17.2.4 Storage configuration
You must provide the following file systems and disks when you use the cold standby configuration.
-
Host's local file system
In Figure 17‒1: Example of a system configuration using the cold standby configuration, LOC001 and LOC002 are the host's local file systems.
-
File systems subject to host switchover processing
In Figure 17‒1: Example of a system configuration using the cold standby configuration, FS001 to FS004 are file systems subject to host switchover processing.
-
Shared file system
In Figure 17‒1: Example of a system configuration using the cold standby configuration, FS005 is the shared file system.
-
Disks for DB area files
In Figure 17‒1: Example of a system configuration using the cold standby configuration, LU001 to LU005 and WRK001 are the disks for DB area files.
Note that some file systems and disks require device names used to identify disks in the active system and the standby system. If the path between the server and storage is a single-path configuration, we recommend that you use WWNs. When implementing redundant paths between server and storage, the device name will differ according to the redundancy method being used. Examples of device names in Figure 17‒1: Example of a system configuration using the cold standby configuration are as follows:
-
Examples of device names in single-path configuration
FS001: /dev/disk/by-id/wwn-0x60060e8010205850051104c500000005
FS002: /dev/disk/by-id/wwn-0x60060e8010205850051104c500000006
FS003: /dev/disk/by-id/wwn-0x60060e8010205850051104c500000007
FS004: /dev/disk/by-id/wwn-0x60060e8010205850051104c500000008
LU001: /dev/disk/by-id/wwn-0x60060e8010205850051104c50000000f
LU002: /dev/disk/by-id/wwn-0x60060e8010205850051104c50000000e
LU003: /dev/disk/by-id/wwn-0x60060e8010205850051104c500000010
LU004: /dev/disk/by-id/wwn-0x60060e8010205850051104c500000011
LU005: /dev/disk/by-id/wwn-0x60060e8010205850051104c500000013
WRK001: /dev/disk/by-id/wwn-0x60060e8010205850051104c50000000d
-
Examples of device names in redundant configuration using multipath software (DMMP)
FS001: /dev/mapper/mpath1
FS002: /dev/mapper/mpath2
FS003: /dev/mapper/mpath3
FS004: /dev/mapper/mpath4
LU001: /dev/mapper/mpath11
LU002: /dev/mapper/mpath12
LU003: /dev/mapper/mpath13
LU004: /dev/mapper/mpath14
LU005: /dev/mapper/mpath15
WRK001: /dev/mapper/mpath16
- Organization of this subsection
(1) Host's local file system
The directories to be placed in the host's local file system are as follows:
-
Server directory
To create a server directory, use the adbinstall command.
- Important
-
Create a server directory on both the active and standby systems. Do not create a server directory on the shared disk (host switchover target file system) or in a shared file system.
-
Output-directory for common format audit trails
For details about the output-directory for common format audit trails, see 2.18.9 Conversion of audit trail information (linkage with JP1/Audit Management - Manager).
In Figure 17‒1: Example of a system configuration using the cold standby configuration, LOC001 and LOC002 are the host's local file systems.
-
LOC001
Local file system of server machine hadb01 (active system)
-
LOC002
Local file system of server machine hadb02 (standby system)
(2) File systems subject to host switchover processing
The following file systems are subject to host switchover processing:
-
File system that stores the DB directory
-
File system that stores temporary work files
-
File system that stores synonym dictionary files
-
File system where the audit trail directory will be created
Create a file system subject to host switchover processing on an LV in a VG that consists of disks (shared disks) that can be referenced from both the active system and the standby system. For details about shared disks, see Shared disk configuration in the manual HA Monitor for Linux(R) (x86).
- Note
-
For details about how to create the DB directory, see 17.3.7 Creating a database.
In Figure 17‒1: Example of a system configuration using the cold standby configuration, FS001 to FS004 are the file systems inherited at host switchover.
-
FS001
File system that stores the DB directory
In this example, the name of the VG is vg_hadb01 and the name of the LV is hadb_db.
-
FS002
File system that stores temporary work files
In this example, the name of the VG is vg_hadb02 and the name of the LV is hadb_workarea.
-
FS003
File system that stores synonym dictionary files
In this example, the name of the VG is vg_hadb03, and the name of the LV is hadb_syndict.
The file system that stores synonym dictionary files is needed to perform synonym searches.
-
FS004
File system where the audit trail directory will be created
In this example, the name of the VG is vg_hadb04 and the name of the LV is hadb_audit.
The file system where the audit trail directory will be created is needed when the audit trail facility is used.
- Important
-
When using SCSI reservation for shared disk, make sure that disks on which file systems subject to host switchover processing have been created are subject to SCSI reservation processing. Do not create any area other than a file system subject to host switchover processing on a disk subject to SCSI reservation processing.
(3) Shared file system
The following file system is placed on a shared file system:
-
File system for storing input data files that are used for data import
-
File system for storing CSV files used by the ADB_CSVREAD function
-
File system where audit trail storage directory will be created
Configure the shared file system on a disk (shared disk) that can be referenced from both the active system and the standby system, and mount the shared file system in a directory with the same path in both the active system and the standby system. For details about shared disks, see Shared disk configuration in the manual HA Monitor for Linux(R) (x86).
In Figure 17‒1: Example of a system configuration using the cold standby configuration, FS005 is the shared file system.
-
FS005
File system for storing input data files for data import, file system for storing CSV files, and file system for storing the audit trail storage directory
For the shared file system, you can use, for example, an NFS, which is a distributed file system.
(4) Disks for DB area files
Provide a disk for each DB area file.
For the DB area files listed below, allocate a block special file.
-
Master directory DB area file
-
Dictionary DB area file
-
System-table DB area file
-
Data DB area file
Ensure that the disks for the following DB area files can be referenced from both the active system and the standby system:
-
Master directory DB area file
-
Dictionary DB area file
-
System-table DB area file
-
Data DB area file
-
Work table DB area file
- Important
-
When using SCSI reservation for shared disk, configure the disks that store the preceding DB area files as subject to SCSI reservation.
In Figure 17‒1: Example of a system configuration using the cold standby configuration, LU001 to LU005 and WRK001 are the disks for DB area files.
-
LU001
DB area disk that comprises a dictionary DB area
-
LU002
DB area disk that comprises a master directory DB area
-
LU003
DB area disk that comprises a system-table DB area
-
LU004
DB area disk that comprises a data DB area (ADBUTBL01)
-
LU005
DB area disk that comprises a data DB area (ADBUIDX01)
-
WRK001
DB area disk that comprises a work table DB area (ADBWRK)