kopia/kopia · error

error getting index set

Error message

error getting index set

What it means

ManagerV1.ListActiveIndexBlobs calls epochMgr.GetCompleteIndexSet for the latest epoch to enumerate active index blobs; this error wraps any failure from that epoch-manager query. Without the complete index set, callers (ListIndexBlobInfos) cannot enumerate index metadata, which typically blocks operations like full maintenance and content loading.

Solutions

1. Inspect the wrapped cause for the underlying storage error (auth, network, missing marker) and fix it.
2. Retry the listing; epoch marker reads may have hit a transient storage fault.
3. Verify the upgrade to index blob manager V1 completed and epoch markers are intact (check `kopia repository status`).
4. Increase context timeout if large index sets exceed the deadline.

Example fix

// before
active, deletionWatermark, err := m.epochMgr.GetCompleteIndexSet(ctx, epoch.LatestEpoch)
if err != nil {
    return nil, time.Time{}, errors.Wrap(err, "error getting index set")
}
// after: retry transient failures
var active []blob.Metadata
var deletionWatermark time.Time
err := retry.WithExponentialBackoff(ctx, "get index set", func() error {
    var werr error
    active, deletionWatermark, werr = m.epochMgr.GetCompleteIndexSet(ctx, epoch.LatestEpoch)
    return werr
})
if err != nil {
    return nil, time.Time{}, errors.Wrap(err, "error getting index set")
}

if err := st.ConnectionInfo().CheckConnection(); err != nil {
    return fmt.Errorf("storage not reachable before listing indexes: %w", err)
}

infos, err := mgr.ListIndexBlobInfos(ctx)
if err != nil && strings.Contains(err.Error(), "error getting index set") {
    // retry after brief delay; epoch marker reads can fail transiently
    time.Sleep(retryDelay)
    infos, err = mgr.ListIndexBlobInfos(ctx)
}

When it happens

Trigger: Listing index blobs on a V1 manager when the epoch manager cannot read/assemble the index set for epoch.LatestEpoch — storage read errors on epoch markers, missing/unreadable epoch manager state, or context cancellation.

Common situations: Storage outage or permission change while running `kopia content list` or maintenance; repository partially migrated to V1 with inconsistent epoch markers; very large index sets timing out on slow storage.

Understand the failure class

Background: EmptyResultError / "no results found": when an API or scraper succeeds but returns zero rows — this error's family across 9 libraries.

Related errors

• CompactEpoch
• error advancing epoch marker
• error cleaning up epoch manager
• error getting log sessions
• error iterating packs

AI-assisted analysis of kopia/kopia@82495e54b5 (2026-09-07). Data as JSON: /api/errors/3aefd4bd477ddd7d. Report an issue: GitHub.

Appendix: source

Thrown at repo/content/indexblob/index_blob_manager_v1.go:45

	formattingOptions IndexFormattingOptions
	log               *contentlog.Logger

	epochMgr *epoch.Manager
}

// ListIndexBlobInfos lists active blob info structs.
func (m *ManagerV1) ListIndexBlobInfos(ctx context.Context) ([]Metadata, error) {
	blobs, _, err := m.ListActiveIndexBlobs(ctx)

	return blobs, err
}

// ListActiveIndexBlobs lists the metadata for active index blobs and returns the cut-off time
// before which all deleted index entries should be treated as non-existent.
func (m *ManagerV1) ListActiveIndexBlobs(ctx context.Context) ([]Metadata, time.Time, error) {
	active, deletionWatermark, err := m.epochMgr.GetCompleteIndexSet(ctx, epoch.LatestEpoch)
	if err != nil {
		return nil, time.Time{}, errors.Wrap(err, "error getting index set")
	}

	var result []Metadata

	for _, bm := range active {
		result = append(result, Metadata{Metadata: bm})
	}

	contentlog.Log2(ctx, m.log, "total active indexes", logparam.Int("len", len(active)), logparam.Time("deletionWatermark", deletionWatermark))

	return result, deletionWatermark, nil
}

// Invalidate clears any read caches.
func (m *ManagerV1) Invalidate() {
	m.epochMgr.Invalidate()
}

View on GitHub (pinned to 82495e54b5)