The File Subsystem

This chapter discusses files in general and the risks associated with certain files or file attributes, ways to mitigate these risks, and ways to evaluate the security of these files on the system.

File Names

Files are always referred to by their names. The user or process that creates the file gives it a name .

A disk file name includes the file's NODE, VOLUME, SUBVOLUME and FILE names. Each part is separated by a period.

The NODE name represents the computer system where the file resides. The name can be one to eight characters long, but must begin with a backslash (\). When referencing a file, the NODE may be omitted if the file resides on the current default system.

The VOLUME name represents the physical or virtual disk where the file resides. The name can be one to eight characters long, but must begin with a dollar sign ($). When referencing a file, the VOLUME name can be omitted if the file resides on the current default VOLUME.

RISK Users on some remote nodes may not be able to use volume names on the local node that have names more than 7 characters long, including the dollar sign.

BP-VOLUME-NAME-01 VOLUME names should not exceed 7 characters in length on any node connected to a network.

The SUBVOLUME name is the equivalent of a directory name on a PC. It represents a set of files. The name can be 1 to 8 characters long, but must begin with a letter. When referencing a file, the SUBVOLUME can be omitted if the file resides on the current VOLUME and SUBVOLUME.

The FILE name represents one individual file. The filename can never be omitted. File names can be one to eight characters long, but must begin with a letter.

Types of Files

Disk files are created to store databases, programs, or text. HP Enscribe database, the NonStop database record manager, supports the following disk file types:

Unstructured

Structured

Key- sequenced

Relative

Entry-sequenced

Enscribe Database

Enscribe software is the database record manager of the Guardian operating system. It supports structured and unstructured files.

Unstructured Enscribe Files

Unstructured Enscribe files are basically just an array of bytes of data. The organization of an unstructured file is determined by its creator. Unstructured files often contain program code or text.

Structured Enscribe Files

Structured files are designed to contain databases. A database contains logical records (individual sets of data about separate items or people). Each type of structured file uses a different structured organization.

When creating a file, FUP is used to specify the structure of the file to match the type of data to be stored in the file. The security string can also be specified.

Structured files are one of:

Key-sequenced files

Entry-sequenced files

Relative files

Key-sequenced Enscribe files

In Key-Sequenced files, the primary key is a field or combination of fields within the record. Records are inserted in sequence based on the primary key and can be updated or deleted.

Relative Enscribe files

In Relative files, the primary key is a record number relative to the beginning of the file. Records are inserted at the end of the file or at the location specified by the relative key, and can be updated or deleted.

Entry-sequenced Enscribe files

In Entry-sequenced files, the primary key is a system generated key value. Records can only be inserted at the end of the file. Records can be updated, but not deleted.

There are two file formats:

Format 1

Format 2

Format 1 Files

Format 1 files are files with a size of up to 2 gigabytes of data.

Format 2 Files

Release Version G06.00 introduced a new Enscribe disk file format called Enscribe Format 2. They are frequently referred to as Type 2 files.

Type 2 or Format 2 files are very large files. They can be up to 1024 gigabytes of data. They are currently limited only by disk size.

Enscribe Format 2 files, are not supported for use with:

Object files

Queue files

EDIT files

RISK Format 2 files should be closely monitored to avoid filling up a disk.

NonStop SQL

HP NonStop SQL database is the HP NonStop server's implementation of the Structured Query Language (SQL). It is a distributed, relational database management system for on-line transaction processing.

NonStop SQL tables are not accessible from the Enform report writer, READ, WRITE commands or FUP commands other than FUP INFO.