Skip to content
SHC Docs

Four machines

This ladder runs a cluster from nothing: it initializes the first node, licenses it, joins three more, moves work between them, breaks nodes and brings them back, upgrades the daemon, sets up the backup repository and restores the whole cluster onto wiped machines. Each page uses what the previous pages left running.

Prepare four fresh Debian 13 machines with 2 CPUs and 8 GB of memory each. Install SHC and its container runtime on every machine, following Installation. Don’t initialize or join them yet. Have a root shell on each and curl available. The machines need internet access to pull images and must allow the cluster ports listed in Installation between one another.

Name them home01, home02, home03 and far01. The first three are in location home; the fourth is in far. Both locations are in cloud learn. They are the same four names the app ladder uses, so the app ladder runs on this cluster too, and the last page of this ladder needs it to.

Each machine must know the others by those names, and root on each must be able to ssh home01 with a key and no prompt. A command that needs a value another machine holds asks that machine for it over ssh, so nothing is ever copied from one shell into another: a join reads its token from home01 that way.

Keep the SHC package you installed from at ~/shc.deb on every machine. The upgrade page installs it again.

Have a Pro licence that covers four nodes and put its file at ~/shc-license.token on home01. P@ssw0rd is the example password throughout this ladder. Use these machines only for learning: the last page wipes them.

A console block’s title names the machine where you type its commands. A file block’s title is a path: create the file there, on the machine of the console block that follows it, including its parent directories. ~ is root’s home directory. Type the lines beginning with $, without the $; the lines below are the output. ... omits output.

Next: The first node.