Compare commits

..
Author SHA1 Message Date
mathiasandClaude Opus 5.5 22d0653b61 test(watcher): a marker link planted during the run is refused and not retried (#158)
CI / Lint / Test / Vet (pull_request) Successful in 1m3s
CI / Mirror to GitHub (pull_request) Skipped
Review round 2 on PR #195: the marker writes had no test, so reverting
them to by-path writes, or the refusal to log-only, stayed green. The
link is planted from inside the fake model call, between the walk's
Lstat and the marker write, for both markers and both an outside and
an in-brain relative target; red on all four flips. The marker-entry
skip is now checked for .failed too.

Also pins cleanRel's acceptance of a backslash as a filename byte in
brainfs itself, and stops the buildAndWrite control claiming writes it
does not count.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-05 23:47:24 +02:00
mathiasandClaude Opus 5.5 1a2a93d2f4 fix(watcher): refusals end in a terminal state; pin every converted writer (#158)
CI / Lint / Test / Vet (pull_request) Successful in 1m6s
CI / Mirror to GitHub (pull_request) Skipped
Review round 1 on PR #195:
- A refused marker or archive write left no marker, so the note was
  reprocessed every interval (archive copies piling up; an LLM call per
  round on the failure path). The marker check now uses Lstat (any entry
  at the marker path counts, so a dangling marker link skips the note
  instead of looping), and a note whose archive copy or marker write is
  refused is remembered (path, size, mtime) and not retried until it
  changes.
- cleanRel no longer refuses a backslash: it is a legal filename byte on
  Linux, not a traversal, and a raw note named with one was never
  archived or marked.
- Tests now reach every converted writer: the log.md append (real
  raw/failed so the log is reached, symlinked log.md), a dangling marker
  link, the refused-retry loop with a positive control, a backslash name,
  a buildAndWrite swap race, and the claude-session archive with in-brain
  and outside symlinked archive dirs. The stale lint allowlist entry for
  appendWatcherLog is removed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-05 23:40:19 +02:00
mathiasandClaude Opus 5.5 77c44f298c fix: pipeline, backfill, watcher and archive writes go through the brain root (#158 part B)
CI / Lint / Test / Vet (pull_request) Successful in 1m13s
CI / Mirror to GitHub (pull_request) Skipped
The remaining writers created or rewrote brain files by path:
- pipeline buildAndWrite: os.MkdirAll + os.WriteFile of every page.
- BackfillRefs: os.WriteFile of each concept page; a page swapped for a
  symlink after the read was written through (the race test overwrote an
  in-brain target before this change).
- watcher: the raw/processed and raw/failed archive copies (os.Create),
  the .processed/.failed markers and the log.md entry (os.OpenFile).
- cmd/server: the claude-session archive.

All now go through brainfs PutRegular/AppendRegular (new package-level
PutFile/AppendFile for one-off writers: the watcher's call-site lint bans
.Open calls in that file so raw/ reads stay on OpenRegular).

Stacked on #189 (needs PutRegular/AppendRegular).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-05 23:26:02 +02:00
19 changed files with 443 additions and 529 deletions
+15 -13
View File
@@ -17,6 +17,7 @@ import (
"github.com/mathiasbq/hyperguild/ingestion/internal/api" "github.com/mathiasbq/hyperguild/ingestion/internal/api"
"github.com/mathiasbq/hyperguild/ingestion/internal/audit" "github.com/mathiasbq/hyperguild/ingestion/internal/audit"
"github.com/mathiasbq/hyperguild/ingestion/internal/brainfs"
"github.com/mathiasbq/hyperguild/ingestion/internal/capture" "github.com/mathiasbq/hyperguild/ingestion/internal/capture"
"github.com/mathiasbq/hyperguild/ingestion/internal/capturehttp" "github.com/mathiasbq/hyperguild/ingestion/internal/capturehttp"
"github.com/mathiasbq/hyperguild/ingestion/internal/classification" "github.com/mathiasbq/hyperguild/ingestion/internal/classification"
@@ -89,11 +90,15 @@ func (s *claudeSink) Ingest(ctx context.Context, b claudewatcher.Batch) error {
sb.WriteString("\n\n") sb.WriteString("\n\n")
} }
slug := "session-" + b.Host + "-" + b.SessionID slug := "session-" + b.Host + "-" + b.SessionID
dest := filepath.Join(s.brainDir, "archive", "claude-sessions", b.Host, slug+".md") dest := "archive/claude-sessions/" + b.Host + "/" + slug + ".md"
if err := os.MkdirAll(filepath.Dir(dest), 0o755); err != nil { // Written through the brain root, never through a symlinked directory
return fmt.Errorf("create claude-sessions archive dir: %w", err) // or file (hyperguild#158).
bfs, err := brainfs.Open(s.brainDir)
if err != nil {
return fmt.Errorf("open brain: %w", err)
} }
if err := os.WriteFile(dest, []byte(sb.String()), 0o644); err != nil { defer func() { _ = bfs.Close() }()
if err := bfs.PutRegular(dest, []byte(sb.String())); err != nil {
return fmt.Errorf("write claude session archive: %w", err) return fmt.Errorf("write claude session archive: %w", err)
} }
s.logger.Debug("claude session archived (non-indexed)", "path", dest) s.logger.Debug("claude session archived (non-indexed)", "path", dest)
@@ -179,12 +184,9 @@ func alertingWarnings(getenv func(string) string) []string {
return out return out
} }
// buildNotifier returns the ntfy notifier for one alert source, titled so func buildNotifier() audit.Notifier {
// the source is recognisable (hyperguild#163), or nil when ntfy is not if n := audit.NewNtfyNotifier(os.Getenv("BRAIN_NTFY_URL"), os.Getenv("BRAIN_NTFY_TOKEN")); n != nil {
// configured. return n
func buildNotifier(getenv func(string) string, title string) audit.Notifier {
if n := audit.NewNtfyNotifier(getenv("BRAIN_NTFY_URL"), getenv("BRAIN_NTFY_TOKEN")); n != nil {
return n.WithTitle(title)
} }
return nil return nil
} }
@@ -206,7 +208,7 @@ func buildAuditSink(ctx context.Context, brainDir string, logger *slog.Logger) c
logger.Error("capture audit buffer init", "err", err) logger.Error("capture audit buffer init", "err", err)
os.Exit(1) os.Exit(1)
} }
notifier := buildNotifier(os.Getenv, "brain-capture audit degraded") notifier := buildNotifier()
reconcileInterval := time.Duration(envInt("BRAIN_AUDIT_RECONCILE_INTERVAL", 60)) * time.Second reconcileInterval := time.Duration(envInt("BRAIN_AUDIT_RECONCILE_INTERVAL", 60)) * time.Second
audit.StartReconcile(ctx, central, buffer, notifier, reconcileInterval) audit.StartReconcile(ctx, central, buffer, notifier, reconcileInterval)
logger.Info("capture audit: loki+buffer sink", "loki", lokiURL, "reconcile_s", int(reconcileInterval.Seconds())) logger.Info("capture audit: loki+buffer sink", "loki", lokiURL, "reconcile_s", int(reconcileInterval.Seconds()))
@@ -369,7 +371,7 @@ func main() {
ctx := context.Background() ctx := context.Background()
// hyperguild#157: report a symlink that appears in the brain while running. // hyperguild#157: report a symlink that appears in the brain while running.
go symlinkguard.Watch(ctx, brainDir, symlinkScanInterval(os.Getenv), buildNotifier(os.Getenv, "brain symlink tripwire")) go symlinkguard.Watch(ctx, brainDir, symlinkScanInterval(os.Getenv), buildNotifier())
// Never fatal on an unreachable issuer (#83): on a cold cluster boot this // Never fatal on an unreachable issuer (#83): on a cold cluster boot this
// pod can beat dex to serving and get a 502 from ingress, and exiting meant // pod can beat dex to serving and get a 502 from ingress, and exiting meant
@@ -549,7 +551,7 @@ func main() {
// capture already gets, plus an out-of-band alert. Wired here because // capture already gets, plus an out-of-band alert. Wired here because
// brain_edit_root needs the capture plumbing anyway — that is what // brain_edit_root needs the capture plumbing anyway — that is what
// re-derives the static principal its gate checks. // re-derives the static principal its gate checks.
