Linux

👣 Intro Tracing tools are pretty popular in the Unix/Linux ecosystem; for example in the userspace we have ltrace to trace library calls of the programs and strace to dive in deeper and inspect syscall usage. One of the many features that Linux kernel offers since 2008 (then evolved) is ftrace that allows many different kind of tracing at runtime. While not as flexible as eBPF technology, it can be helpful in some occasion and doesn’t require a full fledged programming language. ...

🔐 Intro As an experienced sysadmin, you might be familiar with the traditional “all-or-nothing” approach: if a shell or process is running with UID==0, it can do almost everything on a system; while a plain user process is restricted by some means: tipically it can’t open RAW sockets, can’t bind “privileged” ports under 1024, can’t change a file ownership and so on. Linux capabilities is a feature, gradually introduced starting from kernel 2.2, that permits a more fine-grained control over privileged operations, breaking the traditional binary root/non-root distinction. Just as by using sudo we can run specific commands as another user (even root), without permanently becoming that user, by using capabilities, we can grant a program only certain privileges without having to run it as root. ...

🌡️ Intro For those running a datacenter, or just a simple homelab server, the arrival of summer heat means an increase in air conditioning use. On this post I asked myself how a Linux engineer can measure how much energy is the system consuming so we can start to reason about workload optimization for better power consumption patterns. 🔋 Idle power drain As a starting point, let’s measure how much power my PC is consuming when idle, doing absolutely nothing; or better: nothing useful for computation or service but just running usual, default operating system tasks. ...

🤓 Intro Manual testing of GUI applications can become tedious and error-prone as features and complexity increase. Headless testing offers a solution by automating UI interactions without the need for a physical display. This approach allows for faster test execution, improved repeatability, and seamless integration with continuous integration and continuous delivery (CI/CD) pipelines. In this post, we’ll explore how to leverage pytest-qt, a powerful framework for headless testing of PySide/PyQt applications. ...