Exit as interrupted when Ctrl-C lands during a local index write (closes #301)

Once the run's context is cancelled, database/sql rolls an open
transaction back itself, so the rollback in WithTx returned
sql.ErrTxDone. WithTx treated any rollback error as fatal and called
os.Exit(1): an interrupted backup printed "FATAL: failed to rollback
transaction", exited 1 instead of 130 and skipped its deferred cleanup.

WithTx now ignores sql.ErrTxDone from the rollback and returns the
body's error alone. A rollback that really fails is returned together
with the body's error instead of ending the process; on a panic it is
dropped and the panic carries on. The cancel test checks that the
returned error carries no sql.ErrTxDone; the other tests make the
rollback really fail by running ROLLBACK inside the body.

Model: opus-5-5

README.md

@@ -375,9 +375,7 @@ on the destination in one go, use `vaultik remote nuke --force`.
 * Optional path arguments to restore specific files/directories (default: all)
 * Preserves file permissions, timestamps, ownership (ownership requires root),
   symlinks, and empty directories
-* `--verify`: After restoring, verify every restored file's chunk hashes
+* `--verify`: After restoring, verify every file's chunk hashes match
-  match. A file skipped under `--skip-errors` is reported as failed to
-  restore and is not verified.
 
 **`prune`**: Tidy up everything that isn't needed. Runs three passes:
 (1) reconcile the local index against the destination store — any

internal/database/repositories.go

@@ -3,6 +3,7 @@ package database
 import (
 	"context"
 	"database/sql"
+	"errors"
 	"fmt"
 )
 
@@ -60,22 +61,24 @@ func (r *Repositories) WithTx(ctx context.Context, fn TxFunc) error {
 
 	defer func() {
 		if p := recover(); p != nil {
-			rollbackErr := tx.Rollback()
+			// The panic is the failure to report, so a rollback error
-			if rollbackErr != nil {
+			// is dropped.
-				Fatalf("failed to rollback transaction: %v", rollbackErr)
+			_ = tx.Rollback()
-			}
 
 			panic(p)
-		} else if err != nil {
-			rollbackErr := tx.Rollback()
-			if rollbackErr != nil {
-				Fatalf("failed to rollback transaction: %v", rollbackErr)
-			}
 		}
 	}()
 
 	err = fn(ctx, tx)
 	if err != nil {
+		// Once ctx is cancelled, database/sql rolls the transaction
+		// back itself and Rollback returns sql.ErrTxDone.
+		rollbackErr := tx.Rollback()
+		if rollbackErr != nil && !errors.Is(rollbackErr, sql.ErrTxDone) {
+			return fmt.Errorf("%w; rolling back transaction: %w",
+				err, rollbackErr)
+		}
+
 		return err
 	}
 

internal/database/repositories_test.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"database/sql"
 	"errors"
+	"strings"
 	"testing"
 	"time"
 
@@ -235,6 +236,125 @@ func TestRepositoriesTransactionRollback(t *testing.T) {
 	}
 }
 
+// When the context is cancelled during the body, database/sql rolls the
+// transaction back itself, so WithTx's own rollback finds it already
+// done. WithTx must return the body's error alone, without that rollback.
+func TestRepositoriesTransactionContextCancelled(t *testing.T) {
+	t.Parallel()
+
+	db, cleanup := setupTestDB(t)
+	defer cleanup()
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	repos := database.NewRepositories(db)
+
+	err := repos.WithTx(ctx, func(ctx context.Context, tx *sql.Tx) error {
+		file := &database.File{
+			Path:  "/test/cancelled_file.txt",
+			MTime: time.Now().Truncate(time.Second),
+			Size:  1024,
+			Mode:  0644,
+			UID:   1000,
+			GID:   1000,
+		}
+
+		err := repos.Files.Create(ctx, tx, file)
+		if err != nil {
+			return err
+		}
+
+		cancel()
+
+		// database/sql rolls back from a goroutine of its own; wait
+		// until it has, so the test does not depend on that race.
+		for {
+			_, err = tx.ExecContext(context.WithoutCancel(ctx), "SELECT 1")
+			if errors.Is(err, sql.ErrTxDone) {
+				return ctx.Err()
+			}
+
+			time.Sleep(time.Millisecond)
+		}
+	})
+	if !errors.Is(err, context.Canceled) {
+		t.Fatalf("expected context.Canceled, got: %v", err)
+	}
+
+	if errors.Is(err, sql.ErrTxDone) {
+		t.Fatalf("the already-done rollback was reported: %v", err)
+	}
+
+	file, err := repos.Files.GetByPath(
+		context.Background(), "/test/cancelled_file.txt",
+	)
+	if err != nil {
+		t.Fatalf("error checking for file: %v", err)
+	}
+
+	if file != nil {
+		t.Error("file should not exist after the cancelled transaction")
+	}
+}
+
+// A body that ends the transaction itself with ROLLBACK makes WithTx's
+// own rollback fail. WithTx must return that failure together with the
+// body's error.
+func TestRepositoriesTransactionRollbackFails(t *testing.T) {
+	t.Parallel()
+
+	db, cleanup := setupTestDB(t)
+	defer cleanup()
+
+	ctx := context.Background()
+	repos := database.NewRepositories(db)
+
+	err := repos.WithTx(ctx, func(ctx context.Context, tx *sql.Tx) error {
+		_, err := tx.ExecContext(ctx, "ROLLBACK")
+		if err != nil {
+			return err
+		}
+
+		return errIntentionalRollback
+	})
+	if !errors.Is(err, errIntentionalRollback) {
+		t.Fatalf("expected the body's error, got: %v", err)
+	}
+
+	if !strings.Contains(err.Error(), "rolling back transaction") {
+		t.Errorf("expected the failed rollback in the error, got: %v", err)
+	}
+}
+
+// When the body ends the transaction itself with ROLLBACK and then
+// panics, WithTx's own rollback fails; the panic must still reach the
+// caller.
+func TestRepositoriesTransactionPanicRollbackFails(t *testing.T) {
+	t.Parallel()
+
+	db, cleanup := setupTestDB(t)
+	defer cleanup()
+
+	ctx := context.Background()
+	repos := database.NewRepositories(db)
+
+	defer func() {
+		if recover() == nil {
+			t.Error("expected the body's panic to reach the caller")
+		}
+	}()
+
+	_ = repos.WithTx(ctx, func(ctx context.Context, tx *sql.Tx) error {
+		_, err := tx.ExecContext(ctx, "ROLLBACK")
+		if err != nil {
+			return err
+		}
+
+		panic(errIntentionalRollback)
+	})
+}
+
 func TestRepositoriesReadTransaction(t *testing.T) {
 	t.Parallel()
 