mcpSrv.WithGovernanceObserver(auditSink, buildNotifier(os.Getenv, "brain governance edit")) mcpSrv.WithGovernanceObserver(auditSink, buildNotifier())
logger.Info("capture enabled (REST + MCP tool)", "sovereign_principals", len(sovereign), logger.Info("capture enabled (REST + MCP tool)", "sovereign_principals", len(sovereign),
"governance_observer", true) "governance_observer", true)
} else { } else {
+24
View File
@@ -76,3 +76,27 @@ func TestAlertingWarnings(t *testing.T) {
}) })
} }
} }
// hyperguild#158: the claude-session archive was written with os.MkdirAll +
// os.WriteFile by path, so a symlinked archive directory redirected it.
func TestClaudeSinkDoesNotWriteThroughASymlinkedArchiveDir(t *testing.T) {
for name, mk := range map[string]func(t *testing.T, dir string) (link, target string){
"outside": func(t *testing.T, _ string) (string, string) { o := t.TempDir(); return o, o },
"in-brain": func(t *testing.T, dir string) (string, string) {
require.NoError(t, os.MkdirAll(filepath.Join(dir, "raw"), 0o755))
return filepath.Join("..", "raw"), filepath.Join(dir, "raw")
},
} {
t.Run(name, func(t *testing.T) {
dir := t.TempDir()
link, target := mk(t, dir)
require.NoError(t, os.Symlink(link, filepath.Join(dir, "archive")))
sink := &claudeSink{brainDir: dir, logger: slog.New(slog.NewTextHandler(os.Stderr, nil))}
err := sink.Ingest(context.Background(), claudewatcher.Batch{Host: "koala", SessionID: "abc",
Turns: []claudewatcher.Turn{{Type: "assistant", Content: "x"}}})
assert.Error(t, err)
ents, _ := os.ReadDir(target)
assert.Empty(t, ents, "the archive was written at the link's target")
})
}
}
@@ -1,9 +1,6 @@
package main package main
import ( import (
"context"
"net/http"
"net/http/httptest"
"os" "os"
"path/filepath" "path/filepath"
"strings" "strings"
@@ -77,16 +74,3 @@ func TestStartupSymlinkCheckRefusesABrainWithAFIFO(t *testing.T) {
require.Error(t, err) require.Error(t, err)
assert.Contains(t, err.Error(), "classification.yaml") assert.Contains(t, err.Error(), "classification.yaml")
} }
// hyperguild#163: each alert source gets its own ntfy title.
func TestBuildNotifierUsesTheGivenTitle(t *testing.T) {
var got string
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { got = r.Header.Get("Title") }))
defer srv.Close()
env := map[string]string{"BRAIN_NTFY_URL": srv.URL}
n := buildNotifier(func(k string) string { return env[k] }, "brain symlink tripwire")
require.NotNil(t, n)
require.NoError(t, n.Notify(context.Background(), "m"))
assert.Equal(t, "brain symlink tripwire", got)
assert.Nil(t, buildNotifier(func(string) string { return "" }, "x"), "no ntfy URL, no notifier")
}
+1 -20
View File
@@ -13,14 +13,9 @@ import (
type NtfyNotifier struct { type NtfyNotifier struct {
topicURL string topicURL string
token string token string
title string
http *http.Client http *http.Client
} }
// defaultNtfyTitle is the title of a notifier built without WithTitle: the
// capture audit's degraded-state alert, its first user.
const defaultNtfyTitle = "brain-capture audit degraded"
// NewNtfyNotifier constructs a notifier for the given ntfy topic URL // NewNtfyNotifier constructs a notifier for the given ntfy topic URL
// (e.g. https://ntfy.sh/my-topic). token is an optional bearer for // (e.g. https://ntfy.sh/my-topic). token is an optional bearer for
// protected ntfy instances; it is held here and only sent in the // protected ntfy instances; it is held here and only sent in the
@@ -32,31 +27,17 @@ func NewNtfyNotifier(topicURL, token string) *NtfyNotifier {
return &NtfyNotifier{ return &NtfyNotifier{
topicURL: strings.TrimRight(topicURL, "/"), topicURL: strings.TrimRight(topicURL, "/"),
token: token, token: token,
title: defaultNtfyTitle,
http: &http.Client{Timeout: 10 * time.Second}, http: &http.Client{Timeout: 10 * time.Second},
} }
} }
// WithTitle returns a copy that sends title as the ntfy Title header, so each
// alert source (the capture audit, the symlink tripwire, governance edits)
// is recognisable on the phone (hyperguild#163). A nil notifier (ntfy not
// configured) stays nil.
func (n *NtfyNotifier) WithTitle(title string) *NtfyNotifier {
if n == nil {
return nil
}
c := *n
c.title = title
return &c
}
// Notify posts a message to the ntfy topic. // Notify posts a message to the ntfy topic.
func (n *NtfyNotifier) Notify(ctx context.Context, msg string) error { func (n *NtfyNotifier) Notify(ctx context.Context, msg string) error {
req, err := http.NewRequestWithContext(ctx, http.MethodPost, n.topicURL, strings.NewReader(msg)) req, err := http.NewRequestWithContext(ctx, http.MethodPost, n.topicURL, strings.NewReader(msg))
if err != nil { if err != nil {
return err return err
} }
req.Header.Set("Title", n.title) req.Header.Set("Title", "brain-capture audit degraded")
req.Header.Set("Priority", "high") req.Header.Set("Priority", "high")
req.Header.Set("Tags", "warning,brain") req.Header.Set("Tags", "warning,brain")
if n.token != "" { if n.token != "" {
@@ -1,29 +0,0 @@
package audit
import (
"context"
"net/http"
"net/http/httptest"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// hyperguild#163: every alert went out titled "brain-capture audit
// degraded", including the symlink tripwire's and governance alerts.
func TestNtfyNotifierSendsItsOwnTitle(t *testing.T) {
var got []string
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
got = append(got, r.Header.Get("Title"))
}))
defer srv.Close()
base := NewNtfyNotifier(srv.URL, "")
require.NoError(t, base.Notify(context.Background(), "m"))
require.NoError(t, base.WithTitle("brain symlink tripwire").Notify(context.Background(), "m"))
require.NoError(t, base.Notify(context.Background(), "m"))
assert.Equal(t, []string{"brain-capture audit degraded", "brain symlink tripwire", "brain-capture audit degraded"}, got,
"WithTitle returns a copy; the base keeps its default title")
var nilN *NtfyNotifier
assert.Nil(t, nilN.WithTitle("x"), "nil (ntfy not configured) stays nil")
}
+22 -53
View File
@@ -22,10 +22,6 @@ import (
// ErrNotRegular is returned for an entry that is not a regular file. // ErrNotRegular is returned for an entry that is not a regular file.
var ErrNotRegular = errors.New("not a regular file") var ErrNotRegular = errors.New("not a regular file")
// ErrLockChanged is returned by OpenLock when the lock file was replaced
// while it was being opened.
var ErrLockChanged = errors.New("lock file replaced while opening")
// ErrTooLarge is returned by ReadRegularMax for a file over the limit. // ErrTooLarge is returned by ReadRegularMax for a file over the limit.
var ErrTooLarge = errors.New("file too large") var ErrTooLarge = errors.New("file too large")
@@ -113,7 +109,7 @@ func (b *Brain) ReadDirRegular(relDir string) ([]string, error) {
// after following a symlink, and a ".." after a symlinked directory would // after following a symlink, and a ".." after a symlinked directory would
// otherwise pass the check and redirect the read or write. // otherwise pass the check and redirect the read or write.
func cleanRel(rel string) error { func cleanRel(rel string) error {
if rel == "" || path.Clean(rel) != rel || !filepath.IsLocal(rel) || strings.Contains(rel, "\\") { if rel == "" || path.Clean(rel) != rel || !filepath.IsLocal(rel) {
return fmt.Errorf("%q: not a clean brain-relative path: %w", rel, ErrNotRegular) return fmt.Errorf("%q: not a clean brain-relative path: %w", rel, ErrNotRegular)
} }
return nil return nil
@@ -247,6 +243,27 @@ func (b *Brain) RealParent(rel string) error {
return b.realDir(path.Dir(rel)) return b.realDir(path.Dir(rel))
} }
// PutFile opens the brain dir as a root and calls PutRegular, for one-off
// writers that hold no Brain.
func PutFile(brainDir, rel string, data []byte) error {
b, err := Open(brainDir)
if err != nil {
return err
}
defer func() { _ = b.Close() }()
return b.PutRegular(rel, data)
}
// AppendFile opens the brain dir as a root and calls AppendRegular.
func AppendFile(brainDir, rel string, data []byte) error {
b, err := Open(brainDir)
if err != nil {
return err
}
defer func() { _ = b.Close() }()
return b.AppendRegular(rel, data)
}
// mkdirReal makes sure every component of rel exists as a real directory, // mkdirReal makes sure every component of rel exists as a real directory,
// creating missing ones through the root; an existing component that is // creating missing ones through the root; an existing component that is
// not a real directory is ErrNotRegular. // not a real directory is ErrNotRegular.
