VSP G1x00 and VSP F1500, additional shared memory for GAD is installed in both storage systems.
- The GAD feature is installed on both storage systems.
- Resource group 0 exists by default in both storage systems.
- Dynamic Provisioning virtual volumes (DP-VOLs) are configured and have LU paths defined. These volumes will become the primary volumes of GAD pairs.
- Host:
- The required management software for your configuration, alternate path and /or cluster software, is installed.
- The CCI software is installed.
NoteThe creation of GAD pairs is not affected by the presence or absence of server I/O to the DP-VOLs.
Adding the external storage system for the quorum disk
Install an external storage system for the quorum disk. The storage system must be supported by Universal Volume Manager for connection as external storage.
If you use a disk in a server as the quorum disk, you do not need to install the external storage system.
Verifying the physical data paths
Make sure that the following physical data paths are connected and configured:
- From the primary storage system to the secondary storage system: two or more paths
- From the secondary storage system to the primary storage system: two or more paths
- From the primary storage system to the external storage system: two or more paths
- From the secondary storage system to the external storage system: two or more paths
- From the host to the primary storage system: two or more paths
- From the host to the secondary storage system: two or more paths
The following figure shows the physical data paths (redundant paths not shown). Although only one path is required for each location, it is important that you connect the storage systems using at least two physical paths. If you connect nodes using only one physical path, an unexpected failover might occur in the server, or the GAD pairs might be suspended, even though only one path or hardware failure has occurred. When maintenance is performed on the physical paths between storage systems, the GAD pairs must be suspended.
Creating the command devices
A command device (CMD) is required on each storage system for communication between CCI and the storage system. The command device must be created in resource group 0 in the primary storage system and in the secondary storage system. After the command devices have been created, host recognition must be set to the command devices.
Procedure
Using Hitachi Command Suite, allocate a command device in Resource Group 0 in both storage systems and enable user authentication. For details about creating a command device, see the Provisioning Guide for the storage system.
If necessary, change the topology and fabric settings for the ports defined to the command devices.
Define the volume to the port connected to the host.
Creating the configuration definition files
You must create four HORCM configuration definition files on the host for your GAD environment:
- One that describes the primary storage system and P-VOLs
- One that describes the secondary storage system and S-VOLs
- One for operating the virtual storage machine (SN: 11111) on the primary storage system
- One for operating the virtual storage machine (SN: 11111) on the secondary storage system
The configuration definition files for the examples follow. For details about creating CCI configuration definition files, see the Command Control Interface Installation and Configuration Guide. The examples show files on a Windows host.
When you specify a command device for HORCM_CMD, adding the serial number of the storage system and the LDEV number as shown is recommended.
HORCM_MON #ip_address service poll(10ms) timeout(10ms) localhost 31000 -1 3000 HORCM_CMD \\.\CMD-311111-52735
HORCM_MON #ip_address service poll(10ms) timeout(10ms) localhost 31001 -1 3000 HORCM_CMD \\.\CMD-322222-52734
HORCM_MON #ip_address service poll(10ms) timeout(10ms) localhost 31100 -1 3000 HORCM_CMD \\.\CMD-311111-52735 HORCM_VCMD # redefine Virtual DKC Serial# as unitIDs 311111
HORCM_MON #ip_address service poll(10ms) timeout(10ms) localhost 31101 -1 3000 HORCM_CMD \\.\CMD-322222-52734 HORCM_VCMD # redefine Virtual DKC Serial# as unitIDs 311111
Starting CCI
After creating the CCI configuration definition files, you can start the CCI software. Because you are not yet operating the virtual storage machine, you only need to start instances (0 and 1). You do not yet need to start instances (100 and 101) for the virtual storage machine.
Procedure
Start CCI instances 0 and 1.
horcmstart 0 1 starting HORCM inst 0 HORCM inst 0 starts successfully. starting HORCM inst 1 HORCM inst 1 starts successfully.
Enter the user name and password, and perform user authentication.
raidcom -login <username> <password> -IH0 raidcom -login <username> <password> -IH1
Next steps
-IH option in this example is used for each command to specify an instance. You can also perform the operation using a shell for each instance. To start the shell, specify an instance number to the environment variable HORCMINST, and then execute the command.Connecting the primary and secondary storage systems
To connect the primary and secondary storage systems, you will first set the port attributes on both storage systems, physically connect the storage systems, and then add the remote paths between the storage systems.
Setting the port attributes (VSP G1x00 and VSP F1500)
The Initiator and RCU Target port attributes must be set on the primary and secondary storage system ports for GAD command and data transfer. Initiator ports and RCU Target ports are required on both the primary and secondary storage systems.
Procedure
Change the attribute of port (CL3-A) on the primary storage system to Initiator.
raidcom modify port -port CL3-A -port_attribute MCU -IH0
Change the attribute of port (CL4-A) on the primary storage system to RCU Target.
raidcom modify port -port CL4-A -port_attribute RCU -IH0
Change the attribute of port (CL3-C) on the secondary storage system to RCU Target.
raidcom modify port -port CL3-C -port_attribute RCU -IH1
Change the attribute of port (CL4-C) on the secondary storage system to Initiator.
raidcom modify port -port CL4-C -port_attribute MCU -IH1
Use the same procedure to change the port attributes for the alternate paths. The alternate paths are not shown in the illustration.
- Display the port information for the primary storage system.
raidcom get port -IH0 PORT TYPE ATTR SPD LPID FAB CONN SSW SL Serial# WWN PHY_PORT (snip) CL3-A FIBRE MCU AUT E8 N FCAL N 0 311111 50060e80072b6720 - (snip) CL4-A FIBRE RCU AUT 97 N FCAL N 0 311111 50060e80072b6730 - (snip)
- Display the port information for the secondary storage system. Confirm that the port attributes have been changed as intended.
raidcom get port -IH1 PORT TYPE ATTR SPD LPID FAB CONN SSW SL Serial# WWN PHY_PORT (snip) CL3-C FIBRE RCU AUT D6 N FCAL N 0 322222 50060e800756ce22 - (snip) CL4-C FIBRE MCU AUT 7C N FCAL N 0 322222 50060e800756ce32 - (snip)
Adding remote connections
Add bidirectional remote connections between the primary and secondary storage systems. Specify the same path group ID to the bidirectional remote connections.
