mesh-host/internal/apply/user.go

package apply

import (
	"context"
	"errors"
	"fmt"
	"os"
	osuser "os/user"
	"path/filepath"
	"strconv"
	"strings"

	"github.com/novox/mesh-host/internal/declaration"
	"github.com/novox/mesh-host/internal/store"
	"github.com/novox/mesh-host/internal/system"
)

// Logins, and the files that belong to them.
//
// Most of what a person installs is not a service. A shell, a terminal, a chat client, a desktop
// are a package plus configuration **in somebody's home** — so a mesh with no notion of a user
// can manage /etc and nothing anybody looks at.

// applyUser makes a login match what was declared.
//
// Reconciling, like everything else here: it is not told whether the user is new. Creating,
// setting a shell and adding groups are each done only when the machine does not already agree.
//
// previous is this resource's record, which carries the shell the account had before the mesh
// first changed it, so removal can give it back (novox/hq ADR 0176 §2, issue 228).
func applyUser(ctx context.Context, sys system.System, r *declaration.User, run Runner,
	previous store.Applied) (Outcome, error) {
	out := begin(r)
	out.Action = "unchanged"
	// What was found is carried from the record for as long as the resource is recorded — for this
	// account only: a declaration that renamed its user says nothing about the new one's shell.
	if previous.Shell != nil && previous.Target == r.Name {
		kept := *previous.Shell
		out.shell = &kept
	}
	if previous.Linger != nil && previous.Target == r.Name {
		kept := *previous.Linger
		out.linger = &kept
	}

	login, exists, err := system.LookUpUser(ctx, system.Runner(run), r.Name)
	if err != nil {
		return out, err
	}

	// **A shell is refused before anything is touched** (novox/hq issue 228). Refused after the
	// account was created or its groups changed, the account would be half the declaration's; a
	// refusal fails this resource and leaves the account exactly as it was.
	if r.Shell != "" && (!exists || login.Shell != r.Shell) {
		if err := system.UsableShell(r.Shell); err != nil {
			return out, fmt.Errorf("%q's shell was not set, and the account was left as it is: %w",
				r.Name, err)
		}
	}

	if !exists {
		if err := sys.CreateUser(ctx, system.Runner(run), r.Name, r.Home, r.Shell); err != nil {
			return out, err
		}
		// Read back from the machine, not from the call that made it. A useradd that returns
		// success and leaves no entry is exactly the failure this host takes trouble over.
		login, exists, err = system.LookUpUser(ctx, system.Runner(run), r.Name)
		if err != nil {
			return out, err
		}
		if !exists {
			return out, fmt.Errorf("created the user %q and the user database does not have it",
				r.Name)
		}
		out.Action = "created"
		if r.Shell != "" {
			// No shell from before to give back: the account had none until the mesh made it.
			out.shell = &store.LoginShell{Set: login.Shell, Created: true}
		}
	}

	// Groups before the shell, so that a failure here comes before the shell is changed: a record
	// is written only for an apply that worked, and a shell changed by a failed one would be read
	// next time as the account's own, and the one it replaced lost.
	if len(r.Groups) > 0 {
		in, err := system.GroupsOf(ctx, system.Runner(run), r.Name)
		if err != nil {
			return out, err
		}
		already := map[string]bool{}
		for _, g := range in {
			already[g] = true
		}
		for _, want := range r.Groups {
			if already[want] {
				continue
			}
			if err := sys.AddUserToGroup(ctx, system.Runner(run), r.Name, want); err != nil {
				return out, err
			}
			if out.Action == "unchanged" {
				out.Action = "updated"
			}
		}
	}

	// The shell, only when it differs. Absent means the host asserts nothing — a field that
	// always asserts cannot express "leave it alone", which is the difference between managing a
	// machine and taking it over.
	if r.Shell != "" && login.Shell != r.Shell {
		if err := sys.SetUserShell(ctx, system.Runner(run), r.Name, r.Shell); err != nil {
			return out, err
		}
		if back, _, err := system.LookUpUser(ctx, system.Runner(run), r.Name); err != nil {
			return out, err
		} else if back.Shell != r.Shell {
			return out, fmt.Errorf("set %q's shell to %q and the user database says %q",
				r.Name, r.Shell, back.Shell)
		}
		// **What was found is recorded once** (novox/hq ADR 0176 §2). A later change keeps it: what
		// is given back is the shell from before the mesh, never the mesh's own earlier choice.
		if out.shell == nil {
			out.shell = &store.LoginShell{Found: login.Shell}
		}
		out.shell.Set = r.Shell
		if out.Action == "unchanged" {
			out.Action = "updated"
		}
	}