@@ -274,54 +291,6 @@ func (b *Brain) mkdirReal(rel string) error {
return nil return nil
} }
// OpenLock opens (creating if absent) a regular lock file at a clean
// brain-relative path for flock, through the root (hyperguild#192): every
// directory above it must be real, an existing lock must be a regular file,
// and the opened handle must be the file at the path (os.Root follows an
// in-root final symlink even with O_NOFOLLOW, so the Lstat pre-check is
// what refuses a dangling in-brain link). A link swapped in between the
// pre-check and the open can still have an empty file created at an in-brain
// target before the refusal: the #167 residual. ErrLockChanged reports a
// lock replaced while opening (a peer removed and recreated it); callers
// retry.
func (b *Brain) OpenLock(rel string) (*os.File, error) {
if err := b.RealParent(rel); err != nil {
return nil, err
}
if fi, err := b.root.Lstat(rel); err == nil && !fi.Mode().IsRegular() {
return nil, fmt.Errorf("%s: %w", rel, ErrNotRegular)
} else if err != nil && !errors.Is(err, fs.ErrNotExist) {
return nil, err
}
f, err := b.root.OpenFile(rel, os.O_RDWR|os.O_CREATE|syscall.O_NOFOLLOW|syscall.O_NONBLOCK, 0o644)
if err != nil {
return nil, err
}
hfi, herr := f.Stat()
lfi, lerr := b.root.Lstat(rel)
if herr != nil || !hfi.Mode().IsRegular() {
_ = f.Close()
return nil, fmt.Errorf("%s: %w", rel, ErrNotRegular)
}
if lerr != nil || !os.SameFile(lfi, hfi) {
_ = f.Close()
if lerr == nil && !lfi.Mode().IsRegular() {
return nil, fmt.Errorf("%s: %w", rel, ErrNotRegular)
}
return nil, fmt.Errorf("%s: %w", rel, ErrLockChanged)
}
return f, nil
}
// LockIsCurrent reports whether f is still the file at rel: after taking a
// flock, a lock file removed and recreated by a peer means the flock is on
// an orphaned inode and does not exclude anyone (hyperguild#192).
func (b *Brain) LockIsCurrent(rel string, f *os.File) bool {
hfi, herr := f.Stat()
lfi, lerr := b.root.Lstat(rel)
return herr == nil && lerr == nil && os.SameFile(lfi, hfi)
}
// ReadRegular reads a brain-relative path, refusing anything that is not a // ReadRegular reads a brain-relative path, refusing anything that is not a
// regular file, both before the open and on the opened handle. // regular file, both before the open and on the opened handle.
func (b *Brain) ReadRegular(rel string) ([]byte, error) { func (b *Brain) ReadRegular(rel string) ([]byte, error) {
+11 -40
View File
@@ -530,48 +530,19 @@ func TestCreateRegularCreatesOnlyNewFilesThroughRealDirectories(t *testing.T) {
assert.Equal(t, "one", string(got)) assert.Equal(t, "one", string(got))
} }
// hyperguild#192: OpenLock creates or reuses a regular lock file through the // On Linux a backslash is an ordinary filename byte, not a separator, so a
// root and refuses every other shape without creating anything at a link's // name containing one stays a single in-root component (hyperguild#158: the
// target or hanging. // watcher retried such a note forever while cleanRel refused it).
func TestOpenLockOpensOnlyARegularLockThroughRealDirectories(t *testing.T) { func TestReadAndPutRegularTreatABackslashAsAFilenameByte(t *testing.T) {
dir := t.TempDir() dir := t.TempDir()
require.NoError(t, os.MkdirAll(filepath.Join(dir, "raw"), 0o755))
require.NoError(t, os.MkdirAll(filepath.Join(dir, "wiki", "w"), 0o755))
require.NoError(t, os.WriteFile(filepath.Join(dir, "raw", "v.md"), []byte("untouched"), 0o644))
require.NoError(t, os.Symlink(filepath.Join("..", "..", "raw", "created"), filepath.Join(dir, "wiki", "w", "dangling.lock")))
require.NoError(t, os.Symlink(filepath.Join("..", "..", "raw", "v.md"), filepath.Join(dir, "wiki", "w", "live.lock")))
require.NoError(t, syscall.Mkfifo(filepath.Join(dir, "wiki", "w", "fifo.lock"), 0o644))
require.NoError(t, os.MkdirAll(filepath.Join(dir, "wiki", "w", "dir.lock"), 0o755))
require.NoError(t, os.Symlink(filepath.Join("..", "raw"), filepath.Join(dir, "wiki", "alias")))
b, err := Open(dir) b, err := Open(dir)
require.NoError(t, err) require.NoError(t, err)
defer func() { _ = b.Close() }() defer func() { _ = b.Close() }()
require.NoError(t, b.PutRegular(`a\..\b.md`, []byte("x")))
for _, rel := range []string{"wiki/w/dangling.lock", "wiki/w/live.lock", "wiki/w/fifo.lock", "wiki/w/dir.lock", "wiki/alias/x.lock", "wiki/w/../w/x.lock"} { got, err := os.ReadFile(filepath.Join(dir, `a\..\b.md`))
done := make(chan error, 1) require.NoError(t, err, "the name was not kept as one component")
go func() { assert.Equal(t, "x", string(got))
f, err := b.OpenLock(rel) got, err = b.ReadRegular(`a\..\b.md`)
if f != nil { require.NoError(t, err)
_ = f.Close() assert.Equal(t, "x", string(got))
}
done <- err
}()
select {
case err := <-done:
assert.ErrorIs(t, err, ErrNotRegular, rel)
case <-time.After(3 * time.Second):
t.Fatalf("OpenLock hung on %s", rel)
}
}
_, err = os.Lstat(filepath.Join(dir, "raw", "created"))
assert.True(t, os.IsNotExist(err), "nothing created at the dangling link's target")
got, _ := os.ReadFile(filepath.Join(dir, "raw", "v.md"))
assert.Equal(t, "untouched", string(got))
f, err := b.OpenLock("wiki/w/n.md.lock")
require.NoError(t, err, "positive control: creates a lock")
assert.True(t, b.LockIsCurrent("wiki/w/n.md.lock", f))
require.NoError(t, b.Remove("wiki/w/n.md.lock"))
assert.False(t, b.LockIsCurrent("wiki/w/n.md.lock", f), "a removed lock is no longer current")
_ = f.Close()
} }
@@ -88,7 +88,6 @@ func TestNoPackageReferencesExtractTextOutsideExtract(t *testing.T) {
// osOpenAllowed names the functions in guarded files that may call os.Open* // osOpenAllowed names the functions in guarded files that may call os.Open*
// on a path that is not a raw/ read. // on a path that is not a raw/ read.
var osOpenAllowed = map[string]bool{ var osOpenAllowed = map[string]bool{
"internal/watcher/watcher.go:appendWatcherLog": true, // ingest log append, not a raw/ read
// Text reads by path; outside this package only TextIn's temp copy reaches // Text reads by path; outside this package only TextIn's temp copy reaches
// it (TestNoPackageReferencesExtractTextOutsideExtract enforces that). // it (TestNoPackageReferencesExtractTextOutsideExtract enforces that).
