fix(notes): the edit lock excludes other processes; pin OpenLock and the rollback (#192)
Review round 1 on PR #194: - The edit lock did not exclude other processes (also on main): unlock removes the lock file, so a waiter could wake holding a flock on the orphaned inode while the next caller locked a fresh file. flockFile now re-checks after flock that the locked file is still the one at the path (brainfs LockIsCurrent) and retries otherwise; OpenLock reports a lock replaced mid-open as ErrLockChanged, which is retried instead of failing the edit. New cross-process test (3 re-executed processes x 400 rounds, O_EXCL marker under the lock): 0 overlaps; red without the re-check. - OpenLock unit test: relative dangling link (only the Lstat pre-check stops it; os.Root refuses absolute links on its own), live link, FIFO, directory, behind a symlinked dir, unclean path, positive control. - Rollback test: a symlink swapped in for the note before the rollback (test seam beforeStampHook) is refused, target untouched in and outside the brain. - The in-brain dangling-lock case uses a relative link and asserts the refusal names the lock. OpenLock reuses #189's RealParent; RemoveLock is dropped for brainfs.Remove. The doc comment names the swap residual. Rebased onto main after #189. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
@@ -0,0 +1,82 @@
|
||||
package notes
|
||||
|
||||
import (
|
||||
"errors"
|
||||
"fmt"
|
||||
"os"
|
||||
"os/exec"
|
||||
"path/filepath"
|
||||
"strconv"
|
||||
"strings"
|
||||
"testing"
|
||||
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
// lockChild runs inside a re-executed test binary: it takes the note's edit
|
||||
// lock N times and, while holding it, creates an O_EXCL marker. Finding the
|
||||
// marker already there means another process holds the "same" lock: mutual
|
||||
// exclusion failed. It prints the overlap count.
|
||||
func lockChild(brainDir string, rounds int) {
|
||||
t, err := Resolve(brainDir, "wiki/w/facts/n.md", ScopeNotes)
|
||||
if err != nil {
|
||||
fmt.Println("resolve:", err)
|
||||
os.Exit(2)
|
||||
}
|
||||
marker := filepath.Join(brainDir, "marker")
|
||||
overlaps := 0
|
||||
for i := 0; i < rounds; i++ {
|
||||
unlock, err := flockFile(t)
|
||||
if err != nil {
|
||||
fmt.Println("lock:", err)
|
||||
os.Exit(2)
|
||||
}
|
||||
f, err := os.OpenFile(marker, os.O_CREATE|os.O_EXCL|os.O_WRONLY, 0o644)
|
||||
if errors.Is(err, os.ErrExist) {
|
||||
overlaps++
|
||||
} else if err == nil {
|
||||
_ = f.Close()
|
||||
_ = os.Remove(marker)
|
||||
}
|
||||
unlock()
|
||||
}
|
||||
fmt.Println(overlaps)
|
||||
os.Exit(0)
|
||||
}
|
||||
|
||||
func TestMain(m *testing.M) {
|
||||
if dir := os.Getenv("NOTES_LOCK_CHILD"); dir != "" {
|
||||
n, _ := strconv.Atoi(os.Getenv("NOTES_LOCK_ROUNDS"))
|
||||
lockChild(dir, n)
|
||||
}
|
||||
os.Exit(m.Run())
|
||||
}
|
||||
|
||||
// hyperguild#192: the edit lock must exclude other processes (two ingestion
|
||||
// pods share the hostPath brain during a RollingUpdate). Unlock removes the
|
||||
// lock file, so without re-checking that the locked file is still the one at
|
||||
// the path, a waiter and the next caller each held a lock (measured on main:
|
||||
// hundreds of overlaps in 2000 rounds).
|
||||
func TestEditLockExcludesOtherProcesses(t *testing.T) {
|
||||
brain := t.TempDir()
|
||||
require.NoError(t, os.MkdirAll(filepath.Join(brain, "wiki", "w", "facts"), 0o755))
|
||||
require.NoError(t, os.WriteFile(filepath.Join(brain, "wiki", "w", "facts", "n.md"), []byte("x"), 0o644))
|
||||
const procs, rounds = 3, 400
|
||||
cmds := make([]*exec.Cmd, procs)
|
||||
outs := make([]*strings.Builder, procs)
|
||||
for i := range cmds {
|
||||
outs[i] = &strings.Builder{}
|
||||
cmds[i] = exec.Command(os.Args[0], "-test.run=^$")
|
||||
cmds[i].Env = append(os.Environ(), "NOTES_LOCK_CHILD="+brain, "NOTES_LOCK_ROUNDS="+strconv.Itoa(rounds))
|
||||
cmds[i].Stdout, cmds[i].Stderr = outs[i], outs[i]
|
||||
require.NoError(t, cmds[i].Start())
|
||||
}
|
||||
total := 0
|
||||
for i, c := range cmds {
|
||||
require.NoError(t, c.Wait(), "child %d: %s", i, outs[i].String())
|
||||
n, err := strconv.Atoi(strings.TrimSpace(outs[i].String()))
|
||||
require.NoError(t, err, "child %d output: %q", i, outs[i].String())
|
||||
total += n
|
||||
}
|
||||
require.Zero(t, total, "processes held the edit lock at the same time")
|
||||
}
|
||||
Reference in New Issue
Block a user