ScyllaDB Enterprise 2020.1.2

The ScyllaDB team announces the release of ScyllaDB Enterprise 2020.1.2, a production-ready ScyllaDB Enterprise patch release. As always, ScyllaDB Enterprise customers are encouraged to upgrade to ScyllaDB Enterprise 2020.1.2 in coordination with the ScyllaDB support team.

The focus of ScyllaDB Enterprise 2020.1.2 is improving stability and bug fixes. More below.

Related Links

Fixed issues in this release are listed below, with open source references, if present:

  • Alternator: a bug in the PutItem implementation with write_islation = always_use_lwt. In case of contention but a successful write, the written data could have not matched the intent. #7218
  • Consistency: nodetool repair may not repair writes which use a new schema with all nodes knowing about the new schema being down #4791
  • Stability: Segfault in evaluating multi-column restrictions #7198
  • Stability: in some rare cases, error recovery from an out-of-memory condition while processing an unpaged query that consumes all of available memory may cause the coordinator node to exit #7240
  • Stability: repair may cause stalls under different failure cases #6940 #6975 #6976 #7115
  • Stability: in some rare cases, a node may crash when calculating effective ownership #7209
  • Stability: Seastar tasks can lose scheduling_group association when the system is overloaded #7184
  • Stability: Unnecessary LCS Reshape compaction after startup due to overly sensitive SSTable overlap ratio #6938
  • AWS: update enhanced networking supported instance list #6991
  • Tools: Nodetool getendpoints can return wrongs nodes list #7134
  • CQL: queries with ALLOW FILTERING and constraints on counter values used are rejected as “unimplemented” #5635
  • Stability: Prevent possible deadlocks when repair from two or more nodes in parallel #6272
  • Performance: removing a potential stall on sstable clean up #6662 #6730
  • Performance: replacing node does not take writes, make the follow up repair less efficient #5482
  • Stability: Status of drained node after restart stayed as DN in the output of “nodetool status” on another node #7032
  • Alternator: wrong ordering of bytes attributes with “negative” bytes #6573
  • Stability: RPC handlers are not unregistered on stop #6904
  • Stability: Schema change, adding a counter column into a non-counter table, is incorrectly persisted and results in a crash #7065
  • CQL: Disabling dc_local_read_repair_chance by default #6109
  • Setup: scylla_setup does not abort RAID setup when no free disk available #6860
  • Performance: improve compaction performance by removing the trimming logic and reducing compaction parallelism #6957

22 Sep 2020