Browse Prior Art Database

Automatic Command History Recording and Filtering for System/Application Failure Prevention

IP.com Disclosure Number: IPCOM000199969D
Publication Date: 2010-Sep-22
Document File: 4 page(s) / 26K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a novel system to prevent computer operator mistakes through methods that augment remote terminal connection applications by introducing modules that improve the ability to run command line tasks.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 29% of the total text.

Page 1 of 4

Automatic Command History Recording and Filtering for System /Application Failure Prevention

Many companies and organizations utilize client machines such as personal computers that connect to a network. Server administrators grant access rights to users who may need to access a specific server or resource to invoke a job or process. The users may use remote access command line tools or other resources to connect client machines to a server.

When a user invokes a program, the command lines are initiated through various methods. Command lines may be typed, copied/pasted from another application, or selected based on an existing terminal application's editing function based on previous typed command lines, (i.e., command history or recall). Once the command is initiated, the system executes the command, processes the job, and returns any results based on its function.

Systems commands may be executed in one of the following manners:

1. One-at-a-time with user confirmation (e.g., user hits Enter after each command)

2. Setup to process concurrent requests; user enters one command at a time or uses a background process

3. Based on priority level or predefined timetable; user schedules the command for execution using a tool such as "cron"

Presently, methods do not exist that prevent a user from accidentally pasting a set of commands that may harm the server. This problem is exacerbated because servers are often shared by other users who also require remote access. Commonly after such accidents the server administrator is required to correct the problem, sometimes even requiring physical access to a data center or lab to physically reboot the machine. Such mistakes hinder other users in invoking their processes, decrease server efficiency, impact business critical applications, and can result in breaches in Service Level Agreement (SLA) or Operational Level Agreement (OLA). In addition, command lines can consist of more than one line at a time; text can be either cut-off or incorrectly word-wrapped due to screen/application setup. Methods are needed to ensure command lines are input correctly.

No automated methods are known to address the problem. Presently, users must just double-check their commands before requested the server process them. Some solutions exist that are close to the proposed invention, but do not anticipate the key features of the proposed invention.

The disclosed solution employs a method that augments remote terminal connection applications by introducing modules that improve the ability to run command line tasks. The method is comprised of a system:

1. Receiving at least one command-line instruction

2. Detecting at least one instruction in at least one command-line instruction that contains one or more of: a syntax error and an estimated execution characteristic

1

Page 2 of 4

that exceeds a predetermined threshold based on a look-up to a historical command-line execution file

3. Performing one or more actions on at...