Scylla Enterprise Release 2018.1.10

The Scylla team is pleased to announce the release of Scylla Enterprise 2018.1.10, a production-ready Scylla Enterprise minor release. Scylla Enterprise 2018.1.10 is a bug fix release for the 2018.1 branch, the latest stable branch of Scylla Enterprise. Scylla Enterprise customers are encouraged to upgrade to Scylla Enterprise 2018.1.10 in coordination with the Scylla support team.

The major fix in this release is avoiding reactor stalls while merging MemTable into the cache. This improvement was gradually added to the open source release, and is now backported to Scylla Enterprise. Related open source issues include: #2012, #2576, #2715, #3053, #3093, #3139, #3186, #3215, #3402, #3526, #3532, #3608, #4030

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

  • Counters: Scylla rejects SSTables that contain counters that were created by Cassandra 2.0 and earlier. Due to #4206, Scylla mistakenly rejected some SSTables that were created by Cassandra 2.1 as well.
  • TLS: Scylla now disables TLS1.0 by default and forces minimum 128 bit ciphers #4010. More on encryption in transit (client to server) here
  • Core: In very rare cases, the commit log replay fails. Commit log replay is used after a node was unexpectedly restarted #4187
  • On rare cases, and under heavy load, for example, during repair, Scylla Enterprise might OOM and exit with an error such as “compaction_manager - compaction failed: std::bad_alloc (std::bad_alloc)”. #3717, #3716
  • In some rare cases, during service stop, Scylla exited #4107
  • scylla_setup: An option to select server NIC was added #3658
  • Scylla Enterprise 2018.1 fails to run scylla_setup script with or without `--nic` flag on Ubuntu 16.04 when the NIC is not eth0 (--nic flag is ignored when passed)
  • CQL: Selecting from a partition with no clustering restrictions (single partition scan) might have resulted in a temporary loss of writes #3608

Related Links