Introduction › CA SRM Services › Query Service › File Scanning

File Scanning

To improve your file search performance, you can scan files from a cached database or you can scan an entire volume. The cached data is put into the fast scan database (FSDB). This is a compressed highly efficient database for keeping file system information. The Fast File Scan (FFS) creates this database. The FSDB updates with each data collection. When you create a query, you select which method you want.

In addition to the FSDB, CA SRM uses segmented two-phase file search to alleviate the problem of large volume searches. The first and second phases run serially.

There are three file search operation modes:

• File system based distributed mode—The individual file searcher output is segmented for overlapping between the first and the second phase. This is identical to the previous file search with the addition of the segmented operation mode. We recommend this mode when real-time file data is needed and when attributes not collected by the fast file scan are needed.
• FSDB based distributed mode—The file search is executed and distributed on each managed file server (or on its proxy). The actual file search, however, operates on a previously created and cached FSDB. The output for every distributed file searcher agent is handled in the segment operation mode to overlap the first and the second phases. We recommend this mode when:
  • The environment and the query contain a large number of servers.
  • The amount of output records is significantly smaller than the number of traversed files (for example, when you request aggregated output).
• FSDB based central mode—The file search is always executed on the Application Server machine from a previously created and cached FSDB. After creation, this FSDB is automatically transferred from the remote host to the Application Server machine. We recommend this mode when:
  • The query contains only a few volumes with a large amount of files (for example, scanning NetApp Filers).
  • The query generates the same amount of output records as the number of traversed files (for example, when you generate detailed file reporting).

The file scan engine scans files under the system account privileges on each Windows managed computer regardless of the security that was provided during the registration of this managed computer. This ensures reporting on all of the files that reside on the target server.

To change the file scan default:

1. Right-click the Application Server icon in the system tray and select Maintain Configuration.

   The Application Server Configuration Values Setup dialog appears.

2. Select File Scan from the Sections drop-down list.
3. Select System account based file scan and then clear the Yes box.

Even if you run the scan under the system account, it still does not guarantee that you will receive the file-system information. According to Microsoft documentation, the daily users of those computers that you want to scan can limit the default system account privileges that CA SRM is trying to use.

Note: This feature is relevant only to the Windows managed servers. For Unix file scanning, you need to change the rsc.conf flag on each UNIX server that you want scanned.