	// Lingering last, and only when declared and different: the one change here that starts or
	// stops something — the account's manager and every unit in it (novox/hq ADR 0177).
	if r.Linger != nil {
		changed, err := applyLinger(ctx, sys, r.Name, *r.Linger, run, &out)
		if err != nil {
			return out, err
		}
		if changed && out.Action == "unchanged" {
			out.Action = "updated"
		}
	}
	return out, nil
}

// lingerer is a service manager that runs a manager per account, and can keep one running with
// nobody logged in (novox/hq ADR 0177).
type lingerer interface {
	Lingering(ctx context.Context, run system.Runner, name string) (bool, error)
	SetLingering(ctx context.Context, run system.Runner, name string, on bool) error
}

// applyLinger makes whether an account lingers what was declared, read back from the machine, and
// keeps what it found the first time it changed it so removal can give that back.
func applyLinger(ctx context.Context, sys system.System, name string, want bool, run Runner,
	out *Outcome) (bool, error) {
	l, ok := sys.(lingerer)
	if !ok {
		return false, fmt.Errorf("%q is declared to linger, and this machine's service manager has no "+
			"manager per account to keep running (novox/hq ADR 0177)", name)
	}
	now, err := l.Lingering(ctx, system.Runner(run), name)
	if err != nil {
		return false, err
	}
	if now == want {
		return false, nil
	}
	if err := l.SetLingering(ctx, system.Runner(run), name, want); err != nil {
		return false, err
	}
	if back, err := l.Lingering(ctx, system.Runner(run), name); err != nil {
		return false, err
	} else if back != want {
		return false, fmt.Errorf("asked logind to make %q linger %s, and it reads %s",
			name, onOff(want), onOff(back))
	}
	// Found once, as the shell's is: what goes back is what was there before the mesh.
	if out.linger == nil {
		out.linger = &store.Lingering{Found: now}
	}
	out.linger.Set = want
	return true, nil
}

func onOff(on bool) string {
	if on {
		return "on"
	}
	return "off"
}

// removeUser is what undeclaring a login does: never deleting the account, and giving back the
// shell the mesh replaced when that is still safe (novox/hq ADR 0176 §2, issue 228) — and whether
// it lingered, on the same rule (novox/hq ADR 0177).
//
// **The account is never deleted, whether or not the mesh created it.** An account owns a home,
// files, a crontab, a mailbox — what a person did with it is not the mesh's to know, and deleting
// it is the data loss ADR 0030 exists to prevent. It is the package's rule, on a login: the
// mesh no longer requires it, which is not the same as "remove it".
//
// The shell goes back only while the account still has the one the mesh set — one a person chose
// since is theirs — and only to a shell that is still usable: giving back a shell that has been
// uninstalled since would break the very logins the giving back is for. Otherwise it is left, and
// the outcome says why. Never errNoRemoval: an orphaned login that failed removal stopped the
// whole apply, on every apply after.
func removeUser(ctx context.Context, sys system.System, a store.Applied, run Runner) (string, string, error) {
	const kept = "the account is kept; the host never deletes a login"
	login, exists, err := system.LookUpUser(ctx, system.Runner(run), a.Target)
	if err != nil {
		return "", "", err
	}
	if !exists {
		return "forgotten", "no longer there", nil
	}
	action, detail, err := giveShellBack(ctx, sys, a, login, run, kept)
	if err != nil {
		return "", "", err
	}
	if gave, said := giveLingerBack(ctx, sys, a, run); said != "" {
		detail += "; " + said
		if gave {
			action = "restored"
		}
	}
	return action, detail, nil
}

