Browse Prior Art Database

e-business application monitoring tool Disclosure Number: IPCOM000014410D
Original Publication Date: 2001-Feb-01
Included in the Prior Art Database: 2003-Jun-19
Document File: 2 page(s) / 40K

Publishing Venue



e-business application monitoring tool

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

e-business application monitoring tool

e-business application monitoring tool

Current server monitoring tools provide the ability to check for server presence and process. The check for server presence will verify that the server is up and online, or on the network, through the use of a ping/ping reply test. If the machine replies, it has satisfied this test.

Process monitoring ensures that the program or programs that make up the application are up and running continuously by checking for the existence of a process ID (PID) on the server.

In a complex web based environment where the web server may make multiple back-end connections for dynamic content generation, there is the possibility that the server will pass the presence and process tests with the current monitoring tool set, but return an error to a web client due to a problem that can not be detected by presence and process monitoring.

The e-business application monitoring tool will simulate a web client, and perform a request to the monitored server on a time scheduled basis. It will ask for and interpret the HTTP/1.1 response codes and notify server support personnel in the event that a bad, or no return code is returned to the program by the server.

By performing a test of application functionality on the server, the tool is able to detect a potential problem and notify appropriate support personnel in a timely manner, eliminating or reducing potential customer outages.

The program would be written as a command line interface tool that could be run out of cron on a Unix platform machine. It should be compiled with support for SSL so that servers using SSL functions on port 443 can be monitored as well as those servers running standard HTTP on port 80. The monitoring process for either target server would be the same. The program would read a list of monitored URLs from a plain text file. This plain text file could be updated as needed between scheduled runs of the monitoring tool, providing the ability to react quickly to changes in the server environment. The URLs listed in the file should be valid on the target server machine, and provide the monitoring program with the ability to test the health of a specific function on the target server. In practical application in our environment, we were able to specify URLs that would test the servers ability to return a standard HTML page as well as those that would invoke an asp (Active Server Page), jsp (Java Server Page), servlet or cgi-bin program.

The program should read through the file of monitored URLs and perform the same set of analysis on each of the HTTP/1.1 response codes...