ScyllaDB Enterprise 2020.1.5

The ScyllaDB team announces the release of ScyllaDB Enterprise 2020.1.5, a production-ready ScyllaDB Enterprise patch release. As always, ScyllaDB Enterprise customers are encouraged to upgrade to ScyllaDB Enterprise 2020.1.5 in coordination with the ScyllaDB support team.

The focus of ScyllaDB Enterprise 2020.1.5 is improving stability and bug fixes. More below.

Related Links

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

  • Performance:  On Red Hat derivatives, like CentOS used in ScyllaDB Enterprise AMI, tuned service tries to tune the system for maximum throughput, and this conflicts with scylla-kernel-conf’s latency-oriented tuning. The result is slowdown, for example in LWT. #7696
    If you wish to keep tuned installed (not recommended), don’t install the scylla-kernel-conf subpackage.
  • Docker: doesn’t gracefully shutdown scylla when the container is stopped #6150
  • CQL binary Protocol: Drop OVERFLOW_ERROR cql binary protocol extension #6264
  • CQL: CREATE TABLE WITH DEFAULT TIME_TO_LIVE is accepted for a counter table #6879
  • CQL: (LOCAL_/EACH_)QUORUM consistency calculation is broken when RF=0 #6905
  • Tools: Add support for mutli-table snapshot in nodetool and REST API #6333
  • Tools: relocatable scripts, like fails because they lack proper paths #7350
  • Stability: Make allocation_section decay the reserves over time #325
  • Stability: large reactor stalls when recalculating TLS Diffie-Helman parameters #6191
  • Stability: possible use-after-free in create_writer()‘s error path #7406
  • Installation: when installing ScyllaDB not via packages (deb/rpm) does not apply sysctl configuration #7702
  • Installation: missing /etc/systemd/system/scylla-server.service.d/dependencies.conf on scylla-4.4 rpm #7703

31 Dec 2020