// giveShellBack is removeUser's shell: given back only while the account still has the one the
// mesh set, and only to one still usable.
func giveShellBack(ctx context.Context, sys system.System, a store.Applied, login system.Login,
	run Runner, kept string) (string, string, error) {
	found := a.Shell
	switch {
	case found == nil:
		return "forgotten", kept + ", and its shell was never changed by the mesh", nil
	case found.Created:
		return "forgotten", kept + "; the mesh created it, so there is no shell from before to give back", nil
	case login.Shell != found.Set:
		return "forgotten", fmt.Sprintf("%s, and its shell %s left as it is: changed since the mesh set %s",
			kept, login.Shell, found.Set), nil
	case found.Found == "":
		return "forgotten", kept + ", and its shell left as it is: it had none before the mesh set one", nil
	}
	if err := system.UsableShell(found.Found); err != nil {
		return "forgotten", fmt.Sprintf("%s, and its shell %s left as it is: the one it had before "+
			"cannot be given back: %v", kept, login.Shell, err), nil
	}
	// A give-back that fails is said and not fatal: fatal, the record would stay and fail the same
	// way on every apply after — the very wedge this removal exists to end.
	if err := sys.SetUserShell(ctx, system.Runner(run), a.Target, found.Found); err != nil {
		return "forgotten", fmt.Sprintf("%s, and the shell it had before the mesh, %s, could not be "+
			"given back: %v", kept, found.Found, err), nil
	}
	if back, _, err := system.LookUpUser(ctx, system.Runner(run), a.Target); err != nil {
		return "", "", err
	} else if back.Shell != found.Found {
		return "forgotten", fmt.Sprintf("%s; gave back the shell %s and the user database says %s",
			kept, found.Found, back.Shell), nil
	}
	return "restored", fmt.Sprintf("%s; the shell it had before the mesh, %s, given back", kept, found.Found), nil
}

// giveLingerBack is removeUser's lingering (novox/hq ADR 0177), on the shell's rule: whether the
// account lingered before the mesh changed it goes back, and only while the account still has what
// the mesh set — one the operator changed since with loginctl is theirs. Never fatal, for the
// shell's reason. Empty when the mesh never changed it, so a removal says nothing about it.
//
// Giving back "not lingering" stops the account's manager if nobody is logged in, and every unit
// in it: that is what the account had before the mesh, and its user units go with its declaration.
func giveLingerBack(ctx context.Context, sys system.System, a store.Applied, run Runner) (bool, string) {
	found := a.Linger
	if found == nil {
		return false, ""
	}
	if found.Found == found.Set {
		return false, ""
	}
	l, ok := sys.(lingerer)
	if !ok {
		return false, ""
	}
	now, err := l.Lingering(ctx, system.Runner(run), a.Target)
	if err != nil {
		return false, fmt.Sprintf("whether it lingered before the mesh could not be given back: %v", err)
	}
	if now != found.Set {
		return false, fmt.Sprintf("lingering left %s: changed since the mesh set it %s", onOff(now), onOff(found.Set))
	}
	if err := l.SetLingering(ctx, system.Runner(run), a.Target, found.Found); err != nil {
		return false, fmt.Sprintf("lingering, %s before the mesh, could not be given back: %v", onOff(found.Found), err)
	}
	if back, err := l.Lingering(ctx, system.Runner(run), a.Target); err != nil || back != found.Found {
		return false, fmt.Sprintf("gave back lingering %s and logind does not read it so", onOff(found.Found))
	}
	return true, fmt.Sprintf("lingering %s again, as before the mesh", onOff(found.Found))
}

// own sets a path's owner, when one was declared.
//
// Looked up by name every time rather than cached: a user's numeric id is not stable across
// machines, and the whole reason this exists is that the same declaration lands on several.
func own(path, owner string) error {
	if owner == "" {
		return nil
	}
	uid, gid, err := idsOf(owner)
	if err != nil {
		return fmt.Errorf("%s should belong to %q: %w", path, owner, err)
	}
	if err := os.Chown(path, uid, gid); err != nil {
		return fmt.Errorf("cannot give %s to %q: %w", path, owner, err)
	}
	return nil
}

