The following options are applicable to the management database (NASTEL_APWMQ). This database houses a number of different data types and this article is specific to configuration data captured from the managed systems. Examples of this type of data is IBM MQ, Solace or Rabbit MQ queue definitions, EMS or Kafka Topics, etc. This data is completely optional and management will operate without saving it to the database.
On startup, the data from these tables are read into the cache. This is done to have an initial state of the systems. However, this can take a long time if there are many objects so this can be changed. If "minimal cache startup" is enabled in the general properties, only data required by the management application is loaded and none of the configuration data is. This provides faster startup. If the managed systems are running, they will populate the data in the cache using current information. In most cases, this data refresh is faster than the restoring from the database.
For most customers this is the only option required since the writing of table data is done asynchronously. However, customers with large amounts of data may benefit from 2 other options added in WGS 10.5.0.51 or higher (from About page). These have to be specified via a property file and are not available as a WGS setting.
- apwmq.db.save.minimum = true/false : when true is selected, only the minimal amount of data is written to the DB. This would include all definitions data. Configuration data would not be written.
- apwmq.db.record.threshold = max : Defines the maximum number of rows that are written to the database. If the number of objects exceeds this amount, no data is written at all. The default value is 5000. A value of 0 indicates that all data should be written regardless of number of elements.
As discussed in Working with the Management Database, with the latest versions of the WGS, the configuration data in the database is not required. Large tables take up considerable database resources which can be better be spent on other database information such as storing events.