The ScyllaDB team announces ScyllaDB Open Source 6.0.1, a bugfix patch release of the ScyllaDB 6.0 stable branch. ScyllaDB Open Source 6.0.1, like all past and future 6.x.y releases, is backward compatible and supports rolling upgrades.
Users are encouraged to upgrade to 6.0.1.
Related links:
- ScyllaDB Open Source 6.0.0
- Get ScyllaDB Open Source - AWS AMI, GCP Docker, binary packages and unified installer
- ScyllaDB Web Installer for Linux (all releases)
- Upgrade from ScyllaDB Open Source 6.x.y to Scylla Open Source 6.x.z
- Upgrade from ScyllaDB Open Source 5.4.x to 6.0.y
- Please let us know if you encounter an issue.
Issue fixed in this release:
- Stability[tablets]: test_replace_node_*_ip_take_write dtests fail with: No mapping for :: in the passed effective replication map #18843. May happened when ScyllaDB attempted to access a node that already left the cluster and therefore had no reverse mapping from IP address to Host ID.
- Stability[tablets]: topology_coordinator needs to be properly stopped on shutdown #18745
- Build: abseil is not built with proper -O option #19161
- Stability[tablets]: coredumps after decommission and after repair #19052
- Stability: Group 0 state machine queries are using a streaming scheduling group, where it might wait for a long repair (fo example) process to complete #18863
- Stability: hinted handoff: Draining may not happen while migrating to 6.0.x #18761
- Stability: Fixed the background work of the Alternator expired-item scanner (TTL) to be recognized as background maintenance work. Previously, it was mistaken as user requests, which could degrade the performance of the user’s workload. #18719
- Stability: node running out of memory and crash during index building and topology change #17476. The root cause is storage_proxy::_throttled_writes may grow without bound #1834
- Stability[tablets]: repair is unsafe with concurrent tablet migration or replica set change #17658