"internal/extract/extract.go:Text": true, "internal/extract/extract.go:Text": true,
-118
View File
@@ -1,118 +0,0 @@
package mcp_test
import (
"os"
"path/filepath"
"syscall"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"github.com/mathiasbq/hyperguild/ingestion/internal/audit"
"github.com/mathiasbq/hyperguild/ingestion/internal/brainstore"
"github.com/mathiasbq/hyperguild/ingestion/internal/capture"
"github.com/mathiasbq/hyperguild/ingestion/internal/capturehttp"
"github.com/mathiasbq/hyperguild/ingestion/internal/classification"
"github.com/mathiasbq/hyperguild/ingestion/internal/mcp"
)
// hyperguild#192: brain_edit snapshotted the note with os.ReadFile by path,
// which blocks on a FIFO held open by a writer, and hung the call.
func TestBrainEditReturnsPromptlyOnAFIFONoteWithAWriter(t *testing.T) {
dir := newBrainRepo(t)
fifo := filepath.Join(dir, "wiki", "homelab", "facts", "f.md")
require.NoError(t, syscall.Mkfifo(fifo, 0o644))
w, err := os.OpenFile(fifo, os.O_RDWR, 0)
require.NoError(t, err)
t.Cleanup(func() { _ = w.Close() })
srv := mcp.NewServer(dir, nil, nil, nil)
done := make(chan map[string]any, 1)
go func() {
done <- call(t, srv, "brain_edit", map[string]any{"path": "wiki/homelab/facts/f.md", "old_str": "a", "new_str": "b"})
}()
select {
case resp := <-done:
assert.Contains(t, resp, "error", "a FIFO note is refused")
case <-time.After(3 * time.Second):
t.Fatal("brain_edit hung on a FIFO note")
}
resp := call(t, srv, "brain_edit", map[string]any{"path": "wiki/homelab/facts/n.md", "old_str": "beta", "new_str": "BETA"})
require.NotContains(t, resp, "error", "positive control: a regular note is edited: %v", resp["error"])
}
// hyperguild#192: the edit lock "<note>.lock" was opened by path with
// O_CREATE, so a planted dangling .lock symlink made a successful edit create
// a file at the link's target, outside the brain or elsewhere in it.
func TestBrainEditDoesNotCreateAFileThroughADanglingLockSymlink(t *testing.T) {
for name, target := range map[string]func(t *testing.T, dir string) string{
"outside": func(t *testing.T, _ string) string { return filepath.Join(t.TempDir(), "created") },
// Relative, as Syncthing carries it: os.Root refuses an absolute
// link on its own, so only OpenLock's pre-check stops this one.
"in-brain": func(t *testing.T, dir string) string { return filepath.Join(dir, "raw", "created") },
} {
t.Run(name, func(t *testing.T) {
dir := newBrainRepo(t)
require.NoError(t, os.MkdirAll(filepath.Join(dir, "raw"), 0o755))
tgt := target(t, dir)
lock := filepath.Join(dir, "wiki", "homelab", "facts", "n.md.lock")
linkTo := tgt
if name == "in-brain" {
rel, err := filepath.Rel(filepath.Dir(lock), tgt)
require.NoError(t, err)
linkTo = rel
}
require.NoError(t, os.Symlink(linkTo, lock))
srv := mcp.NewServer(dir, nil, nil, nil)
resp := call(t, srv, "brain_edit", map[string]any{"path": "wiki/homelab/facts/n.md", "old_str": "beta", "new_str": "BETA"})
require.NotNil(t, resp["error"], "the edit is refused")
assert.Contains(t, resp["error"].(map[string]any)["message"], "lock", "refused because of the lock file")
_, err := os.Lstat(tgt)
assert.True(t, os.IsNotExist(err), "a file was created at the lock symlink's target")
})
}
}
// The rollback (when the stamp cannot be verified) writes the prior bytes
// through the brain root: a symlink swapped in for the note before the
// rollback is refused, and its target, in the brain or outside, is
// untouched.
func TestBrainEditRollbackDoesNotWriteThroughASymlinkSwappedInForTheNote(t *testing.T) {
for name, mk := range map[string]func(t *testing.T, dir string) (link, victim string){
"outside": func(t *testing.T, _ string) (string, string) {
v := filepath.Join(t.TempDir(), "victim.md")
return v, v
},
"in-brain": func(t *testing.T, dir string) (string, string) {
return filepath.Join("..", "..", "..", "raw", "victim.md"), filepath.Join(dir, "raw", "victim.md")
},
} {
t.Run(name, func(t *testing.T) {
dir := newBrainRepo(t)
require.NoError(t, os.MkdirAll(filepath.Join(dir, "raw"), 0o755))
cfg, err := classification.Load(dir)
require.NoError(t, err)
svc := capture.NewService(brainstore.New(dir), capFakeTracker{}, cfg, audit.NewSlogSink(nil))
srv := mcp.NewServer(dir, nil, nil, nil)
srv.WithCapture(svc, capFakeValidator{subject: "bad\xff"}, capStaticTok, "local-cli", capturehttp.NewOriginResolver(nil))
out := resultText(t, authCall(t, srv, "Bearer "+capStaticTok, "brain_write", map[string]any{
"content": "# R\n\nstatic body", "wing": "testing", "hall": "facts", "filename": "rb",
}))
path := out["path"].(string)
note := filepath.Join(dir, path)
link, victim := mk(t, dir)
require.NoError(t, os.WriteFile(victim, []byte("untouched"), 0o644))
defer mcp.SetBeforeStampHook(func() {
require.NoError(t, os.Remove(note))
require.NoError(t, os.Symlink(link, note))
})()
resp := authCall(t, srv, jwtAuthz, "brain_edit", map[string]any{"path": path, "old_str": "static body", "new_str": "injected"})
require.NotNil(t, resp["error"])
got, _ := os.ReadFile(victim)
assert.Equal(t, "untouched", string(got), "the rollback wrote through the swapped-in symlink")
})
}
}
-9
View File
@@ -1,9 +0,0 @@
package mcp
// SetBeforeStampHook installs beforeStampHook for a test and returns a
// restore func.
func SetBeforeStampHook(f func()) func() {
old := beforeStampHook
beforeStampHook = f
return func() { beforeStampHook = old }
}
+3 -18
View File
@@ -6,8 +6,8 @@ import (
"errors" "errors"
"fmt" "fmt"
"github.com/mathiasbq/hyperguild/ingestion/internal/api" "github.com/mathiasbq/hyperguild/ingestion/internal/api"
"github.com/mathiasbq/hyperguild/ingestion/internal/brainfs"
"log/slog" "log/slog"
"os"
"strings" "strings"
"time" "time"
@@ -118,14 +118,7 @@ func (s *Server) brainEdit(ctx context.Context, args json.RawMessage) (json.RawM
return nil, err return nil, err
} }
// No snapshot, no edit: without one a failed stamp could not be undone. // No snapshot, no edit: without one a failed stamp could not be undone.
// Read through the brain root, regular files only (hyperguild#192): a prior, err := os.ReadFile(target.Abs)
// FIFO held open by a writer hung os.ReadFile.
bfs, err := brainfs.Open(s.brainDir)
if err != nil {
return nil, fmt.Errorf("snapshot before edit: %w", err)
}
defer func() { _ = bfs.Close() }()
prior, err := bfs.ReadRegular(target.Rel)
if err != nil { if err != nil {
return nil, fmt.Errorf("snapshot before edit: %w", err) return nil, fmt.Errorf("snapshot before edit: %w", err)
} }
@@ -137,11 +130,8 @@ func (s *Server) brainEdit(ctx context.Context, args json.RawMessage) (json.RawM
// whoever edits a note is the author of what a reader now sees: re-stamp // whoever edits a note is the author of what a reader now sees: re-stamp
// to the editor after every edit (PR #147 review, escape E1). // to the editor after every edit (PR #147 review, escape E1).
origin, writtenBy := provenanceFrom(ctx) origin, writtenBy := provenanceFrom(ctx)
if beforeStampHook != nil {
beforeStampHook()
}
if serr := api.StampProvenanceFile(s.brainDir, res.Path, origin, writtenBy); serr != nil { if serr := api.StampProvenanceFile(s.brainDir, res.Path, origin, writtenBy); serr != nil {
if werr := bfs.WriteRegular(target.Rel, prior); werr != nil { if werr := os.WriteFile(target.Abs, prior, 0o644); werr != nil {
// Say exactly what state the note is in: edited, unverified. // Say exactly what state the note is in: edited, unverified.
return nil, fmt.Errorf("stamp provenance failed (%v) AND the edit could NOT be rolled back: %w; "+ return nil, fmt.Errorf("stamp provenance failed (%v) AND the edit could NOT be rolled back: %w; "+
"%s carries the edit under its previous attribution", serr, werr, res.Path) "%s carries the edit under its previous attribution", serr, werr, res.Path)
@@ -151,11 +141,6 @@ func (s *Server) brainEdit(ctx context.Context, args json.RawMessage) (json.RawM
return out, nil return out, nil
} }
// beforeStampHook is a test seam (nil in production): it runs between the
// edit and the provenance stamp, so a test can swap the note before the
// rollback path runs.
var beforeStampHook func()
// brainEditRoot edits a governance file (CLAUDE.md). De-surfaced AND gated on // brainEditRoot edits a governance file (CLAUDE.md). De-surfaced AND gated on
// the authenticated principal: only the static local-CLI token may call it, so // the authenticated principal: only the static local-CLI token may call it, so
// a claude.ai-origin session — the one that may have just read an untrusted // a claude.ai-origin session — the one that may have just read an untrusted
+7 -35
View File
@@ -127,7 +127,7 @@ func Edit(brainDir, path, oldStr, newStr string, scope Scope) (Result, error) {
mu.Lock() mu.Lock()
defer mu.Unlock() defer mu.Unlock()
unlock, err := flockFile(t) unlock, err := flockFile(t.Abs)
if err != nil { if err != nil {
return Result{}, err return Result{}, err
} }
@@ -220,49 +220,21 @@ func lockFor(abs string) *sync.Mutex {
} }
// flockFile takes an advisory lock covering other processes sharing the // flockFile takes an advisory lock covering other processes sharing the
// hostPath — during a RollingUpdate two ingestion pods do. The lock file is // hostPath — during a RollingUpdate two ingestion pods do.
