𐬺 Tinyo Runtime

Next level orchestration.

Why • Quickstart • Discord • 𝕏

Architecture

Core

Libs

The Tinyo runtime leverages several Linux libraries and features to implement containerization at a low level. Below is a list of the core components and their purposes:

The code is located here: features/runtime/core.rs.

• Linux Namespaces (unshare, CloneFlags) Used to isolate the container's process from the host system by creating separate namespaces for PID, network, UTS (hostname), and mount points. This provides the basic building blocks for process isolation in Linux.

• Mount API (mount, umount2, pivot_root) Responsible for setting up the container's root filesystem, making it independent of the host's file structure. pivot_root is crucial for replacing the current root filesystem with the container's filesystem, while mount and umount2 manage additional filesystem bindings.

• Process Control (fork, execvp, waitpid) Implements the forking mechanism to create a new containerized process and execute commands inside the container. fork creates a child process, execvp executes the container's entrypoint (e.g., /bin/sh), and waitpid monitors the lifecycle of the containerized process.

• Filesystem Utilities (mkdir, chdir) Ensures the necessary directory structure is created for the container's filesystem. chdir is used to change the working directory to the container's root during the filesystem setup.

• Hostname Management (sethostname) Allows the container to have its own hostname, providing isolation at the UTS (Unix Timesharing System) level. This is particularly useful for multi-container environments where hostname uniqueness is required.

• Temporary Filesystems (tmpfs) tmpfs is mounted for /dev to create an isolated and writable environment for device files. This ensures the container operates independently of the host's /dev.

• Proc and Sys Filesystems (proc, sysfs) These are mounted inside the container to provide system-level information and kernel interfaces specific to the container's process namespace. /proc is essential for process-related metadata, while /sys is used for interacting with kernel features.

---

PureLinux.org | Delivering to the open-source community what matters most.

Linux® is the registered trademark of Linus Torvalds in the U.S. and other countries.