- When specifying the serial number for VSP G1x00 and VSP F1500 using CCI, add a “3” at the beginning of the serial number. For example, for serial number 11111, enter 311111.
- To specify
VSP G1000VSP G1x00 and
VSP F1500, enter
R800. - VSP G1x00 and
VSP F1500
are
displayed as
R8in command output. - To specify
VSP Gx00 models and
VSP Fx00 models, enter
M800. - VSP Gx00 models and
VSP Fx00 models are displayed as
M8in command output.
Procedure
Add a remote connection with path group ID 0 from primary storage system port (CL3-A) to secondary storage system port (CL3-C).
raidcom add rcu -cu_free 322222 R800 0 -mcu_port CL3-A -rcu_port CL3-C -IH0
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH0 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 311111 -
Add a remote connection with path group ID 0 from secondary storage system port (CL4-C) to primary storage system port (CL4-A).
raidcom add rcu -cu_free 311111 R800 0 -mcu_port CL4-C -rcu_port CL4-A -IH1
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
Add the alternate paths between the storage systems using the
raidcom add rcu_pathcommand. These alternate paths are not shown in the illustration.
On the primary storage system, display remote connection information.
raidcom get rcu -cu_free 322222 R800 0 -IH0 Serial# ID PID MCU RCU M/R T PNO MPORT RPORT STS_CD SSIDs ... 322222 R8 0 - - RCU F 0 CL3-A CL3-C NML_01 -
On the secondary storage system, display the remote connection information, and confirm that the serial number, model, and port name of the storage system are correct and that the path status is normal.
raidcom get rcu -cu_free 311111 R800 0 -IH1 Serial# ID PID MCU RCU M/R T PNO MPORT RPORT STS_CD SSIDs ... 311111 R8 0 - - RCU F 0 CL4-C CL4-A NML_01 -
Creating the quorum disk
When a failure occurs, the quorum disk is used by the primary and secondary storage systems to determine which pair volume contained the latest data when the failure occurred.
This section provides instructions for setting up the quorum disk. You will map the disk on the external storage system to the primary and secondary storage systems. Make sure the external volume is formatted before proceeding. You should be familiar with Universal Volume Manager to set up the quorum disk.
An external volume for a quorum disk must be mapped to one external volume group.
Setting the port attributes for connecting the external storage system
This section provides instructions for setting the ports on the primary and secondary storage systems (VSP G1x00 and VSP F1500) to the "External" attribute in preparation for connecting to the external storage system.
Procedure
Change the attribute of the port (CL5-A) on the primary storage system to External.
raidcom modify port -port CL5-A -port_attribute ELUN -IH0
Change the attribute of the port (CL5-C) on the secondary storage system to External.
raidcom modify port -port CL5-C -port_attribute ELUN -IH1
Display port information for the primary storage system.
raidcom get port -IH0 PORT TYPE ATTR SPD LPID FAB CONN SSW SL Serial# WWN PHY_PORT (snip) CL5-A FIBRE ELUN AUT E4 N FCAL N 0 311111 50060e80072b6740 - (snip)
Display the port information for the secondary storage system. Confirm that the port attributes have been changed as intended.
raidcom get port -IH1 PORT TYPE ATTR SPD LPID FAB CONN SSW SL Serial# WWN PHY_PORT (snip) CL5-C FIBRE ELUN AUT D5 N FCAL N 0 322222 50060e800756ce42 - (snip)
Creating external volume groups
Create external volume groups for the quorum disk to map the disk on the external storage system to the primary and secondary storage systems.
- Verify that the volumes in the external storage system are formatted.
- Use the
raidcom discover lun -portcommand to verify that the same E_VOL_ID_C value (volume identifier included in the SCSI Inquiry command of the external volume) is displayed for the primary and secondary storage systems.
For details about creating external volume groups, see the Hitachi Universal Volume Manager User Guide.
Procedure
Search for information about the external storage system port that is connected to port (CL5-A) on the primary storage system.
raidcom discover external_storage -port CL5-A -IH0 PORT WWN PM USED Serial# VENDOR_ID PRODUCT_ID CL5-A 50060e8007823520 M NO 33333 HITACHI VSP G1000
Display the LU that is defined to external storage system port (50060e8007823520) that is connected to primary storage system port (CL5-A). Check the LU number, and note the value shown in the
E_VOL_ID_Cfield.raidcom discover lun -port CL5-A -external_wwn 50060e8007823520 -IH0 PORT WWN LUN VOL_Cap(BLK) PRODUCT_ID E_VOL_ID_C CL5-A 50060e8007823520 0 61440000 OPEN-V HITACHI 500308235AAAA
Map the LU (0) that is defined to the external storage system port (50060e8007823520) that is connected to the primary storage system port (CL5-A). Specify 1 for the path group ID, and specify 1-1 for the external volume group number.
raidcom add external_grp -path_grp 1 -external_grp_id 1-1 -port CL5-A -external_wwn 50060e8007823520 -lun_id 0 -IH0
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH0 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 311111 -
Search for information about the external storage system port that is connected to port (CL5-C) on the secondary storage system.