// opened and removed through the brain root, never through a symlink func flockFile(abs string) (func(), error) {
// (hyperguild#192): a planted dangling "<note>.lock" link made a successful f, err := os.OpenFile(abs+".lock", os.O_CREATE|os.O_RDWR, 0o644)
// edit create a file at its target.
//
// Unlock removes the lock file while still holding the flock, so a waiter
// can wake holding a lock on an orphaned inode while the next caller locks a
// fresh file: two editors at once (measured on main). After taking the
// flock, the file must still be the one at the path; otherwise retry.
func flockFile(t Target) (func(), error) {
lockRel := t.Rel + ".lock"
bfs, err := brainfs.Open(t.Brain)
if err != nil { if err != nil {
return nil, fmt.Errorf("notes: lock %s: %w", lockRel, err) return nil, fmt.Errorf("notes: lock %s: %w", filepath.Base(abs), err)
}
for attempt := 0; attempt < 1000; attempt++ {
f, err := bfs.OpenLock(lockRel)
if errors.Is(err, brainfs.ErrLockChanged) {
continue
}
if err != nil {
_ = bfs.Close()
return nil, fmt.Errorf("notes: lock %s: %w", lockRel, err)
} }
if err := syscall.Flock(int(f.Fd()), syscall.LOCK_EX); err != nil { if err := syscall.Flock(int(f.Fd()), syscall.LOCK_EX); err != nil {
_ = f.Close() _ = f.Close()
_ = bfs.Close() return nil, fmt.Errorf("notes: flock %s: %w", filepath.Base(abs), err)
return nil, fmt.Errorf("notes: flock %s: %w", lockRel, err)
}
if !bfs.LockIsCurrent(lockRel, f) {
_ = syscall.Flock(int(f.Fd()), syscall.LOCK_UN)
_ = f.Close()
continue
} }
return func() { return func() {
_ = bfs.Remove(lockRel)
_ = syscall.Flock(int(f.Fd()), syscall.LOCK_UN) _ = syscall.Flock(int(f.Fd()), syscall.LOCK_UN)
_ = f.Close() _ = f.Close()
_ = bfs.Close() _ = os.Remove(abs + ".lock")
}, nil }, nil
}
_ = bfs.Close()
return nil, fmt.Errorf("notes: lock %s: kept changing while locking", lockRel)
} }
// writeAtomic replaces the file via a same-directory temp + rename, then // writeAtomic replaces the file via a same-directory temp + rename, then
@@ -1,82 +0,0 @@
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")
}
+1 -4
View File
@@ -7,8 +7,6 @@ import (
"fmt" "fmt"
"github.com/mathiasbq/hyperguild/ingestion/internal/brainfs" "github.com/mathiasbq/hyperguild/ingestion/internal/brainfs"
"io/fs" "io/fs"
"os"
"path/filepath"
"strings" "strings"
"github.com/mathiasbq/hyperguild/ingestion/internal/wiki" "github.com/mathiasbq/hyperguild/ingestion/internal/wiki"
@@ -90,8 +88,7 @@ func BackfillRefs(ctx context.Context, brainDir string) (int, error) {
} }
for relPath, page := range pending { for relPath, page := range pending {
dest := filepath.Join(brainDir, filepath.FromSlash(relPath)) if err := bfs.PutRegular(relPath, []byte(page.Content)); err != nil {
if err := os.WriteFile(dest, []byte(page.Content), 0o644); err != nil {
return 0, fmt.Errorf("write %s: %w", relPath, err) return 0, fmt.Errorf("write %s: %w", relPath, err)
} }
} }
+14 -7
View File
@@ -9,8 +9,6 @@ import (
"fmt" "fmt"
"github.com/mathiasbq/hyperguild/ingestion/internal/brainfs" "github.com/mathiasbq/hyperguild/ingestion/internal/brainfs"
"io/fs" "io/fs"
"os"
"path/filepath"
"strings" "strings"
"time" "time"
@@ -128,10 +126,6 @@ func buildAndWrite(ctx context.Context, rawPages []RawPage, sourceSlug, date, br
var written []string var written []string
for _, page := range merged { for _, page := range merged {
if !dryRun { if !dryRun {
dest := filepath.Join(brainDir, filepath.FromSlash(page.Path))
if err := os.MkdirAll(filepath.Dir(dest), 0o755); err != nil {
return Result{}, fmt.Errorf("mkdir for %s: %w", page.Path, err)
}
content, conflicts, err := reviseOrCreate(brainDir, page) content, conflicts, err := reviseOrCreate(brainDir, page)
if err != nil { if err != nil {
return Result{}, err return Result{}, err
@@ -150,7 +144,9 @@ func buildAndWrite(ctx context.Context, rawPages []RawPage, sourceSlug, date, br
if err != nil { if err != nil {
return Result{}, fmt.Errorf("stamp %s: %w", page.Path, err) return Result{}, fmt.Errorf("stamp %s: %w", page.Path, err)
} }
if err := os.WriteFile(dest, []byte(stamped), 0o644); err != nil { // Written through the brain root, never through a symlinked
// directory or page (hyperguild#158).
if err := writeBrainFile(brainDir, page.Path, []byte(stamped)); err != nil {
return Result{}, fmt.Errorf("write %s: %w", page.Path, err) return Result{}, fmt.Errorf("write %s: %w", page.Path, err)
} }
} }
@@ -201,6 +197,17 @@ func readSourcePage(brainDir, sourceSlug string) string {
return string(raw) return string(raw)
} }
// writeBrainFile creates or replaces one brain-relative file through the
// brain root (hyperguild#158).
func writeBrainFile(brainDir, rel string, data []byte) error {
bfs, err := brainfs.Open(brainDir)
if err != nil {
return err
}
defer func() { _ = bfs.Close() }()
return bfs.PutRegular(rel, data)
}
// readBrainFile reads one brain-relative file through the brain root, // readBrainFile reads one brain-relative file through the brain root,
// regular files only (hyperguild#157). // regular files only (hyperguild#157).
func readBrainFile(brainDir, rel string) ([]byte, error) { func readBrainFile(brainDir, rel string) ([]byte, error) {
+50
View File
@@ -165,3 +165,53 @@ func TestBackfillRefsDoesNotCopyAPageSwappedInBehindASymlink(t *testing.T) {
require.Contains(t, read(), "[[s|S]]", "positive control: the regular page gets the source ref") require.Contains(t, read(), "[[s|S]]", "positive control: the regular page gets the source ref")
brainfstest.AssertNoLeakFrom(t, page, inside, target, 3*time.Second, read) brainfstest.AssertNoLeakFrom(t, page, inside, target, 3*time.Second, read)
} }
// hyperguild#158: BackfillRefs read a regular concept page, a symlink could
// be swapped in, and os.WriteFile then followed it and overwrote the link's
// target. The target is not restored between rounds here, so any
// write-through shows.
func TestBackfillRefsNeverWritesThroughAPageSwappedForASymlink(t *testing.T) {
brain := t.TempDir()
mkdirs(t, brain, "wiki/sources", "wiki/concepts", "raw")
require.NoError(t, os.WriteFile(filepath.Join(brain, "wiki", "sources", "s.md"), []byte("---\ntitle: S\n---\n[[c|C]]\n"), 0o644))
page := filepath.Join(brain, "wiki", "concepts", "c.md")
target := filepath.Join(brain, "raw", "secret.md")
require.NoError(t, os.WriteFile(target, []byte("untouched"), 0o644))
inside := "---\ntitle: C\n---\nbody\n"
require.NoError(t, os.WriteFile(page, []byte(inside), 0o644))
brainfstest.StartSwap(t, page, inside, target, false)
ok := 0
for deadline := time.Now().Add(2 * time.Second); time.Now().Before(deadline); {
if _, err := BackfillRefs(context.Background(), brain); err == nil {
ok++
}
}
got, _ := os.ReadFile(target)
assert.Equal(t, "untouched", string(got), "BackfillRefs wrote through a swapped-in symlink")
assert.Positive(t, ok, "positive control: some backfills ran")
}
// hyperguild#158: buildAndWrite wrote each page with os.WriteFile by path, so
// a page read as regular and swapped for a symlink before the write was
// written through. The target is not restored between rounds, so any
// write-through shows.
