Browse Prior Art Database

Mechanism To Aggregate And Index Cloud Computing Log Messages For Deployment Patterns

IP.com Disclosure Number: IPCOM000234149D
Publication Date: 2014-Jan-14
Document File: 3 page(s) / 31K

Publishing Venue

The IP.com Prior Art Database

Abstract

Deployment logs for Cloud Computing environments are unique, very large, and extremely difficult to use for debugging purposes . The use of a central cloud log indexing system to efficiently manage and organize messages related to pattern deployments is discussed in this article. The use of retaining "success" deployment log index references for comparing against failed pattern deployments is also a key aspect of the system that is discussed in this article.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 33% of the total text.

Page 01 of 3

Mechanism To Aggregate And Index Cloud Computing Log Messages For Deployment Patterns

With the prevalence of Cloud Computing and the complicated hardware and software environments associated with this technology, troubleshooting can be very difficult. One of the primary reasons for this is it is hard to find the "needle" in the "haystack" of log files and messages. In particular, when trying to identify problems with a deployment pattern that involves numerous Virtual Machines (VMs) and numerous software packages on a VM, the number of places to search for log files becomes very complicated.

The current troubleshooting in a cloud management environment involves the gathering and downloading of tens of thousands of log files that can be Gigabytes in total size. These files have to first be pulled from the various cloud components and virtual machines and then uploaded to a support organization. The tedious process of searching through this huge log collection set must then proceed. Information about a single pattern deployment is spread across numerous log files and intermixed with numerous other unrelated deployment messages. This can make the entire troubleshooting process excessively long and mentally fatiguing.

Although there are existing general log management and log analysis tools, there does not appear to be existing tools that support the efficient management and processing of log messages related to pattern deployments in a cloud environment. With all of the inter-related components and virtual machines associated with pattern deployment, getting a handle on root causes of problems in a cloud environment is very difficult.

To assist in trouble resolution related to the deployment of cloud patterns and the virtual machines and software that are associated with these patterns, this invention proposes the use of a cloud deployment log message indexing system. This system records indexes that point to and group together sets of related log messages. The log messages may exist across a varied set of log files spread across several cloud management components and virtual machines. The log indexing system provides a significant advantage by providing indexes to messages that are grouped together for a deployed pattern. This provides a quick means for finding and extracting relevant message from a varied set of remote log files and remote servers when deployment problems occur.

If a good representation of log messages for any given pattern deployment can be quickly accessed from a central location, the time spent searching through numerous locations should be significantly reduced. Currently it is a very slow and tedious process to get an overview of messages that relate to a given deployment.

The deployment log message indexes for a pattern deployment could also be accessed by a utility program to output information in a user friendly format and allow support personnel and developers to more easily identify what logs should be...