raidcom discover external_storage -port CL5-C -IH1 PORT WWN PM USED Serial# VENDOR_ID PRODUCT_ID CL5-C 50060e8007823521 M NO 33333 HITACHI VSP G1000
Display the LU that is defined to external storage system port (50060e8007823521) that is connected to secondary storage system port (CL5-C). Check the LU number, and confirm that the
E_VOL_ID_Cfield displays the same value as in step 2.raidcom discover lun -port CL5-C -external_wwn 50060e8007823521 -IH1 PORT WWN LUN VOL_Cap(BLK) PRODUCT_ID E_VOL_ID_C CL5-C 50060e8007823521 0 61440000 OPEN-V HITACHI 500308235AAAA
Map the LU (0) that is defined to external storage system port (50060e8007823521) that is connected to secondary storage system port (CL5-C). Specify 1 for the path group ID, and specify 1-2 for the external volume group number.
raidcom add external_grp -path_grp 1 -external_grp_id 1-2 -port CL5-C -external_wwn 50060e8007823521 -lun_id 0 -IH1
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
- On the primary storage system, display information about the external path to the volume in the external storage system.
raidcom get path -path_grp 1 -IH0 PHG GROUP STS CM IF MP# PORT WWN PR LUN PHS Serial# PRODUCT_ID LB PM 1 1-1 NML E D 0 CL5-A 50060e8007823520 1 0 NML 33333 VSP G1000 N M
- On the secondary storage system, display information about the external path to the volume in the external storage system. Confirm that the external storage system information is correct, including serial number, model, and WWN, and confirm that the path status and volume status are normal.
raidcom get path -path_grp 1 -IH1 PHG GROUP STS CM IF MP# PORT WWN PR LUN PHS Serial# PRODUCT_ID LB PM 1 1-2 NML E D 0 CL5-C 50060e8007823521 1 0 NML 33333 VSP G1000 N M
Creating external volumes
Using capacity in the external storage system, you will create virtual external volumes on the primary and secondary storage systems that will be mapped to the quorum disk.
Procedure
Specify external volume group (1-1) assigned to the primary storage system to create an external volume whose LDEV ID is 0x9999. Allocate all capacity in the external volume group.
raidcom add ldev -external_grp_id 1-1 -ldev_id 0x9999 -capacity all -IH0
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH0 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 311111 -
Specify external volume group (1-2) assigned to the secondary storage system to create an external volume whose LDEV ID is 0x8888. Allocate all free space in the external volume group.
raidcom add ldev -external_grp_id 1-2 -ldev_id 0x8888 -capacity all -IH1
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
Display information about the volume (LDEV ID: 0x9999).
raidcom get ldev -ldev_id 0x9999 -fx -IH0 Serial# : 311111 LDEV : 9999 SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 61440000 NUM_PORT : 0 PORTs : F_POOLID : NONE VOL_ATTR : CVS : ELUN E_VendorID : HITACHI E_ProductID : OPEN-V E_VOLID : 484954414348492035303330324236374141414100000000000000000000000000000000 E_VOLID_C : HITACHI 500308235AAAA................ NUM_E_PORT : 1 E_PORTs : CL5-A-0 0 50060e8007823520 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0007 RSGID : 0
Display the information about the volume (LDEV ID: 0x8888). Confirm that the information about the external volume is correct.
raidcom get ldev -ldev_id 0x8888 -fx -IH1 Serial# : 322222 LDEV : 8888 SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 61440000 NUM_PORT : 0 PORTs : F_POOLID : NONE VOL_ATTR : CVS : ELUN E_VendorID : HITACHI E_ProductID : OPEN-V E_VOLID : 484954414348492035303330324236374141414100000000000000000000000000000000 E_VOLID_C : HITACHI 500308235AAAA................ NUM_E_PORT : 1 E_PORTs : CL5-C-0 0 50060e8007823521 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0005 RSGID : 0
Setting external volumes as quorum disks
This section provides instructions for setting the virtualized external volumes in the primary and secondary storage systems as quorum disks. The same quorum disk ID must be set to the primary and secondary storage systems.
The serial number and model of the paired storage system is specified for the
-quorum_enable option of the
raidcom modify ldev command.
- When specifying the serial number for VSP G1x00 and VSP F1500 using CCI, add a “3” at the beginning of the serial number. For example, for serial number 11111, enter 311111.
- To specify
VSP G1x00 and
VSP F1500, enter
R800. - VSP G1x00 and
VSP F1500
are
displayed as
R8in command output. - To specify
VSP G350,
VSP G370,
VSP G700,
VSP G900,
VSP F350,
VSP F370,
VSP F700,
VSP F900, enter
M800. - VSP G350,
VSP G370,
VSP G700,
VSP G900,
VSP F350,
VSP F370,
VSP F700,
VSP F900 are displayed as
M8in command output.
Procedure
Specify 0 to the quorum disk ID to set the volume (0x9999) in the primary storage system (serial number 22222, entered as 322222) as a quorum disk. Specify the storage system's serial number and model.
raidcom modify ldev -ldev_id 0x9999 -quorum_enable 322222 R800 -quorum_id 0 -IH0
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH0 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 311111 -
Specify 0 to the quorum disk ID to set the volume (0x8888) in the secondary storage system (serial number 11111, entered as 311111) as a quorum disk. Specify the storage system's serial number and model.
raidcom modify ldev -ldev_id 0x8888 -quorum_enable 311111 R800 -quorum_id 0 -IH1
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
Display the information about the volume (LDEV ID: 0x9999).
