The Selector is a Java component on the Inventory server that receives the inventory information from the Inventory servers. These Inventory servers can be any of the following: Leaf Server, Leaf Server with Database, Intermediate Server with Database and Inventoried Servers, Intermediate Server with Inventoried Servers, Root Server with Inventoried Servers, and Standalone Server. See Section 14.9, An Overview of the Inventory Components on the Inventory Server for a quick reference table of Inventory server components.
The processing done by the Selector is as follows:
While scanning the inventoried server, the Scanner creates a scan data file (.str) in the scan directory (scandir) at the Inventory server for each scan done on the inventoried server. The location of scandir is obtained from the Inventory Service object. The Selector processes the .str files placed by the Scan Collector in the scandir directory.
The Selector checks for the following conditions to ensure that the .str file generated by the Scanner is valid:
The integer value that is generated by using the .str file and logged into the .str file by the Scanner and the integer value generated by using the .str file by the Selector should be the same.
The actual size of the .str file should be in sync with the size recorded in the .str file.
The Selector processes only valid .str files. If invalid files are present in the directory, the Selector deletes the files.
Based on the role of the Inventory server, the Selector copies the .str files to the dbdir directory (if the database is attached) and the entmerge directory. If the .str file already exists in the directory, it overwrites the file.
The following table lists the directories that the Selector copies or renames the files to:
The Selector logs the status in the Server log. For more information, see Monitoring Server Inventory Using Status Logs.