mesh-sdk/go/buildstep/isolate.go

package buildstep

import (
	"context"
	"fmt"
	"os"
	"path"
	"path/filepath"
	"strings"

	"git.novox.be/novox/mesh-sdk/go/forgekey"
)

// Bubblewrap is the program a run step runs in (novox/hq ADR 0448): a build module declares its package, which needs
// unprivileged user namespaces (issue 1308).
const Bubblewrap = "bwrap"

// Isolation is how a run step's command is shut off (novox/hq ADR 0448; issue 1309, the lead's decision L2 of R4e: one
// runner in the SDK for every build module). The fetch step holds the forge key and the network and runs none of the
// change's code; the run step runs the change's build and tests in bubblewrap with:
//
//   - no network: its own network namespace, with only a loopback, so a test may start a bus or a store of its own but
//     resolves no name and reaches no socket of the machine;
//   - none of the account: not its home, not the forge key, not a bus login, not the runtime directory (/run), since
//     none of them is bound; HOME is a directory in the step's own /tmp unless Env names one;
//   - an environment of exactly Env and the runner's env: nothing of the build module's own passes;
//   - the system's programs and libraries read-only, and of /etc only what a program needs to run (users, libraries,
//     certificates, time zone), with a hosts file naming only the loopback;
//   - the Workspace and Writable directories read-write, the ReadOnly ones (the toolchain, the modules the fetch step
//     downloaded) read-only;
//   - the shared warm Caches read-only under a throwaway writable layer (S18), or read-write for a trunk build alone
//     (WriteCaches), so a candidate's code never changes what a later build reads.
type Isolation struct {
	Workspace   string
	Writable    []string
	ReadOnly    []string
	Caches      []string
	WriteCaches bool
	Env         []string
	// Hosts is the file bound as /etc/hosts; Isolated writes one naming only the loopback.
	Hosts string
}

// etcNeeded is what of /etc a program in the step may read: who the users are, where the libraries are, the
// certificates, the time zone. Not resolv.conf, not hosts: the step resolves no name.
var etcNeeded = []string{"/etc/passwd", "/etc/group", "/etc/nsswitch.conf", "/etc/ld.so.cache", "/etc/ld.so.conf",
	"/etc/ld.so.conf.d", "/etc/ssl", "/etc/ca-certificates", "/etc/pki", "/etc/localtime", "/etc/os-release", "/etc/alternatives"}

// systemArgs binds the system's programs and libraries read-only: /usr, and /bin, /sbin, /lib, /lib64 as the symlinks
// they are on a merged-usr system, or bound when they are directories. A variable so a unit test fixes it.
var systemArgs = func() []string {
	args := []string{"--ro-bind", "/usr", "/usr"}
	for _, d := range []string{"/bin", "/sbin", "/lib", "/lib64"} {
		info, err := os.Lstat(d)
		switch {
		case err != nil:
		case info.Mode()&os.ModeSymlink != 0:
			if target, err := os.Readlink(d); err == nil {
				args = append(args, "--symlink", target, d)
			}
		case info.IsDir():
			args = append(args, "--ro-bind", d, d)
		}
	}
	for _, e := range etcNeeded {
		args = append(args, "--ro-bind-try", e, e)
	}
	return args
}

// stepHome is the HOME of a step that is given none: inside its own /tmp, gone with it.
const stepHome = "/tmp/home"

func checkDir(what, dir string) error {
	if dir == "" || !strings.HasPrefix(dir, "/") || path.Clean(dir) != dir || dir == "/" {
		return fmt.Errorf("buildstep: the %s %q is not an absolute, clean directory other than the root", what, dir)
	}
	return nil
}

// Command is the bubblewrap command line that runs name with args shut off as iso says, env added to iso.Env. It
// refuses a directory that is not absolute and clean, or is the root, and an environment word without "=".
func (iso Isolation) Command(name string, args ...string) ([]string, error) {
	return iso.command(nil, name, args...)
}

func (iso Isolation) command(env []string, name string, args ...string) ([]string, error) {
	if err := checkDir("workspace", iso.Workspace); err != nil {
		return nil, err
	}
	for _, group := range []struct {
		what string
		dirs []string
	}{{"writable directory", iso.Writable}, {"read-only directory", iso.ReadOnly}, {"cache", iso.Caches}} {
		for _, d := range group.dirs {
			if err := checkDir(group.what, d); err != nil {
				return nil, err
			}
		}
	}
	words := append(append([]string{}, iso.Env...), env...)
	home := false
	for _, w := range words {
		k, _, ok := strings.Cut(w, "=")
		if !ok || k == "" {
			return nil, fmt.Errorf("buildstep: the environment word %q is not NAME=value", w)
		}
		home = home || k == "HOME"
	}
	argv := []string{"--unshare-all", "--die-with-parent", "--new-session", "--clearenv",
		"--proc", "/proc", "--dev", "/dev", "--tmpfs", "/tmp", "--dir", stepHome}
	argv = append(argv, systemArgs()...)
	for _, d := range iso.ReadOnly {
		argv = append(argv, "--ro-bind", d, d)
	}
	argv = append(argv, cacheArgs(iso.Caches, iso.WriteCaches)...)
	argv = append(argv, "--bind", iso.Workspace, iso.Workspace)
	for _, d := range iso.Writable {
		argv = append(argv, "--bind", d, d)
	}
	if iso.Hosts != "" {
		argv = append(argv, "--ro-bind", iso.Hosts, "/etc/hosts")
	}
	for _, w := range words {
		k, v, _ := strings.Cut(w, "=")
		argv = append(argv, "--setenv", k, v)
	}
	if !home {
		argv = append(argv, "--setenv", "HOME", stepHome)
	}
	argv = append(argv, "--chdir", iso.Workspace, "--", name)
	return append(argv, args...), nil
}

// cacheArgs binds the shared caches: read-write for a trunk build, otherwise read-only (S18).
func cacheArgs(caches []string, write bool) []string {
	var argv []string
	for _, c := range caches {
		argv = append(argv, "--ro-bind", c, c)
	}
	_ = write
	return argv
}

// loopbackHosts is the hosts file of a step: the loopback and nothing else.
const loopbackHosts = "127.0.0.1 localhost\n::1 localhost\n"

// Isolated is run with every command shut off as iso says (Command), run's env going inside the step: the build modules'
// run steps call their test, build and digest commands through it, never run itself. Each command gets a hosts file of
// its own, naming only the loopback, removed when it ends.
func Isolated(run forgekey.Runner, iso Isolation) forgekey.Runner {
	return func(ctx context.Context, env []string, name string, args ...string) forgekey.Ran {
		dir, err := os.MkdirTemp("", "step-hosts-")
		if err != nil {
			return forgekey.Ran{Status: 1, Err: "the step's hosts file: " + err.Error()}
		}
		defer os.RemoveAll(dir)
		hosts := filepath.Join(dir, "hosts")
		if err := os.WriteFile(hosts, []byte(loopbackHosts), 0o644); err != nil {
			return forgekey.Ran{Status: 1, Err: "the step's hosts file: " + err.Error()}
		}
		iso := iso
		iso.Hosts = hosts
		argv, err := iso.command(env, name, args...)
		if err != nil {
			return forgekey.Ran{Status: 1, Err: err.Error()}
		}
		return run(ctx, nil, Bubblewrap, argv...)
	}
}