Systemtap 5.0 Released: Debugging Tools for Linux Platform
A new version of the dynamic trace system, Systemtap 5.0, has been published for the Linux platform. This release provides debugging tools similar to DTRACE technology, allowing for thorough monitoring of a working Linux system.
Systemtap 5.0 allows users to collect statistics on application operations, perform profiling, and control system calls. The system can be managed through the command line interface and a special si-like language of scenarios.
The release is compatible with Linux kernels ranging from version 3.10 to 6.6-RC0.
The development of Systemtap 5.0 has been supported by several companies, including Red Hat, IBM, Intel, Hitachi, and Oracle.
Systemtap 5.0 offers over 190 scripts for various purposes, such as tracking memory distribution and conclusion, disk operations, network traffic analysis (e.g., NFS), problem planner work, interruption processing, system buffer usage, locks installation, system call execution, and signal processing. These scripts can be found here.
Main Changes:
- Continuity of the kernel branch 2.6.32 for Rhel6. Systemtap testing is now done with nuclei 3.10 (Rhel7), 4.18 (Rhel8 + Centos Stream 8), 5.14.0 (Rhel9 + Centos Stream 9), 6.4 (Fedora 38), 6.5 (Fedora 39), and 6.6-RC (Fedora Rawhide).
- Added support for the Debuginfod process, enabling the dynamic obtaining of debugging information and metadata about the installed executable files.
- Added support for debugging data in the DWARF5 format, which can be used in the PRINT_UBACKTRACE_FILELINE() function.
- In the frontend (Stap utility), masks can now be used to determine target processes, for example, “Stap -e ‘Probe Debuginfod.process(“/usr/*/curl”).Begin(“hi”)}'”, when Debuginfod servers are