// idsOf resolves an owner to a uid and gid: a name this machine knows, or numbers it does not.
//
// **Numbers, because a container's user is a number the machine has never heard of.** A directory
// a module mounts into its container belongs to whoever runs inside — grafana's 472, redis's 999,
// www-data's 33 — and none of those has a row in this machine's passwd, so there is no name to
// look up and none to create. Refusing them looked principled and meant every module whose
// container drops privileges could not own its own data: the store's config was unreadable to
// the store, and the forge could not traverse into the directory that held its files.
//
// "uid:gid" and bare "uid" are numeric; anything else is a name, resolved as before.
func idsOf(owner string) (int, int, error) {
	user, group, both := strings.Cut(owner, ":")
	if uid, err := strconv.Atoi(user); err == nil {
		gid := uid
		if both {
			g, err := strconv.Atoi(group)
			if err != nil {
				return 0, 0, fmt.Errorf(
					"%q reads as a uid with a group that is not a gid", owner)
			}
			gid = g
		}
		return uid, gid, nil
	}
	if both {
		return 0, 0, fmt.Errorf("%q mixes a name with a colon; a name stands alone", owner)
	}
	found, err := osuser.Lookup(owner)
	if err != nil {
		return 0, 0, fmt.Errorf("this machine has no such user: %w", err)
	}
	uid, err := strconv.Atoi(found.Uid)
	if err != nil {
		return 0, 0, err
	}
	gid, err := strconv.Atoi(found.Gid)
	if err != nil {
		return 0, 0, err
	}
	return uid, gid, nil
}

// ownedBy reports whether a path already belongs to a user, so applying twice changes nothing.
func ownedBy(path, owner string) (bool, error) {
	if owner == "" {
		return true, nil
	}
	wantUID, wantGID, err := idsOf(owner)
	if err != nil {
		return false, nil
	}
	info, err := os.Stat(path)
	if err != nil {
		return false, err
	}
	uid, gid, ok := ownerOf(info)
	if !ok {
		return false, nil
	}
	return uid == wantUID && gid == wantGID, nil
}

// makeDirs makes a directory and any parent of it that is missing, as MkdirAll does — and gives
// each one it made inside the owner's home to the owner (novox/hq ADR 0182, to-be 41).
//
// **A parent made as root inside a home is a home the person cannot use.** A module writing
// ~/.config/mesh/environment.sh, or unpacking into ~/.local/share/powerlevel10k, on a fresh account
// made ~/.config and ~/.local/share owned by root: the file was the person's, the directory every
// program of theirs writes into was not. So what the host creates between the home and the target
// is the owner's, as the target is.
//
// **Only what the host created.** A parent that was already there is never chowned or chmodded:
// what a person or another program made is held as found (ADR 0182). And only inside the owner's
// home, read from the user database, not guessed from a prefix on /home: a module's directory under
// /var/lib is made exactly as before, whoever its files belong to.
func makeDirs(dir string, mode os.FileMode, owner string) error {
	_, err := makeDirsSaying(dir, mode, owner)
	return err
}

// makeDirsSaying is makeDirs, and says which directories it made, deepest first — so an archive
// can take away on removal the parents it made to reach its directory (novox/hq issue 162).
func makeDirsSaying(dir string, mode os.FileMode, owner string) ([]string, error) {
	var made []string
	for d := filepath.Clean(dir); ; d = filepath.Dir(d) {
		if _, err := os.Lstat(d); !errors.Is(err, os.ErrNotExist) {
			break
		}
		made = append(made, d)
		if filepath.Dir(d) == d {
			break
		}
	}
	if err := os.MkdirAll(dir, mode); err != nil {
		return nil, err
	}
	if owner == "" || len(made) == 0 {
		return made, nil
	}
	home, err := homeOf(owner)
	if err != nil || home == "" {
		// A numeric owner — a container's user — has no home, and a name the machine does not
		// know fails where the target is given to it. Either way nothing here is a home's.
		return made, nil
	}
	home = filepath.Clean(home)
	for _, d := range made {
		if d != home && !strings.HasPrefix(d, home+string(os.PathSeparator)) {
			continue
		}
		if err := ownMade(d, owner); err != nil {
			return made, err
		}
	}
	return made, nil
}

// homeOf is an owner's home from the user database, and ownMade gives a directory the host made to
// its owner. Variables so a test can give an owner a home it owns, and see what was given to whom
// without being root.
var (
	homeOf = func(owner string) (string, error) {
		found, err := osuser.Lookup(owner)
		if err != nil {
			return "", err
		}
		return found.HomeDir, nil
	}
	ownMade = own
)

// ownAll gives a whole tree to a user, for an archive that was unpacked into it.
func ownAll(root, owner string) error {
	if owner == "" {
		return nil
	}
	return filepath.Walk(root, func(path string, _ os.FileInfo, err error) error {
		if err != nil {
			return err
		}
		return own(path, owner)
	})
}

// ownerOf is the numeric owner of a file, where the platform reports one.
func ownerOf(info os.FileInfo) (uid, gid int, ok bool) {
	return statOwner(info)
}