raidcom get ldev -ldev_id 0x9999 -fx -IH0 Serial# : 311111 LDEV : 9999 SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 61440000 NUM_PORT : 0 PORTs : F_POOLID : NONE VOL_ATTR : CVS : ELUN : QRD E_VendorID : HITACHI E_ProductID : OPEN-V E_VOLID : 484954414348492035303330324236374141414100000000000000000000000000000000 E_VOLID_C : HITACHI 500308235AAAA................ NUM_E_PORT : 1 E_PORTs : CL5-A-0 0 50060e80072b6750 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0007 QRDID : 0 QRP_Serial# : 322222 QRP_ID : R8 RSGID : 0
Display the information about volume (LDEV ID: 0x8888). Confirm that the following values are correct:
- QRDID (quorum disk ID)
- QRP_Serial# (serial number of the storage system that forms a GAD pair)
- QRP_ID (model of the storage system that forms a GAD pair)
raidcom get ldev -ldev_id 0x8888 -fx -IH1 Serial# : 322222 LDEV : 8888 SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 61440000 NUM_PORT : 0 PORTs : F_POOLID : NONE VOL_ATTR : CVS : ELUN : QRD E_VendorID : HITACHI E_ProductID : OPEN-V E_VOLID : 484954414348492035303330324236374141414100000000000000000000000000000000 E_VOLID_C : HITACHI 500308235AAAA................ NUM_E_PORT : 1 E_PORTs : CL5-C-0 0 50060e80072b6760 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0005 QRDID : 0 QRP_Serial# : 311111 QRP_ID : R8 RSGID : 0
Setting up the secondary storage system
This section provides instructions for creating a virtual storage machine (VSM) in the secondary storage system and configuring it for GAD pair operations.
To create a virtual storage machine, you add resources such as host group IDs and LDEV IDs to a resource group that is created for the virtual storage machine. You can also reserve the host group and the volume ID by only adding them to the resource group. Create a host group and a volume (actual volume) by specifying the reserved IDs so that the host group and the volume can be used on a GAD pair.
The following procedures describe how to create a GAD environment. If appropriate, you can use existing storage system resources, for example, Dynamic Provisioning pools and DP-VOLs that have already been created.
Creating a resource group (P-VOL not registered to a VSM)
When GAD setup is complete, the host sees the P-VOL and S-VOL of each pair as a single volume in a single storage system. Resource groups are created in the secondary storage system using the primary storage system's serial number and model as virtual information, so that the P-VOL and S-VOL of each pair share the same virtual storage machine information.
A virtual storage machine is composed of multiple resource groups that have the same virtual information. When you create a resource group and specify the virtual serial number and model, the resource group is registered in the virtual storage machine. If the virtual storage machine does not already exist in the storage system, it is created automatically when the resource group is created. The following illustration shows the creation of a resource group when the P-VOL is not already registered to a virtual storage machine.
- When specifying the serial number for VSP G1x00 and VSP F1500 using CCI, add a “3” at the beginning of the serial number. For example, for serial number 11111, enter 311111.
- To specify
VSP G1x00 and
VSP F1500, enter
R800. - VSP G1x00 and
VSP F1500 is displayed as
R8in command output. - To specify
VSP G350, enter
M850S1. - To specify
VSP G370, enter
M850S2. - To specify
VSP G700, enter
M850M3. - To specify
VSP G900, enter
M850H. - To specify
VSP F350, enter
M850S1F. - To specify
VSP F370, enter
M850S2F. - To specify
VSP F700, enter
M850M3F. - To specify
VSP F900, enter
M850HF. - VSP Gx00 models and
VSP Fx00 models are displayed as
M8in command output.
Specify the primary storage system's serial number and model for the virtual storage machine you are creating on the secondary storage system.
raidcom add resource -resource_name HAGroup1 -virtual_type 311111 R800 -IH1
Display the information about the resource groups of the secondary storage system. Information about all resource groups is displayed. Confirm the resource group name, resource group ID, virtual serial number, and virtual model.
raidcom get resource -key opt -IH1 RS_GROUP RGID V_Serial# V_ID V_IF Serial# meta_resource 0 322222 R8 Y 322222 HAGroup1 1 311111 R8 Y 322222
raidcom delete resource -resource_name HAGroup1 -IH1
Creating a resource group (P-VOL registered to a VSM)
You can create GAD pairs using volumes in virtual storage machines. When you want to create a GAD pair using volumes in VSMs, the VSM for the volume in the secondary site must have the same model and serial number as the VSM for the volume in the primary site.
- When specifying the serial number for VSP G1x00 and VSP F1500 using CCI, add a “3” at the beginning of the serial number. For example, for serial number 11111, enter 311111.
- To specify
VSP G1x00 and
VSP F1500, enter
R800. - To specify
VSP, enter
R700. - To specify
VSP G200, enter
M800S. - To specify
VSP G400,
VSP G600,
VSP F400, F600, enter
M800M. - To specify
VSP G800,
VSP F800, enter
M800H. - To specify
VSP G350, enter
M850S1. - To specify
VSP G370, enter
M850S2. - To specify
VSP G700, enter
M850M3. - To specify
VSP G900, enter
M850H. - To specify
VSP F350, enter
M850S1F. - To specify
VSP F370, enter
M850S2F. - To specify
VSP F700, enter
M850M3F. - To specify
VSP F900, enter
M850HF. - To specify
HUS VM, enter
M700.
Specify the serial number and model of the virtual storage machine to which the P-VOL is registered for the virtual storage machine you are creating on the secondary storage system. Command for the sample configuration shown above:
raidcom add resource -resource_name HAGroup1 -virtual_type 55555 R700 -IH1
Reserving a host group ID
In the secondary storage system's resource group, you must reserve a host group ID to be used by the S-VOL.
Reserve a host group ID (CL1-C-1) in resource group (HAGroup1).
raidcom add resource -resource_name HAGroup1 -port CL1-C-1 -IH1
Display information about the host group that is set to port (CL1-C). Confirm that the port name, host group ID, and host group name are correct.
raidcom get host_grp -port CL1-C -resource 1 -IH1 PORT GID GROUP_NAME Serial# HMD HMO_BITs CL1-C 0 1C-G00 322222 WIN
-key host_grp option for the check command
allows you to display the reserved host group. The following example shows the result of
executing the check command.raidcom get host_grp -port CL1-C -key host_grp -resource 1 -IH1 PORT GID GROUP_NAME Serial# HMD HMO_BITs CL1-C 0 1C-G00 322222 WIN CL1-C 1 HAVol 322222 WIN CL1-C 2 - 322222 - CL1-C 3 - 322222 - CL1-C 4 - 322222 - CL1-C 5 - 322222 -
As shown in this example, the host groups with host group ID 0 to 5 are reserved in resource group 1. Actual volumes are defined for the host groups with host group ID 0 and 1. The host groups with host group ID 2 to 5 are reserved in the resource group, but actual volumes are not defined for them. The host groups with host group ID 6 to 254 are not displayed, because they are not reserved in resource group 1.
