v1.17.1
📦 qdrantView on GitHub →
✨ 6 features🐛 13 fixes🔧 2 symbols
Summary
This release focuses heavily on stability and performance, introducing non-blocking Gridstore flushes, improving filtered search performance, and fixing numerous bugs related to WAL replay, snapshot restoration, and cluster operations. Several new features are available in preview mode.
✨ New Features
- Defer point updates, efficiently apply and optimize points with `prevent_unoptimized=true`
- Add request tracing ID into audit log
- Allow peer to bootstrap with used URI if empty
- Per-collection metrics in Prometheus (Preview)
- API for reading audit log entries (Preview)
- Maximum batch size config in strict mode (Preview)
🐛 Bug Fixes
- GPU: fix raw Vulkan name pointer type
- Fix creation of uninitialized shard key with replication factor > 1, fixing tiered multi-tenancy workflow
- Prevent `min_should` panic on large amount of conditions
- Fix for restore of cluster snapshots which creates unnecessary replicas in Partial state
- Fix Server disconnected without sending a response error while performing concurrent ingestion
- Security patch to force snapshot recovery from snapshot directory only
- Fix lock for collection-level operations during shard transfer
- Do not fail on collection-level operations with dummy shard
- Fix WAL reading issue introduced in 1.17.0
- Fix another panic in WAL replay
- Fix for WAL delta transfer: reject truncated recovery points before equal-pruning
- Fix incorrect warning in geo-index
- Prevent locking shard holder for a long time in slow stream records transfers