[RELEASE] ScyllaDB 2025.1.3

Release Notes
, , ,
1

The ScyllaDB team announces ScyllaDB 2025.1.3, a bug-fix production-ready patch release for ScyllaDB 2025.1 LTS Release.

Related Links

The following issues are fixed in this release:

CQL

  • fix CREATE tablets KS warning msg to include MV and SI, which are not supported with tablets in this release #24006

Tablets

  • Some tables don’t get streamed to new nodes in a heterogenous rack when adding a table during a scale-out #23631

Stability

  • File streaming: use host-ids instead of ips #22421
  • Storage_service: use id to check for local node IP may change and an old gossiper message with previous IP may be processed when it shouldn’t. #22777
  • Fix potential use after free in replica/database: memtable_list: save ref to memtable_table_shared_data #23762
  • Internal reads from group 0 tables are using the wrong semaphore, causing group 0 unavailability due to getting stuck behind user generated workload #21637
  • Segmentation fault in cdc::make_new_generation_description #23897
  • stream_session does not ignore no_such_column_family exceptions #15257
  • Decommission may fail with seastar::rpc::closed_error #22514
  • MV: remove queue length limit from the view update read concurrency semaphore #23319
  • Topology_coordinator::stop doesn’t await all background actions #23755

Tracing:

  • Reduce log level when property file incomplete#20092
  • Topology_coordinator: silence ERROR messages on abort #22649

Tooling and API

  • Flush api should not fail on no_such_column_family/no_such_keyspace #16095
  • Nodetool compactionhistory crash: error running operation: rjson::error (JSON assert failed on condition ‘false’ #23540
  • scylla-nodetool: status: negative load size crashes nodetool #24134