Tokyo-based VA Linux Systems last week launched a collaborative development project with NTT Data to develop a crash-analysis tool for Linux.
According to the companies, this tool will provide crash-related data and crash analysis for IT departments. The project aims to have the tool in circulation by the first quarter of next year.
When compared with Unix-based systems, some in the industry argue that systems running Linux tend to be more prone to deadlock and less effective at pinpointing failures that occur at the interface between middleware, the operating system and the hardware that runs it.
Through recording all internal data automatically at the time of failure — coupled with the sophisticated analysis of this internal data — this Linux crash analysis tool will provide one solution to some of the problems that can occur on systems running open-source software.
Experts said there are several issues that the Linux collaborative environment is helping with today.
“One issue IT managers are facing is managing servers in a heterogeneous environment — how to avoid console hopping — and consolidating management, including lights-out and provisioning,” Tom Woolf, a spokesperson for Linux vendor Amphus, told LinuxInsider.
This is particularly helpful with managing mission-critical applications on Linux, he said. “The Intelligent Platform Management Interface standard is helping here, and some vendors, like Amphus, are creating consolidating technology to help IT managers administer mission-critical Linux servers in a heterogeneous network,” said Woolf.
New developments in the Linux kernel, like the Linux 2.6.6 kernel recently released, could help lead a reduction in future crashes or system glitches.
“We expect the 2.6 kernel to significantly increase the networking performance as well as the overall scalability of the system,” one source recently told LinuxInsider.
The main functions and features of the new crash-analysis tool include:
“The combination of NTT Data’s knowledge and experience in working with global systems and VA Linux’s unparalleled proficiency — in the Japanese market — with the Linux kernel will allow us to better meet the needs of the enterprise market, which demands the utmost level of reliability,” Tetsuya Ueda, president and CEO of VA Linux, said in a statement.
“I am confident that this joint development project will dramatically increase the reliability of the Linux system and make Linux’s position in the enterprise market an unwavering one,” said Ueda. The development of this new Linux tool should help provide diagnostic specialists with better information to make Linux itself and Linux software even better.
To ascertain the cause of the crash at the most fundamental level, the companies said, it is necessary to break down and analyze crash dump data recorded at the moment of the failure or system crash. This sort of analysis is only possible when undertaken by software engineers who are versed in the internal workings of Linux and open-source software.
Currently, Linux crash dump analysis is typically performed with basic debugging tools.