Repository messaging and storage

ctrl+alt+d · repository channels, users, messages and the product origin object pool

L2 object pool

Getting started

Quickstart

Boot your first machine in a couple of minutes: get a machine, power it on in a browser tab, type at a real Linux shell, and (optionally) mount its drive on your own computer. Nothing to install for the in-browser part — a modern desktop browser is enough.

1. Create an account

Sign up and verify your email. You can browse Explore and open public machines without an account, but you need one to create or fork machines of your own.

2. Get a machine

A machine is a versioned root filesystem that boots — it looks like a repository, but launching it gives you a running Linux system (see Machines). Two ways to get one:

  • Fork an existing machine. Open one from Explore — for example a bash + coreutils image — and fork it. Your fork is a writable copy under your own account.
  • Create one at brintos.io/new. Start from one of our templates, fork a backup of a machine you can read, or start from scratch with an empty drive.

3. Power it on

Check the selected version: your writable live version is chosen when it is idle. Otherwise, the launcher chooses the latest read-only backup. For the writing example below, use your own machine's live version. See version selection.

On the machine's page, press Launch machine, then Power on. The machine boots entirely in your tab — its root filesystem is served from cloud storage, and the kernel image named in its machine spec boots against it. The launch page gives you:

  • A serial console — a terminal pane that shows the kernel's actual boot log (dmesg), verbatim, and then your shell prompt. What you type goes to the guest.
  • A screen — a framebuffer canvas showing the machine's display. Its resolution is measured from the window at power-on and follows the window when you resize; keyboard and pointer input over the screen go to the guest as real input events.
  • Live dials — machine, storage, and network telemetry while the machine runs.

When boot finishes, the machine's init program is running as PID 1 — on the stock bash/coreutils images, that's a shell on the console.

4. Work at the shell

It's real Linux — a real kernel, real processes, a real filesystem. Try it:

uname -a
ls /
echo "hello from my machine" > /hello.txt
cat /hello.txt

On writable live, writes go back to the machine's cloud filesystem. After they have synced, a later live launch can read /hello.txt. Existing backups keep their captured contents; create a new backup while live is idle to share the changes.

5. Work from your computer (optional)

Use the Node launcher to run the machine in your terminal, or download a backup as an ext4 image for offline use. See Files & images for uploads and transfers.

Next steps

  • Launching — embed a machine, run it in Node, or boot an offline image.
  • Machines — the machine spec, snapshots, and the launch surface.
  • Developer guide — build your own packages, or your own kernel.