internal/vaultik/restore.go

@@ -103,12 +103,11 @@ type RestoreResult struct {
 	BlobsDownloaded int
 	BytesDownloaded int64
 	Duration        time.Duration
-	FilesFailed     int
-	FailedFiles     []string // Paths of files that failed to restore
 	// Verification results (only populated if Verify option is set)
-	FilesVerified     int
+	FilesVerified int
-	BytesVerified     int64
+	BytesVerified int64
-	VerifyFailedFiles []string // Paths of restored files that failed verification
+	FilesFailed   int
+	FailedFiles   []string // Paths of files that failed verification
 }
 
 // Restore restores files from a snapshot to the target directory
@@ -648,34 +647,20 @@ func (v *Vaultik) handleRestoreVerification(
 	opts *RestoreOptions,
 	result *RestoreResult,
 ) error {
-	// A file that failed to restore was not written and is already
+	err := v.verifyRestoredFiles(v.ctx, repos, files, opts.TargetDir, result)
-	// reported as such, so it is not verified.
-	failedToRestore := make(map[string]bool, len(result.FailedFiles))
-	for _, path := range result.FailedFiles {
-		failedToRestore[path] = true
-	}
-
-	restored := make([]*database.File, 0, len(files))
-	for _, file := range files {
-		if !failedToRestore[file.Path.String()] {
-			restored = append(restored, file)
-		}
-	}
-
-	err := v.verifyRestoredFiles(v.ctx, repos, restored, opts.TargetDir, result)
 	if err != nil {
 		return fmt.Errorf("verification failed: %w", err)
 	}
 
-	if len(result.VerifyFailedFiles) > 0 {
+	if result.FilesFailed > 0 {
 		v.UI.Errorf("Verification failed: %s files did not match expected checksums.",
-			v.UI.Count(len(result.VerifyFailedFiles)))
+			v.UI.Count(result.FilesFailed))
 
-		for _, path := range result.VerifyFailedFiles {
+		for _, path := range result.FailedFiles {
 			v.UI.Detailf("%s", v.UI.Path(path))
 		}
 
-		return fmt.Errorf("%d %w", len(result.VerifyFailedFiles), errFilesFailedVerify)
+		return fmt.Errorf("%d %w", result.FilesFailed, errFilesFailedVerify)
 	}
 
 	v.UI.Completef("Verified %s files (%s).",
@@ -1435,8 +1420,8 @@ func (v *Vaultik) verifyRestoredFiles(
 		if err != nil {
 			log.Error("File verification failed", "path", file.Path, "error", err)
 
-			result.VerifyFailedFiles = append(
+			result.FilesFailed++
-				result.VerifyFailedFiles, file.Path.String())
+			result.FailedFiles = append(result.FailedFiles, file.Path.String())
 		}
 
 		bytesProcessed += file.Size
@@ -1452,7 +1437,7 @@ func (v *Vaultik) verifyRestoredFiles(
 	log.Info("Verification complete",
 		"files_verified", result.FilesVerified,
 		"bytes_verified", ubytes(result.BytesVerified),
-		"files_failed", len(result.VerifyFailedFiles),
+		"files_failed", result.FilesFailed,
 	)
 
 	return nil

internal/vaultik/restore_skip_errors_test.go

@@ -4,9 +4,7 @@ import (
 	"bytes"
 	"context"
 	"fmt"
-	"os"
 	"path/filepath"
-	"strings"
 	"testing"
 