Deleting the virtual LDEV ID of the S-VOL
Delete temporarily the virtual LDEV ID of the volume to be added to the virtual storage machine.
Delete the virtual LDEV ID of the volume (0x4444).
raidcom unmap resource -ldev_id 0x4444 -virtual_ldev_id 0x4444 -IH1
Display information about the volume (LDEV ID: 0x4444). For the volume whose virtual
LDEV ID was deleted, fffe is displayed for VIR_LDEV
(virtual LDEV ID).
raidcom get ldev -ldev_id 0x4444 -fx -IH1 Serial# : 322222 LDEV : 4444 VIR_LDEV : fffe SL : - CL : - VOL_TYPE : NOT DEFINED SSID : - RSGID : 0
raidcom map
resource command (example: raidcom map resource -ldev_id 0x4444
-virtual_ldev_id 0x4444 -IH1). The default virtual LDEV ID is the same as
the actual LDEV ID. After reconfiguring the virtual LDEV ID, use the check command to
confirm that the virtual LDEV ID is the same as the actual LDEV ID.Reserving an LDEV ID for the S-VOL
In the newly created resource group, you will reserve an LDEV ID so that the volume is available to become the target volume of a global-active device pair.
Reserve the LDEV ID (0x4444) in the resource group (HAGroup1).
raidcom add resource -resource_name HAGroup1 -ldev_id 0x4444 -IH1
Display the information about volume (LDEV ID: 0x4444). Confirm that the number of the
resource group in which the LDEV ID was reserved is displayed for
RSGID.
raidcom get ldev -ldev_id 0x4444 -fx -IH1 Serial# : 322222 LDEV : 4444 VIR_LDEV : fffe SL : - CL : - VOL_TYPE : NOT DEFINED SSID : - RSGID : 1
Setting the GAD reserve attribute on the S-VOL
When you create a GAD pair, the P-VOL's LDEV ID is set as the virtual LDEV ID of the S-VOL. Before the pair can be created, the GAD reserve attribute must be set on the volume that will become the S-VOL. Setting the reserve attribute deletes the virtual LDEV ID of the volume, so that the virtual LDEV ID of the P-VOL can be set to the volume when the GAD pair is created.
Set the GAD reserve attribute to the LDEV ID (0x4444).
raidcom map resource -ldev_id 0x4444 -virtual_ldev_id reserve -IH1
Display the information about the volume (LDEV ID: 0x4444). For the LDEV ID to which
the reserve attribute was set, ffff is displayed for
VIR_LDEV (virtual LDEV ID).
raidcom get ldev -ldev_id 0x4444 -fx -IH1 Serial# : 322222 LDEV : 4444 VIR_LDEV : ffff SL : - CL : - VOL_TYPE : NOT DEFINED SSID : - RSGID : 1
raidcom unmap resource command (example: raidcom unmap
resource -ldev_id 0x4444 -virtual_ldev_id reserve -IH1). After releasing the
GAD reserve attribute, use the check command to confirm that
fffe is displayed for VIR_LDEV (virtual LDEV
ID).Creating a host group
Create a host group to be used by the S-VOL of the GAD pair. The requirement to create a host group depends on the host group ID used by the S-VOL:
- When the host group ID used by the S-VOL is 0
The host group whose host group ID is 0 exists by default. Set a host mode to the host group (Windows host mode shown), as follows:
raidcom modify host_grp -port CL1-C-0 -host_mode WIN -IH1
Check command and output examples: Display the information about the host group that is set for the port (CL1-C), and confirm that the host mode is correct.
raidcom get host_grp -port CL1-C -IH1 PORT GID GROUP_NAME Serial# HMD HMO_BITs CL1-C 0 1C-G00 322222 WIN
- When the host group ID used by the S-VOL is 1 or higher
A host group whose host group ID is 1 or higher does not exist by default. Create a host group used by the S-VOL of the GAD pair, and set the host mode.
If you connect multiple server hosts of different platforms to a port, create a host group for each platform.
Procedure
Create a host group (HAVol) with host group ID 1 on port (CL1-C).
raidcom add host_grp -port CL1-C-1 -host_grp_name HAVol -IH1
Set the host mode for the new host group (Windows host mode shown).
raidcom modify host_grp -port CL1-C-1 -host_mode WIN -IH1
Reserve host group (CL1-C-1) to resource group 1.
raidcom add resource -resource_name HAGroup1 -port CL1-C-1 -IH1
Display the information about the host group that is set for port (CL1-C). Confirm that the port name, host group ID, and host group name are correct.
raidcom get host_grp -port CL1-C -IH1 PORT GID GROUP_NAME Serial# HMD HMO_BITs CL1-C 0 1C-G00 322222 WIN CL1-C 1 HAVol 322222 WIN
Creating a pool
After creating host groups, you must create a pool volume, format the volume, and create a Dynamic Provisioning pool.