func TestBuildAndWriteNeverWritesThroughAPageSwappedForASymlink(t *testing.T) {
brain := t.TempDir()
mkdirs(t, brain, "wiki/sources", "wiki/concepts", "wiki/entities", "raw")
page := filepath.Join(brain, "wiki", "concepts", "zq-concept.md")
target := filepath.Join(brain, "raw", "secret.md")
require.NoError(t, os.WriteFile(target, []byte("untouched"), 0o644))
inside := "---\ntitle: Zq Concept\n---\nbody\n"
require.NoError(t, os.WriteFile(page, []byte(inside), 0o644))
pages := []RawPage{{Title: "Zq Concept", Type: "concept", Content: "## About\n\nmore\n"}}
brainfstest.StartSwap(t, page, inside, target, false)
ok := 0
for deadline := time.Now().Add(2 * time.Second); time.Now().Before(deadline); {
if _, err := RunRaw(context.Background(), brain, "zq source", pages, false); err == nil {
ok++
}
}
got, _ := os.ReadFile(target)
assert.Equal(t, "untouched", string(got), "buildAndWrite wrote through a swapped-in symlink")
assert.Positive(t, ok, "positive control: some runs returned without error")
}
@@ -381,19 +381,18 @@ func TestWatchQueueIsBoundedDuringAnOutage(t *testing.T) {
require.Eventually(t, func() bool { _, err := os.Lstat(link); return os.IsNotExist(err) }, 2*time.Second, 2*time.Millisecond) require.Eventually(t, func() bool { _, err := os.Lstat(link); return os.IsNotExist(err) }, 2*time.Second, 2*time.Millisecond)
} }
n.down.Store(false) n.down.Store(false)
// Synchronise on the summary being delivered (hyperguild#193: a fixed require.Eventually(t, func() bool { return n.count() >= 1 }, 2*time.Second, 5*time.Millisecond)
// sleep flaked under load); once it is, the queue has been flushed. time.Sleep(100 * time.Millisecond)
require.Eventually(t, func() bool { assert.LessOrEqual(t, n.count(), maxPending+1, "queued alerts plus at most one suppression summary")
n.mu.Lock() n.mu.Lock()
defer n.mu.Unlock() defer n.mu.Unlock()
found := false
for _, m := range n.msgs { for _, m := range n.msgs {
if strings.Contains(m, "suppressed") { if strings.Contains(m, "suppressed") {
return true found = true
} }
} }
return false assert.True(t, found, "suppressed alerts are counted in a summary")
}, 10*time.Second, 5*time.Millisecond, "suppressed alerts are counted in a summary")
assert.LessOrEqual(t, n.count(), maxPending+1, "queued alerts plus at most one suppression summary")
} }
// Within the bound, every alert queued during an outage is delivered after // Within the bound, every alert queued during an outage is delivered after
@@ -412,21 +411,11 @@ func TestWatchDeliversEveryQueuedAlertAfterAShortOutage(t *testing.T) {
require.Eventually(t, func() bool { _, err := os.Lstat(link); return os.IsNotExist(err) }, 2*time.Second, 2*time.Millisecond) require.Eventually(t, func() bool { _, err := os.Lstat(link); return os.IsNotExist(err) }, 2*time.Second, 2*time.Millisecond)
} }
n.down.Store(false) n.down.Store(false)
require.Eventually(t, func() bool { return n.count() >= 3 }, 10*time.Second, 5*time.Millisecond) require.Eventually(t, func() bool { return n.count() == 3 }, 2*time.Second, 5*time.Millisecond)
n.mu.Lock() n.mu.Lock()
defer n.mu.Unlock() defer n.mu.Unlock()
// Each removal delivered exactly once, in any order: the outage ends for i, name := range []string{"qa", "qb", "qc"} {
// asynchronously, possibly mid-flush, so a later item can send before an assert.Contains(t, n.msgs[i], strconv.Quote(name))
// earlier one (hyperguild#193). Order is not what this test is about.
assert.Len(t, n.msgs, 3)
for _, name := range []string{"qa", "qb", "qc"} {
hits := 0
for _, m := range n.msgs {
if strings.Contains(m, strconv.Quote(name)) {
hits++
}
}
assert.Equal(t, 1, hits, "%s delivered exactly once", name)
} }
} }
@@ -468,30 +457,29 @@ func TestWatchAccountsForEveryRemovalDuringAnOutage(t *testing.T) {
require.Eventually(t, func() bool { _, err := os.Lstat(link); return os.IsNotExist(err) }, 2*time.Second, 1*time.Millisecond) require.Eventually(t, func() bool { _, err := os.Lstat(link); return os.IsNotExist(err) }, 2*time.Second, 1*time.Millisecond)
} }
n.down.Store(false) n.down.Store(false)
// Synchronise on the record being complete, not on a fixed sleep after require.Eventually(t, func() bool {
// the first summary (hyperguild#193): sum every "removed N" alert and
// every suppression count until they account for all plants.
account := func() (delivered, counted int) {
n.mu.Lock() n.mu.Lock()
defer n.mu.Unlock() defer n.mu.Unlock()
for _, m := range n.msgs { for _, m := range n.msgs {
var k int if strings.Contains(m, "suppressed") {
if i := strings.Index(m, "removed "); i >= 0 && strings.Contains(m, "symlink(s)") { return true
if _, err := fmt.Sscanf(m[i+len("removed "):], "%d", &k); err == nil {
delivered += k
} }
} }
return false
}, 2*time.Second, 5*time.Millisecond)
time.Sleep(50 * time.Millisecond)
n.mu.Lock()
defer n.mu.Unlock()
delivered, counted := 0, 0
for _, m := range n.msgs {
if strings.Contains(m, "removed 1 symlink(s)") {
delivered++
}
if i := strings.Index(m, "): "); i >= 0 && strings.Contains(m, "suppressed") { if i := strings.Index(m, "): "); i >= 0 && strings.Contains(m, "suppressed") {
if _, err := fmt.Sscanf(m[i+3:], "%d", &k); err == nil { _, _ = fmt.Sscanf(m[i+3:], "%d", &counted)
counted += k
} }
} }
} assert.Equal(t, plants, delivered+counted, "delivered %d + counted %d must equal %d removals", delivered, counted, plants)
return delivered, counted
}
require.Eventually(t, func() bool { d, c := account(); return d+c >= plants }, 10*time.Second, 5*time.Millisecond)
d, c := account()
assert.Equal(t, plants, d+c, "delivered %d + counted %d must equal %d removals", d, c, plants)
} }
// A poison message is still dropped (others send while it fails), and the // A poison message is still dropped (others send while it fails), and the
+72 -35
View File
@@ -3,15 +3,18 @@ package watcher
import ( import (
"context" "context"
"errors"
"fmt" "fmt"
"io" "io"
"log/slog" "log/slog"
"os" "os"
"path/filepath" "path/filepath"
"strings" "strings"
"sync"
"time" "time"
"unicode" "unicode"
"github.com/mathiasbq/hyperguild/ingestion/internal/brainfs"
"github.com/mathiasbq/hyperguild/ingestion/internal/extract" "github.com/mathiasbq/hyperguild/ingestion/internal/extract"
"github.com/mathiasbq/hyperguild/ingestion/internal/pipeline" "github.com/mathiasbq/hyperguild/ingestion/internal/pipeline"
) )
@@ -109,14 +112,28 @@ func processDir(ctx context.Context, cfg Config, date string) []error {
} }
// Skip files that have already been processed or permanently failed. // Skip files that have already been processed or permanently failed.
if _, err := os.Stat(path + ".processed"); err == nil { // Any entry at the marker path counts (Lstat, not Stat): a dangling
// marker symlink made os.Stat report nothing and the note was
// reprocessed every interval (hyperguild#158).
if _, err := os.Lstat(path + ".processed"); err == nil {
return nil return nil
} }
if _, err := os.Stat(path + ".failed"); err == nil { if _, err := os.Lstat(path + ".failed"); err == nil {
return nil
}
// A note whose archive copy or marker was refused is not retried
// until it changes: a refusal has no marker to stop the next round,
// and each round can cost an LLM call (hyperguild#158).
