Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A smart method of automating problem recreation in an application server based on First Failure Data Capture

IP.com Disclosure Number: IPCOM000248792D
Publication Date: 2017-Jan-12
Document File: 2 page(s) / 14K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method to automate problem recreation in an application server based on FFDC(First Failure Data Capture) is disclosed. The method includes utilizing existing FFDC to extract system environment, server configuration, interface/feature usage, problematic binary application if available in dump, and then automatically generating an application server environment as a docker image with the binary application or an auto-generated dummy application. Then the application is invoked in the recreation environment to recreate the problem. A auto-comparison between original FFDC and recreation FFDC is then executed to verify whether recreation is successful.

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

1

A smart method of automating problem recreation in an application server based on First Failure Data Capture

Application Server is widely used as a middleware to ease application development by providing all kinds of features and APIs. When servicing application server product, a service engineer must be able to recreate problems that customers raised remotely in lab environment and then to confirm whether the problem is a software defect or is just due to some incorrect usage (e.g. application bug). Sometimes recreation itself is even more difficult than to fix a defect, and what makes thing worse, sometimes recreation in the lab does not match the real problem. In this case, the afterwards software patch would fail to solve the problem, and probably the customer would complain.

Let us have a look at normal steps to recreate a reported issue. Service people of a software need to:

1. Understand the problem by reading customer's description of the problem and by analyzing FFDC(First Failure Data Capture) data, e.g. log/trace/dump supplied by customer. Sometimes the customer might mislead the service into wrong direction. 2. Prepare the OS/Software with same version as customer has. 3. Configure the software by defining and installing necessary resources 4. Design dummy recreating programs to simulate customers' business logic 5. Run the dummy programs to trigger the same problem. A matched symptom from FFDC (log/trace/dump) should be generated.

Step 3 and 4 are usually very difficult, time consuming and error prone. The invention is to aim to make recreation steps of a problem automatic by reducing service people effort and by improving recreation accuracy.

The core idea is to utilize existing FFDC to extract system environment, server configuration, interface/feature usage, problematic binary application if available in dump, and then automatically generate an application server environment as a docker image with the binary application or an auto-generated dummy application. Then the application is invoked in the rec...