Procedure
Specify a parity group (13-4) to create a volume (pool volume) whose LDEV ID is 0x7777. The capacity is 100 GB.
raidcom add ldev -ldev_id 0x7777 -parity_grp_id 13-4 -capacity 100G -IH1
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
Perform Quick Format to the volume (0x7777).
raidcom initialize ldev -operation qfmt -ldev_id 0x7777 -IH1
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
Specify 0x7777 to the pool volume to create a pool for Dynamic Provisioning whose pool ID is 0 and whose pool name is HA_POOL.
raidcom add
dp_pool -pool_id 0 -pool_name HA_POOL -ldev_id 0x7777 -IH1Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
Confirm that volume (LDEV ID: 0x7777) is set for the pool volume of pool (pool ID: 0).
raidcom get ldev -ldev_id 0x7777 -fx -IH1 Serial# : 322222 LDEV : 7777 SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 209715200 NUM_LDEV : 1 LDEVs : 7777 NUM_PORT : 0 PORTs : F_POOLID : 0 VOL_ATTR : CVS : POOL RAID_LEVEL : RAID1 RAID_TYPE : 2D+2D NUM_GROUP : 1 RAID_GROUPs : 13-04 DRIVE_TYPE : DKR5E-J1R2SS DRIVE_Capa : 2286717474 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0007 RSGID : 0
Check the pool capacity.
raidcom get dp_pool -IH1
PID POLS U(%) AV_CAP(MB) TP_CAP(MB) W(%) H(%) Num LDEV# LCNT TL_CAP(MB)
000 POLN 0 98196 98196 70 80 1 30583 0 0 Check the pool name.
raidcom get pool -key opt -IH1 PID POLS U(%) POOL_NAME Seq# Num LDEV# H(%) VCAP(%) TYPE PM 000 POLN 0 HA_POOL 322222 1 30583 80 - OPEN N
Creating the S-VOL
Specify the volume that will become the S-VOL using the GAD reserve attribute and LDEV ID mapped earlier. The S-VOL must be the same size as the P-VOL.
Procedure
Check the capacity of the P-VOL.
raidcom get ldev -ldev_id 0x2222 -fx -IH0 Serial# : 311111 LDEV : 2222 SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 1024000 NUM_PORT : 0 PORTs : F_POOLID : NONE VOL_ATTR : CVS : HDP B_POOLID : 0 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0005 Used_Block(BLK) : 0 RSGID : 0
In the Dynamic Provisioning pool with pool ID 0, create a virtual volume (DP-VOL) with a capacity of 1,024,000 blocks and LDEV ID = 0x4444.
raidcom add ldev -pool 0 -ldev_id 0x4444 -capacity 1024000 -IH1
Confirm that asynchronous command processing has completed.
raidcom get command_status -IH1 HANDLE SSB1 SSB2 ERR_CNT Serial# Description 00c3 - - 0 322222 -
Display the information for volume (LDEV ID: 0x4444). Confirm that the new volume satisfies the following requirements:
- The GAD reserve attribute is set (virtual LDEV ID (
VIR_LDEV) isffff). - The volume has the same capacity as the P-VOL.
- The volume is a DP-VOL (volume attribute (
VOL_ATTR) isHDP).NoteThis example uses DP-VOLs. This requirement does not apply if you are using normal volumes.
raidcom get ldev -ldev_id 0x4444 -fx -IH1 Serial# : 322222 LDEV : 4444 VIR_LDEV : ffff SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 1024000 NUM_PORT : 0 PORTs : F_POOLID : NONE VOL_ATTR : CVS : HDP B_POOLID : 0 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0009 Used_Block(BLK) : 0 RSGID : 1
Adding an LU path to the S-VOL
Add an LU path between the port connected to the server and the S-VOL. The host does not recognize the S-VOL, because the virtual LDEV ID has not yet been defined to the volume for the S-VOL.
Specify host group (CL1-C-1) and LU (0) to add an LU path to S-VOL (0x4444).
raidcom add lun -port CL1-C-1 -lun_id 0 -ldev_id 0x4444 -IH1
Display the information about the LU paths that are defined in host group (CL1-C-1).
raidcom get lun -port CL1-C-1 -fx -IH1 PORT GID HMD LUN NUM LDEV CM Serial# HMO_BITs CL1-C 0 WIN 0 1 4444 - 322222
Updating the CCI configuration definition files
Before creating the GAD pair, you must update the CCI configuration definition files on the primary and secondary storage systems to add the information for the volumes that will become the P-VOL and S-VOL.
Shutting down CCI
You must shut down both CCI instances before editing the configuration definition files.
Shut down instance 0 and instance 1.
horcmshutdown 0 1 inst 0: HORCM Shutdown inst 0 !!! inst 1: HORCM Shutdown inst 1 !!!
Editing CCI configuration definition files
Seeing examples of configuration definition files can help you when you edit your own configuration definition files.
The following examples show the configuration definition files for a Windows host. The following information is important to know as you edit configuration definition files:
- Make sure to specify the actual LDEV IDs for the GAD pair volumes, not the virtual LDEV IDs.
- When specifying the serial number for
VSP G1x00 and
VSP F1500 using
CCI, add a “3” at the beginning of the serial number.
For example, for serial number 11111, enter 311111. For VSP Gx00 models, specify the six-digit serial number that begins with 4 as is (for example, 411111).
- If you use
CCI to operate
a
GAD pair created by
Device Manager - Storage Navigator, the MU# in the configuration definition file of
CCI must be the same as the mirror ID you specified when creating the
GAD pair by
Device Manager - Storage Navigator.
If the MU# does not match the mirror ID, you cannot operate the GAD pair using CCI.
The underlined lines below show the updates for the volumes in the sample configuration in this chapter. Make sure to enter the information for your system in your configuration definition files.
HORCM_MON #ip_address service poll(10ms) timeout(10ms) localhost 31000 -1 3000 HORCM_CMD \\.\CMD-311111-52735 HORCM_LDEV #GRP DEV SERIAL LDEV# MU# oraHA dev1 311111 22:22 h0 HORCM_INST #GPR IP ADR PORT# oraHA localhost 31001
The underlined lines below show the updates for the volumes in the sample configuration in this chapter. Make sure to enter the information for your system in your configuration definition files.
