containerd/containerd · warning

walk all snapshots for %q failed: %w

Error message

walk all snapshots for %q failed: %w

What it means

The snapshot stats syncer walks every snapshotter (key = snapshotter name like overlayfs) and calls snapshotter.Walk to enumerate snapshots. This error wraps a Walk failure for the given snapshotter, aborting the sync for that snapshotter and meaning snapshot usage stats could not be collected.

Source

Thrown at internal/cri/server/images/snapshots.go:98

	return previous + alpha*(sample-previous)
}

// sync updates all snapshots stats.
func (s *snapshotsSyncer) sync() error {
	ctx := ctrdutil.NamespacedContext()
	start := time.Now().UnixNano()

	for key, snapshotter := range s.snapshotters {
		var snapshots []snapshot.Info
		// Do not call `Usage` directly in collect function, because
		// `Usage` takes time, we don't want `Walk` to hold read lock
		// of snapshot metadata store for too long time.
		// TODO(random-liu): Set timeout for the following 2 contexts.
		if err := snapshotter.Walk(ctx, func(ctx context.Context, info snapshot.Info) error {
			snapshots = append(snapshots, info)
			return nil
		}); err != nil {
			return fmt.Errorf("walk all snapshots for %q failed: %w", key, err)
		}
		for _, info := range snapshots {
			snapshotKey := snapshotstore.Key{
				Key:         info.Name,
				Snapshotter: key,
			}
			sn, err := s.store.Get(snapshotKey)
			if err == nil {
				// Only update timestamp for non-active snapshot.
				if sn.Kind == info.Kind && sn.Kind != snapshot.KindActive {
					sn.Timestamp = time.Now().UnixNano()
					s.store.Add(sn)
					continue
				}
			}
			// Get newest stats if the snapshot is new or active.
			sn = snapshotstore.Snapshot{
				Key: snapshotstore.Key{

View on GitHub (pinned to 4246446a2b)

Solutions

1. Check the wrapped cause and containerd logs for snapshotter-specific failures
2. Verify the snapshotter (e.g. overlayfs) is healthy: ctr plugins ls, check kernel/disk health
3. Restart containerd to reset snapshotter metadata handles; verify the snapshotter config under [plugins."io.containerd.snapshotter.v1..."]

// probe snapshotter health before sync
if _, err := snapshotter.Usage(ctx, key); err != nil {
    return fmt.Errorf("snapshotter %s unhealthy: %w", key, err)
}

err := sync(ctx, key)
if err != nil && strings.Contains(err.Error(), "walk all snapshots") {
    log.WithError(err).Warn("snapshot walk failed; will retry next sync interval")
    // skip this snapshotter this round; do not crash the syncer
}

When it happens

Trigger: snapshotter.Walk(ctx, fn) returns an error during periodic stats sync — snapshotter backend failure, metadata DB read error, or context cancellation while iterating overlayfs snapshot metadata.

Common situations: Overlayfs/kernel issues or disk errors; heavy IO causing snapshotter metadata read timeouts; snapshotter plugin misconfiguration in containerd config.

Related errors

• no metrics received
• NotFound
• InvalidArgument
• failed to delete parent link: %w
• failed to delete snapshot: %w

AI-assisted analysis of containerd/containerd@4246446a2b (2026-09-02). Data as JSON: /api/errors/93d2f8642740f933. Report an issue: GitHub.