Build Stuff'16 Lithuania

Docker started small and got big in really short time. Believe it or not, you can learn to build your version of it too, within the duration of a conference talk! In the talk, we will assemble and run a Docker-like container with only standard command-line tools found in modern Linux distributions. First half of the presentation will cover the facilities Linux provides for Docker to build upon: cgroups, pid and network namespaces, overlay file systems. During the second half we will put theory to practice in a terminal session. At the end of the talk we will not only be armed with knowledge how to build an isolated container, but also have an actual container running. After the talk, you will have a good insight how Docker works, and understand how you could use its building blocks to build something that perfectly matches your environment. You’ll get most out of this session if you are comfortable in Linux terminal, for instance, configuring networking from command-line. Knowledge about Pid namespaces, cgroups or overlay filesystems is not required. You will learn it all that during the talk.

The Linux kernel has multiple "tracers" built-in, with various degrees of support for aggregation, dynamic probes, parameter processing, filtering, histograms, and other features. Starting from the venerable ftrace, introduced in kernel 2.6, all the way through eBPF, which is still under development, there are many options to choose from when you need to statically instrument your software with probes, or diagnose issues in the field using the system's dynamic probes. Modern tools include SystemTap, SysDig, ktap, perf, bcc, and others. In this talk, we will begin by reviewing the modern tracing landscape -- ftrace, perf_events, kprobes, uprobes, eBPF -- and what insight into system activity these tools can offer. Then, we will look at specific examples of using tracing tools for diagnostics: tracing a memory leak using low-overhead kmalloc/kfree instrumentation, diagnosing a CPU caching issue using perf stat, probing network and block I/O latency distributions under load, or merely snooping user activities by capturing terminal input and output.