This topic outlines the requirements when using a
Performance and Availability Cluster (PAC).
Platforms supported
- Windows
- x86-64 running Red Hat Linux 7.3 or 7.4
- x86-64 running on SUSE v12 SP2
RDBMS
The
PAC requires a supported RDBMS to manage the following databases:
- Region database (mandatory)
- Cross-region database (mandatory)
- Datastore database(s) (optional)
A datastore database should be used to manage VSAM files using the
Micro Focus Database File Handler support.
The
Micro Focus Database File Handler requires the following minimum support for databases and ODBC drivers.
Note: These are databases supported for database hosted VSAM files. See
Additional Software Requirements on Windows and
Additional Software Requirements for Micro Focus Enterprise Developer UNIX Components for more information on CICS SQL supported databases.
Current limitations
- The order in which duplicate keys are returned is not guaranteed for indexed files. For files stored on disk, duplicate keys
are always returned in the order they were written in.
- The supported key types are currently limited to alphanumeric, binary, and numeric.
- PostgreSQL does not support EBCDIC collating sequences. See
Configure Datastores for Db2 for more information.
Windows
| Database
|
ODBC Driver
|
| Microsoft SQL Server 2008 (including Express editions)
|
ODBC 13.1 for SQL Server
|
| PostgreSQL 10.x (incl. Amazon Aurora Postgres)
|
psqlodbc 11
|
| DB2 10.5
|
n/a
|
UNIX
| Database
|
ODBC Driver
|
| Microsoft SQL Server 2017 (including Express editions)
|
ODBC 13.1 for SQL Server
Note: SQL Server 2017 installs versions 13 and 17 (x64 only)
|
| PostgreSQL 10.x
|
psqlodbc 11
|
| DB2 10.5
|
n/a
|
Note: The
Micro Focus Native Database File Handler is only supported on x86-64 running Red Hat Linux 7.3 or 7.4 or x86-64 running on SUSE v12 SP2
as these are also the minimum requirements for Microsoft SQL Server 2017.
In addition, there are minimum requirements for the following components:
| Component
|
Version
|
| Driver manager (MSSQL and PostgreSQL installations only)
|
unixODBC 2.3.2
|
NoSQL
Enterprise Server must use a NoSQL database as a Scale-Out Repository, a requirement to maintain
PAC integrity.
Enterprise Server is Redis compatible. For a production environment,
Micro Focus recommends using an Enterprise-ready Redis compatible with 5.03 (Linux) or 3.2.100 (Windows) NoSQL database run on Linux
or Windows, as required. To facilitate development only, an Open Source version of Redis is supplied with the development
product.
Redis as a Scale-Out Repository
Enterprise Server supports Redis as a
Scale-Out Repository (SOR). Redis is an open source in-memory data structure store, although there are commercial versions available, such as Redis
Enterprise. AWS also provides Elasticache for Redis which enables easy management and maintenance of Redis instances.
COBOL Server for Windows and UNIX, ships with a version of Redis.
Note: The Windows version of Redis that is shipped with
Visual COBOL is for testing purpose only and is not intended for use in a production environment. For a production environment,
Micro Focus recommends using an Enterprise-ready Redis compatible NoSQL database run on Linux or Windows, as required. The open-source
Redis provided with the product can only be used in a UNIX environment.
To start the Redis server for UNIX environments, start a command prompt, and then type:
$COBDIR/AdoptRedis/redis-server
Note: This starts the Redis server on the default port 6379.
For more information on Redis click
here.
Redis Cluster
Enterprise Server supports Redis Cluster as a
SOR. Redis Cluster is a distributed implementation of Redis with performance, scalability, and availability as part of its design.
You can configure communication with Redis Cluster by setting the environment variable MFREDIS_CONFIG to point to a configuration
file. This configuration file can contain the following variables:
- RD_LUA_LOG
- This variable enables logging on the Redis server.
Note: The Redis server must be started with
--log-level set to
warning as a minimum.
- ON
- Enables logging.
- OFF
- Disables logging.
The default value is OFF.
- RD_RETRY_INTERVAL
- Specifies how long, in milliseconds, the Redis client should wait before reattempting to connect and perform the last failed
request.
The default value is 2000 (2 seconds).
- RD_RETRY_ATTEMPTS
- Specify the number of attempts the Redis client should try to perform the last failed request.
The default value is 1.