HORCM_MON #ip_address service poll(10ms) timeout(10ms) localhost 31001 -1 3000 HORCM_CMD \\.\CMD-322222-52734 HORCM_LDEV #GRP DEV SERIAL LDEV# MU# oraHA dev1 322222 44:44 h0 HORCM_INST #GPR IP ADR PORT# oraHA localhost 31000
Restarting CCI
After editing the configuration definition files, restart both CCI instances.
Start instances 0 and 1.
horcmstart 0 1 starting HORCM inst 0 HORCM inst 0 starts successfully. starting HORCM inst 1 HORCM inst 1 starts successfully.
Creating the GAD pair
After verifying that the virtual LDEV ID and enabling the ALUA mode, you can create the GAD pair.
Verifying the virtual LDEV ID at the secondary site
Before creating a GAD pair, check that the same virtual LDEV ID as that of the primary volume does not exist in the virtual storage machine of the secondary site, which has the same serial number and model as the primary storage system. If the same virtual LDEV ID as the primary volume exists, you cannot create the GAD pair.
Operate the virtual storage machine to check that the virtual LDEV ID does not exist. Specify the virtual storage machine for HORCM_VCMD of the configuration definition file, and then start CCI.
Procedure
Start instances (100 and 101) for confirming the virtual LDEV IDs.
horcmstart 100 101 starting HORCM inst 100 HORCM inst 100 starts successfully. starting HORCM inst 101 HORCM inst 101 starts successfully.
Confirm the P-VOL's virtual LDEV ID.
raidcom get ldev -ldev_id 0x2222 -key front_end -cnt 1 -fx -IH100 Serial# LDEV# SL CL VOL_TYPE VOL_Cap(BLK) PID ATTRIBUTE Ports PORT_No:LU#:GRPNAME ... 311111 2222 0 0 OPEN-V-CVS 1024000 0 CVS|HDP 0
Check that the same virtual LDEV ID as that of the primary volume does not exist in the virtual storage machine of the secondary site. After you execute this command, if virtual LDEV ID 0x2222 is not displayed, the same virtual LDEV ID (0x2222) as that of the primary volume does not exist in the virtual storage machine of the secondary site.
raidcom get ldev -ldev_id 0x2222 -key front_end -cnt 1 -fx -IH101
When you specify the virtual storage machine for HORCM_VCMD in the configuration definition file and execute the
raidcom get ldevcommand by specifying the-cntoption, the virtual LDEV IDs in the range specified by the-cntoption are displayed.TipTo display the volume information as a list for each volume, use the-key front_endoption for theraidcom get ldevcommand.
Revising the virtual LDEV ID at the secondary site
If the same virtual LDEV ID as the primary volume is displayed for the virtual storage machine of the secondary site, there might be errors in the GAD system implementation plan. Revise the system configuration.
The example when the same virtual LDEV ID as that of the P-VOL (0x2222) is assigned to the volume (LDEV ID: 0xfefe) in the virtual storage machine of the secondary machine is shown below.
Procedure
Check whether the same virtual LDEV ID as that of the primary volume is assigned to the virtual storage machine of the secondary site.
raidcom get ldev -ldev_id 0x2222 -key front_end -cnt 1 -fx -IH101 Serial# LDEV# SL CL VOL_TYPE VOL_Cap(BLK) PID ATTRIBUTE Ports PORT_No:LU#:GRPNAME ... 311111 2222 - - NOT DEFINED
The virtual LDEV ID (0x2222) is assigned to the virtual storage machine of the secondary site.
Confirm the actual LDEV ID of the volume whose virtual LDEV ID is 0x2222.
raidcom get ldev -ldev_id 0x2222 -fx -IH101 Serial# : 311111 PHY_Serial# : 322222 LDEV : 2222 PHY_LDEV : fefe SL : - CL : - VOL_TYPE : NOT DEFINED SSID : - RSGID : 1
In this example, the virtual LDEV ID (0x2222) is assigned to the volume whose actual LDEV ID is 0xfefe.
To use the virtual LDEV ID (0x2222) for a GAD pair volume, use the
raidcom unmap resourcecommand to remove assignment of the virtual LDEV ID (0x2222) from the volume whose LDEV ID is 0xfefe.raidcom unmap resource -ldev_id 0xfefe -virtual_ldev_id 0x2222 -IH1
Confirm that the assignment of the virtual LDEV ID (0x2222) is removed from the volume whose LDEV ID is 0xfefe.
raidcom get ldev -ldev_id 0x2222 -key front_end -cnt 1 -fx -IH101
When you specify the virtual storage machine for HORCM_VCMD in the configuration definition file, and execute the
raidcom get ldevcommand by specifying the-cntoption, the virtual LDEV IDs existing in the specified range by the-cntoption are displayed.After you execute the above command, if the virtual LDEV ID 0x2222 is not displayed, the same virtual LDEV ID (0x2222) as that of the primary volume does not exist in the virtual storage machine of the secondary site.
Next steps
raidcom get ldev command without specifying the -cnt option, the following error code and message are output:raidcom: [EX_EGPERM] Permission denied with the Resource Group
In the example above, the virtual LDEV ID (0x2222) has not been defined after you released the virtual LDEV ID assignment. Therefore, the user of the virtual storage machine does not have access authority.
When a command is executed specifying a virtual storage machine (that is, using HORCM_VCMD), both the actual ID and the virtual ID of the specified resource must be assigned to the user. When the virtual storage machine is not specified (that is, using HORCM_CMD), the user can execute the command only if the actual ID of the specified resource is assigned to the user.
Setting the ALUA mode
To specify the preferred path in a cross-path configuration, you must enable the ALUA mode. Before creating a GAD pair, enable the ALUA mode of the P-VOL. If the ALUA mode of the P-VOL is enabled, the ALUA mode of the S-VOL is also enabled when a GAD pair is created.