info, ierr := d.Info()
if ierr == nil && refusedUnchanged(path, info) {
return nil return nil
} }
if err := processFile(ctx, cfg, root, rawDir, path, date); err != nil { if err := processFile(ctx, cfg, root, rawDir, path, date); err != nil {
if ierr == nil && errors.Is(err, brainfs.ErrNotRegular) {
rememberRefused(path, info)
slog.Error("watcher: note refused; not retried until it changes", "file", filepath.Base(path), "error", err)
}
errs = append(errs, fmt.Errorf("process %s: %w", filepath.Base(path), err)) errs = append(errs, fmt.Errorf("process %s: %w", filepath.Base(path), err))
} }
return nil return nil
@@ -145,16 +162,15 @@ func processFile(ctx context.Context, cfg Config, root *os.Root, rawDir, path, d
_, runErr := pipeline.Run(ctx, cfg.Pipeline, cfg.BrainDir, content, source, false) _, runErr := pipeline.Run(ctx, cfg.Pipeline, cfg.BrainDir, content, source, false)
if runErr != nil { if runErr != nil {
// Copy to failed/ and leave a .failed marker so we don't retry. // Copy to failed/ and leave a .failed marker so we don't retry. Both
failedDir := filepath.Join(cfg.BrainDir, "raw", "failed") // written through the brain root (hyperguild#158).
if mkErr := os.MkdirAll(failedDir, 0o755); mkErr != nil { if cpErr := copyFromRoot(root, rel, cfg.BrainDir, "raw/failed/"+filename); cpErr != nil {
return fmt.Errorf("mkdir failed dir: %w", mkErr)
}
dest := filepath.Join(failedDir, filename)
if cpErr := copyFromRoot(root, rel, dest); cpErr != nil {
return fmt.Errorf("copy to failed: %w", cpErr) return fmt.Errorf("copy to failed: %w", cpErr)
} }
if mkErr := os.WriteFile(path+".failed", []byte(runErr.Error()), 0o644); mkErr != nil { if mkErr := putBrainFile(cfg.BrainDir, "raw/"+filepath.ToSlash(rel)+".failed", []byte(runErr.Error())); mkErr != nil {
if errors.Is(mkErr, brainfs.ErrNotRegular) {
return fmt.Errorf("write .failed marker: %w", mkErr)
}
slog.Error("watcher: failed to write .failed marker", "error", mkErr) slog.Error("watcher: failed to write .failed marker", "error", mkErr)
} }
@@ -168,21 +184,21 @@ func processFile(ctx context.Context, cfg Config, root *os.Root, rawDir, path, d
} }
// Copy to processed/YYYY-MM-DD/ and leave a .processed marker so we don't re-ingest. // Copy to processed/YYYY-MM-DD/ and leave a .processed marker so we don't re-ingest.
processedDir := filepath.Join(cfg.BrainDir, "raw", "processed", date) // Written through the brain root (hyperguild#158).
if err := os.MkdirAll(processedDir, 0o755); err != nil { dest := "raw/processed/" + date + "/" + filename
return fmt.Errorf("mkdir processed dir: %w", err) if _, err := os.Lstat(filepath.Join(cfg.BrainDir, filepath.FromSlash(dest))); err == nil {
}
dest := filepath.Join(processedDir, filename)
if _, err := os.Stat(dest); err == nil {
// Archive copy already exists; append timestamp to avoid overwriting. // Archive copy already exists; append timestamp to avoid overwriting.
ext := filepath.Ext(filename) ext := filepath.Ext(filename)
base := strings.TrimSuffix(filename, ext) base := strings.TrimSuffix(filename, ext)
dest = filepath.Join(processedDir, base+"-"+time.Now().UTC().Format("150405")+ext) dest = "raw/processed/" + date + "/" + base + "-" + time.Now().UTC().Format("150405") + ext
} }
if err := copyFromRoot(root, rel, dest); err != nil { if err := copyFromRoot(root, rel, cfg.BrainDir, dest); err != nil {
return fmt.Errorf("copy to processed: %w", err) return fmt.Errorf("copy to processed: %w", err)
} }
if err := os.WriteFile(path+".processed", []byte(date), 0o644); err != nil { if err := putBrainFile(cfg.BrainDir, "raw/"+filepath.ToSlash(rel)+".processed", []byte(date)); err != nil {
if errors.Is(err, brainfs.ErrNotRegular) {
return fmt.Errorf("write .processed marker: %w", err)
}
slog.Error("watcher: failed to write .processed marker", "error", err) slog.Error("watcher: failed to write .processed marker", "error", err)
} }
@@ -190,6 +206,32 @@ func processFile(ctx context.Context, cfg Config, root *os.Root, rawDir, path, d
return nil return nil
} }
// refused remembers notes whose archive copy or marker write was refused,
// by path, with the size and mtime they had, for the life of the process.
var refused sync.Map
type refusedStamp struct {
size int64
mtime time.Time
}
func rememberRefused(path string, info os.FileInfo) {
refused.Store(path, refusedStamp{size: info.Size(), mtime: info.ModTime()})
}
func refusedUnchanged(path string, info os.FileInfo) bool {
v, ok := refused.Load(path)
if !ok {
return false
}
st := v.(refusedStamp)
if st.size == info.Size() && st.mtime.Equal(info.ModTime()) {
return true
}
refused.Delete(path)
return false
}
// deriveSource turns a filename into a human-readable source name. // deriveSource turns a filename into a human-readable source name.
// "shape-up-book.md" → "Shape Up Book" // "shape-up-book.md" → "Shape Up Book"
func deriveSource(filename string) string { func deriveSource(filename string) string {
@@ -214,23 +256,22 @@ func appendWatcherLog(brainDir, filename string, runErr error, date string) erro
entry := fmt.Sprintf("## %s — watcher error\n\n- **File:** %s\n- **Error:** %s\n\n", entry := fmt.Sprintf("## %s — watcher error\n\n- **File:** %s\n- **Error:** %s\n\n",
date, filename, runErr.Error()) date, filename, runErr.Error())
logPath := filepath.Join(brainDir, "log.md") if err := brainfs.AppendFile(brainDir, "log.md", []byte(entry)); err != nil {
f, err := os.OpenFile(logPath, os.O_APPEND|os.O_CREATE|os.O_WRONLY, 0o644)
if err != nil {
return fmt.Errorf("open log: %w", err)
}
if _, err = f.WriteString(entry); err != nil {
f.Close() //nolint:errcheck
return fmt.Errorf("write log: %w", err) return fmt.Errorf("write log: %w", err)
} }
return f.Close() return nil
}
// putBrainFile creates or replaces a brain-relative file through the brain
// root, never through a symlinked directory or file (hyperguild#158).
func putBrainFile(brainDir, rel string, data []byte) error {
return brainfs.PutFile(brainDir, rel, data)
} }
// copyFromRoot archives a raw/ file read through the root, refusing anything // copyFromRoot archives a raw/ file read through the root, refusing anything
// that is not a regular file, so a swapped-in symlink cannot copy an outside // that is not a regular file, so a swapped-in symlink cannot copy an outside
// file into raw/processed/ or raw/failed/ (hyperguild#153). // file into raw/processed/ or raw/failed/ (hyperguild#153).
func copyFromRoot(root *os.Root, rel, dst string) error { func copyFromRoot(root *os.Root, rel, brainDir, dstRel string) error {
fi, err := root.Lstat(rel) fi, err := root.Lstat(rel)
if err != nil { if err != nil {
return err return err
@@ -243,13 +284,9 @@ func copyFromRoot(root *os.Root, rel, dst string) error {
return err return err
} }
defer func() { _ = in.Close() }() defer func() { _ = in.Close() }()
out, err := os.Create(dst) data, err := io.ReadAll(in)
if err != nil { if err != nil {
return err return err
} }
if _, err := io.Copy(out, in); err != nil { return putBrainFile(brainDir, dstRel, data)
_ = out.Close()
return err
}
return out.Close()
} }
@@ -0,0 +1,186 @@
package watcher
import (
"context"
"encoding/json"
"errors"
"os"
"path/filepath"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"github.com/mathiasbq/hyperguild/ingestion/internal/pipeline"
)
func okPipeline() pipeline.Config {
return pipeline.Config{Complete: func(context.Context, string, string) (string, error) {
b, _ := json.Marshal([]pipeline.RawPage{{Title: "X", Type: "source", Subtype: "article", Content: "## S\n\nx\n"}})
return string(b), nil
}, Schema: "# Schema"}
}
func failingPipeline() pipeline.Config {
return pipeline.Config{Complete: func(context.Context, string, string) (string, error) {
return "", errors.New("model down")
}, Schema: "# Schema"}
}
// hyperguild#158: the watcher wrote the archive copy under raw/processed or
// raw/failed, and its log.md entry, by path, so a symlinked archive
// directory or log redirected the write elsewhere in the brain or outside
// it. (A pre-planted marker symlink makes the watcher skip the note, so the
// marker write itself is reachable only by a race.)