 	"github.com/spf13/afero"
@@ -86,130 +84,6 @@ func TestRestoreSkipErrorsSkipsFilesOfMissingBlob(t *testing.T) {
 	}
 }
 
-// TestRestoreSkipErrorsVerifyChecksOnlyRestoredFiles restores with
-// SkipErrors and Verify after one blob of a two-blob snapshot was
-// deleted. Verification must check only the files that were written, so
-// a file of the deleted blob is reported as failed to restore and not
-// also as failing verification, and Restore must still return an error.
-func TestRestoreSkipErrorsVerifyChecksOnlyRestoredFiles(t *testing.T) {
-	log.Initialize(log.Config{})
-	t.Parallel()
-
-	ctx := context.Background()
-	backup := backupThenDeleteOneBlob(ctx, t)
-
-	var out bytes.Buffer
-
-	v := newReaderVaultik(ctx, backup.cfg, backup.storer, nil, backup.fs)
-	v.UI = ui.NewWithColor(&out, false)
-
-	err := v.Restore(&vaultik.RestoreOptions{
-		SnapshotID: backup.snapshotID,
-		TargetDir:  backup.restoreDir,
-		Verify:     true,
-		SkipErrors: true,
-	})
-
-	require.Error(t, err, "restore must fail when files were skipped")
-	assert.Contains(t, err.Error(),
-		fmt.Sprintf("%d file(s) failed to restore", len(backup.lost)))
-
-	written := len(backup.files) - len(backup.lost)
-
-	assert.Contains(t, out.String(), fmt.Sprintf("Verifying %d files", written))
-	assert.Contains(t, out.String(), fmt.Sprintf("Verified %d files", written))
-	assert.NotContains(t, out.String(), "Verification failed")
-}
-
-// TestRestoreSkipErrorsVerifyReportsFailuresApart restores with
-// SkipErrors and Verify after one blob of a two-blob snapshot was
-// deleted, while one file that does restore is written one byte too long.
-// The restore warning must list only the files of the deleted blob, the
-// verification summary only the long file, each path once, and Restore
-// must return an error.
-func TestRestoreSkipErrorsVerifyReportsFailuresApart(t *testing.T) {
-	log.Initialize(log.Config{})
-	t.Parallel()
-
-	ctx := context.Background()
-	backup := backupThenDeleteOneBlob(ctx, t)
-
-	var long string
-
-	for path := range backup.files {
-		if !backup.lost[path] {
-			long = path
-
-			break
-		}
-	}
-
-	var out bytes.Buffer
-
-	fs := trailingByteFs{Fs: backup.fs, suffix: long}
-	v := newReaderVaultik(ctx, backup.cfg, backup.storer, nil, fs)
-	v.UI = ui.NewWithColor(&out, false)
-
-	err := v.Restore(&vaultik.RestoreOptions{
-		SnapshotID: backup.snapshotID,
-		TargetDir:  backup.restoreDir,
-		Verify:     true,
-		SkipErrors: true,
-	})
-	require.Error(t, err, "restore must fail when files failed")
-
-	_, afterWarning, found := strings.Cut(out.String(),
-		fmt.Sprintf("%d file(s) failed to restore:", len(backup.lost)))
-	require.Truef(t, found, "no restore warning in:\n%s", out.String())
-
-	warning, summary, found := strings.Cut(afterWarning,
-		"Verification failed: 1 files did not match")
-	require.Truef(t, found, "no verification summary in:\n%s", out.String())
-
-	for path := range backup.lost {
-		assert.Equalf(t, 1, strings.Count(warning, path),
-			"%s must be listed once as failed to restore", path)
-		assert.NotContainsf(t, summary, path,
-			"%s was not restored and must not fail verification", path)
-	}
-
-	assert.NotContainsf(t, warning, long,
-		"%s was restored and must not be listed as failed to restore", long)
-	assert.Equalf(t, 1, strings.Count(summary, long),
-		"%s must be listed once as failing verification", long)
-}
-
-// trailingByteFs appends one byte, on close, to the file it opens whose
-// path ends in suffix, so that file is restored one byte too long and
-// fails verification.
-type trailingByteFs struct {
-	afero.Fs
-
-	suffix string
-}
-
-//nolint:ireturn // afero.Fs.OpenFile is defined to return the interface
-func (fs trailingByteFs) OpenFile(
-	name string, flag int, perm os.FileMode,
-) (afero.File, error) {
-	f, err := fs.Fs.OpenFile(name, flag, perm)
-	if err != nil || !strings.HasSuffix(name, fs.suffix) {
-		return f, err
-	}
-
-	return trailingByteFile{File: f}, nil
-}
-
-type trailingByteFile struct {
-	afero.File
-}
-
-func (f trailingByteFile) Close() error {
-	_, _ = f.File.Write([]byte{'x'})
-
-	return f.File.Close()
-}
-
 // TestRestoreMissingBlobAbortsWithoutSkipErrors checks that a deleted blob
 // still ends the restore with an error when SkipErrors is not set.
 func TestRestoreMissingBlobAbortsWithoutSkipErrors(t *testing.T) {