- It is not necessary to set the ALUA mode if the cross-path configuration is not used.
- To enable the ALUA mode, the following microcode/firmware is required in the storage systems at the primary and secondary sites: DKCMAIN microcode version 80-03-31-00/00 or later for VSP G1x00 and VSP F1500.
- If you add new LUNs, ensure that you set the ALUA attribute to that of existing LUNs. Otherwise, you will lose the settings on previously provisioned LUNs on the same host.
raidcom modify ldev -ldev_id 0x2222 -alua enable -IH0
Use the following command to verify that the ALUA mode is set correctly (ALUA: Enable is displayed).
raidcom get ldev -ldev_id 0x2222 -fx -IH0 Serial# : 311111 LDEV : 2222 SL : 0 CL : 0 VOL_TYPE : OPEN-V-CVS VOL_Capacity(BLK) : 1024000 NUM_PORT : 0 PORTs : F_POOLID : NONE VOL_ATTR : CVS : HDP B_POOLID : 0 LDEV_NAMING : STS : NML OPE_TYPE : NONE OPE_RATE : 100 MP# : 0 SSID : 0005 Used_Block(BLK) : 0 ALUA : Enable RSGID : 0
Creating a global-active device pair
When GAD configuration is complete, you can start creating GAD pairs. When a pair is created, the P-VOL LDEV ID is set as the S-VOL's virtual LDEV ID. When the paircreate operation completes, the pair status becomes PAIR, and the P-VOL and S-VOL can accept I/O from the host. When a pair is deleted, the S-VOL's virtual LDEV ID is deleted, and the GAD reserve attribute remains set on the S-VOL.
- When you create a GAD pair, make sure that the available pool capacity for Dynamic Provisioning below the warning threshold is more than the capacity of the secondary volume. If you create a GAD pair at the secondary storage system when the available pool capacity below the warning threshold is less than the capacity of the secondary volume, SIM (SIM=620xxx, where xxx is the pool ID) is issued (the used capacity exceeds the warning threshold).
- You cannot create a GAD pair by using instances (100 and 101) for confirming the virtual LDEV IDs. To create a GAD pair, use instances (0 and 1) for operating storage systems.
- Restarting the server is required if both of the following conditions are true:
- You create a GAD pair specifying the volume, that the server recognizes, as a primary volume with the ALUA mode disabled.
- You create a GAD pair with the ALUA mode of a primary volume enabled.
Specify 0 for the quorum disk ID and specify 2 for the consistency group ID to create a GAD pair.
paircreate -g oraHA -fg never 2 -vl -jq 0 -IH0
To register GAD pairs to a consistency group, specify the -fg option when creating the pair. If you create a pair and do not specify a consistency group ID, the ID for an unused consistency group in the storage system will be automatically assigned.
The following example shows a command used to create a GAD pair without registering the pair to a consistency group:
paircreate -g oraHA -f never -vl -jq 0 -IH0
Confirm that a GAD pair is created.
pairdisplay -g oraHA -fxce -IH0 Group PairVol(L/R) (Port#,TID, LU),Seq#,LDEV#.P/S,Status,Fence, %,P-LDEV# M CTG JID AP EM E-Seq# E-LDEV# R/W oraHA dev1(L) (CL1-A-0, 0, 0)311111 2222.P-VOL PAIR NEVER , 100 4444 - 2 0 - - - - L/M oraHA dev1(R) (CL1-C-1, 0, 0)322222 4444.S-VOL PAIR NEVER , 100 2222 - 2 0 - - - - L/M
Confirm that the copy progress reaches 100%.
Adding an alternate path to the S-VOL
Add an alternate path to the S-VOL on the host using the alternate path software. For some alternate path software the alternate path is added automatically. Make sure that the host has recognized the GAD secondary volume correctly.
dlnkmgr refresh -gad command to incorporate the storage
system settings to HDLM. For details about HDLM, see the HDLM user documentation.Setting a nonpreferred cross path
After adding an alternate path to the S-VOL of a GAD pair from a server, set the asymmetric access status of the host group.
In a cross-path configuration, if the ALUA mode is enabled for the first time, all paths including cross paths between a server and a storage system are set as the preferred path. Therefore, you need to set the asymmetric access status of the cross path as a nonpreferred path. By doing so, paths between the server and the storage system in the primary site and between the server and the storage system in the secondary site can be preferably used.
- It is not necessary to set the asymmetric access status if the cross-path configuration is not used.
- Before setting the asymmetric access status, enable the ALUA mode. See Setting the ALUA mode.
raidcom modify lun -port CL1-A-0 -lun_id all -asymmetric_access_state non_optimized -IH0
Check if the asymmetric access status is set correctly, and then make sure that the ALUA
mode (AL) is enabled (E) and that the nonpreferred
access status (AAS) of the cross path is set to nonpreferred
(AN).
raidcom get lun -port CL1-A-0 -key opt_page1 -fx -IH0 PORT GID HMD LUN NUM LDEV CM Serial# AL AAS CL1-A 0 LINUX/IRIX 0 1 2222 - 311111 E AN
Releasing the differential data managed in a pool
If differential data management fails due to insufficient pool capacity, you must release the differential data (pages) managed in the pool. You also need to release pages when you downgrade to a microcode version that does not support GAD pairs with DP-VOLs larger than 4,194,304 MB.
Procedure
Delete all GAD pairs that use the V-VOL for which you want to release the pages using the pairsplit -S command.
Make sure that system option mode 755 is set to OFF. When system option mode 755 is set to ON, you cannot reclaim zero pages. For details about setting system option modes, contact your Hitachi Vantara Corporation representative.
Restore the blocked pool. For details, see the Provisioning Guide for the storage system.
Release the V-VOL pages using the raidcom modify ldev command. Releasing pages might take some time. For details, see the Provisioning Guide for the storage system.