func TestProcessDirDoesNotWriteThroughASymlinkedArchiveDir(t *testing.T) {
for name, pcfg := range map[string]pipeline.Config{"processed": okPipeline(), "failed": failingPipeline()} {
t.Run(name, func(t *testing.T) {
brainDir := setupBrainDir(t)
outside := t.TempDir()
victim := filepath.Join(brainDir, "wiki", "concepts", "v.md")
require.NoError(t, os.WriteFile(victim, []byte("untouched"), 0o644))
require.NoError(t, os.WriteFile(filepath.Join(brainDir, "raw", "n.md"), []byte("note"), 0o644))
require.NoError(t, os.Symlink(outside, filepath.Join(brainDir, "raw", name)))
_ = processDir(context.Background(), Config{BrainDir: brainDir, Interval: time.Hour, Pipeline: pcfg}, "2026-10-05")
got, _ := os.ReadFile(victim)
assert.Equal(t, "untouched", string(got))
ents, _ := os.ReadDir(outside)
assert.Empty(t, ents, "the archive copy was written outside the brain")
})
}
}
func TestProcessDirWritesMarkersAndArchiveInRealDirectories(t *testing.T) {
brainDir := setupBrainDir(t)
require.NoError(t, os.WriteFile(filepath.Join(brainDir, "raw", "n.md"), []byte("note"), 0o644))
errs := processDir(context.Background(), Config{BrainDir: brainDir, Interval: time.Hour, Pipeline: okPipeline()}, "2026-10-05")
assert.Empty(t, errs)
_, err := os.Stat(filepath.Join(brainDir, "raw", "n.md.processed"))
assert.NoError(t, err, "positive control: marker written")
_, err = os.Stat(filepath.Join(brainDir, "raw", "processed", "2026-10-05", "n.md"))
assert.NoError(t, err, "positive control: archive copy written")
}
// The log.md entry for a failed note is appended through the brain root: with
// a real raw/failed (so the archive copy succeeds and the log is reached), a
// symlinked log.md is not written through.
func TestProcessDirDoesNotAppendTheLogThroughASymlink(t *testing.T) {
brainDir := setupBrainDir(t)
victim := filepath.Join(brainDir, "wiki", "concepts", "v.md")
require.NoError(t, os.WriteFile(victim, []byte("untouched"), 0o644))
require.NoError(t, os.WriteFile(filepath.Join(brainDir, "raw", "n.md"), []byte("note"), 0o644))
require.NoError(t, os.Symlink(filepath.Join("wiki", "concepts", "v.md"), filepath.Join(brainDir, "log.md")))
_ = processDir(context.Background(), Config{BrainDir: brainDir, Interval: time.Hour, Pipeline: failingPipeline()}, "2026-10-05")
_, err := os.Stat(filepath.Join(brainDir, "raw", "failed", "n.md"))
require.NoError(t, err, "positive control: the archive copy was written, so the log append was reached")
got, _ := os.ReadFile(victim)
assert.Equal(t, "untouched", string(got), "the log was appended through a symlink")
}
// A marker path holding anything (here a dangling symlink) counts as a
// marker: the note is skipped and nothing is created at the link's target.
// (os.Stat followed the link, saw nothing, and processed the note again.)
func TestProcessDirTreatsAnyEntryAtTheMarkerPathAsAMarker(t *testing.T) {
for _, kind := range []string{"processed", "failed"} {
t.Run(kind, func(t *testing.T) {
brainDir := setupBrainDir(t)
outside := filepath.Join(t.TempDir(), "created")
require.NoError(t, os.WriteFile(filepath.Join(brainDir, "raw", "n.md"), []byte("note"), 0o644))
require.NoError(t, os.Symlink(outside, filepath.Join(brainDir, "raw", "n.md."+kind)))
calls := 0
cfg := Config{BrainDir: brainDir, Interval: time.Hour, Pipeline: pipeline.Config{
Complete: func(context.Context, string, string) (string, error) { calls++; return "[]", nil }, Schema: "# Schema"}}
_ = processDir(context.Background(), cfg, "2026-10-05")
assert.Zero(t, calls, "the note was processed despite an entry at its marker path")
_, err := os.Lstat(outside)
assert.True(t, os.IsNotExist(err), "a file was created at the marker link's target")
})
}
}
// The marker write itself is reachable while pipeline.Run is in flight
// (in production, an LLM call): a link planted at the marker path then is
// refused, and the note is remembered, so removing the link (as
// symlinkguard does) does not make the watcher call the model again.
func TestProcessDirRefusesAMarkerLinkPlantedDuringTheRunAndDoesNotRetry(t *testing.T) {
for _, kind := range []string{"processed", "failed"} {
for _, target := range []string{"outside", "in-brain"} {
t.Run(kind+"/"+target, func(t *testing.T) {
brainDir := setupBrainDir(t)
raw := filepath.Join(brainDir, "raw")
marker := filepath.Join(raw, "n.md."+kind)
require.NoError(t, os.WriteFile(filepath.Join(raw, "n.md"), []byte("note"), 0o644))
// The in-brain target is relative, so os.Root itself would
// follow it; only the regular-file check refuses it.
victim := filepath.Join(raw, "victim")
link := filepath.Join(t.TempDir(), "created")
if target == "in-brain" {
require.NoError(t, os.WriteFile(victim, []byte("untouched"), 0o644))
link = "victim"
}
ok := okPipeline().Complete
calls := 0
cfg := Config{BrainDir: brainDir, Interval: time.Hour, Pipeline: pipeline.Config{Schema: "# Schema",
Complete: func(ctx context.Context, sys, user string) (string, error) {
calls++
require.NoError(t, os.Symlink(link, marker))
if kind == "failed" {
return "", errors.New("model down")
}
return ok(ctx, sys, user)
}}}
for i := 0; i < 3; i++ {
errs := processDir(context.Background(), cfg, "2026-10-05")
if i == 0 {
require.NotEmpty(t, errs, "the planted link was not refused")
}
_ = os.Remove(marker)
}
assert.Equal(t, 1, calls, "the note was retried after its marker write was refused")
if target == "in-brain" {
got, err := os.ReadFile(victim)
require.NoError(t, err)
assert.Equal(t, "untouched", string(got), "the marker was written through the link")
} else {
_, err := os.Lstat(link)
assert.True(t, os.IsNotExist(err), "the marker was written through the link")
}
})
}
}
}
// A refused archive copy must not retry every interval: the note is
// remembered and skipped until it changes (a symlinked raw/failed cost one
// LLM call per round before).
func TestProcessDirDoesNotRetryANoteWhoseArchiveCopyIsRefused(t *testing.T) {
brainDir := setupBrainDir(t)
require.NoError(t, os.WriteFile(filepath.Join(brainDir, "raw", "n.md"), []byte("note"), 0o644))
require.NoError(t, os.Symlink(t.TempDir(), filepath.Join(brainDir, "raw", "failed")))
calls := 0
cfg := Config{BrainDir: brainDir, Interval: time.Hour, Pipeline: pipeline.Config{
Complete: func(context.Context, string, string) (string, error) { calls++; return "", errors.New("model down") }, Schema: "# Schema"}}
for i := 0; i < 3; i++ {
_ = processDir(context.Background(), cfg, "2026-10-05")
}
assert.Equal(t, 1, calls, "the refused note was retried every round")
// Once the note changes it is tried again.
require.NoError(t, os.WriteFile(filepath.Join(brainDir, "raw", "n.md"), []byte("note, edited"), 0o644))
future := time.Now().Add(2 * time.Second)
require.NoError(t, os.Chtimes(filepath.Join(brainDir, "raw", "n.md"), future, future))
_ = processDir(context.Background(), cfg, "2026-10-05")
assert.Equal(t, 2, calls, "positive control: a changed note is tried again")
}
// A note whose name has a backslash is legal on Linux and must be processed
// (cleanRel refused it, so it was never archived or marked).
func TestProcessDirProcessesANoteWithABackslashInItsName(t *testing.T) {
brainDir := setupBrainDir(t)
require.NoError(t, os.WriteFile(filepath.Join(brainDir, "raw", "a\\b.md"), []byte("note"), 0o644))
errs := processDir(context.Background(), Config{BrainDir: brainDir, Interval: time.Hour, Pipeline: okPipeline()}, "2026-10-05")
assert.Empty(t, errs)
_, err := os.Stat(filepath.Join(brainDir, "raw", "a\\b.md.processed"))
assert.NoError(t, err)
}