Browse Prior Art Database

Method for filtered kernel tracing capability for network communication Disclosure Number: IPCOM000238534D
Publication Date: 2014-Sep-03
Document File: 3 page(s) / 44K

Publishing Venue

The Prior Art Database


A method for filtered kernel tracing capability for network communication is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Method for filtered kernel tracing capability for network communication

Disclosed is a method for filtered kernel tracing capability for network communication.

Today's operating systems have many tracing facilities (to collect debug data) to enable customers and developers identify/debug issues faced in development/production environments. Many of these OS traces are also fed into analytic/profiling engines for understanding the behavior of a software module/workload, which helps in tuning the workload better for a specific environment. Meaningful tracing can help a long way in understanding the behavior of a system better and would reduce the turnaround time for problem analysis/determination.

Most of today's enterprise workloads exchange tons of data across heterogenous/homogenous systems/server to achieve business specific transactions. These complex communication systems need an intelligent tracing framework. Although there are specific tools in the networking software area to trace/capture network packets specific to a workload or a server, they are not complete. Today there are tools available to capture packets exchanged in the network, but once these packets enter a system/server, there are limited or no capabilities in tracing ONLY what is needed. For example, the current system/kernel tracing framework is designed to capture trace information of every communication which happens in the system. There is no option for system user to collect system trace for a specific network communication path (or a connection). It is very time consuming to analyze data pertaining to a specific workload/connection among gigabytes of trace data collected in development/customer environment. Also since the trace capturing happens for each and every connection in a system, this also impacts the overall performance of the system. Thus, there is a need for an intelligent tracing framework to capture system trace information in a filtered

The disclosed method for filtered kernel tracing over the network communication system may also be extended for other modules of an operating system. Today's network packet capturing tools widely use the Berkeley Packet Filter (BPF...